mobile learning anytime everywhere - university of west ...stu.westga.edu/~bthibau1/medt 8484-...

Mobile learninganytime everywhereA book of papers from MLEARN 2004Edited by Jill Attewell and Carol Savill-Smith

MLEARN2 0 0 4

Published by the Learning and Skills Development Agency

www.LSDA.org.uk

Registered with the Charity Commissioners

LSDA is committed to providing publications that are accessible to all. To request additional copies of this publication or a different format please contact:

Information Services Learning and Skills Development Agency Regent Arcade House 19–25 Argyll Street London W1F 7LS UK

Tel +44 (0)20 7297 9144 Fax +44 (0)20 7297 9242 [email protected]

Copyeditors: Helen Lund and Susannah Wight Designer: David Shaw and associates Outside front cover photograph: Creatas Images Printer: Blackmore Ltd, Shaftesbury, Dorset

CIMS 052232SP/10/05/1800

ISBN 1–84572–344–9

© Learning and Skills Development Agency 2005

You are welcome to copy this publication for internal use within your organisation.Otherwise, no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,electronic, electrical, chemical, optical,photocopying, recording or otherwise, without prior written permission of the copyright owner.

Further information

For further information about the issuesdiscussed in this publication please contact:

Jill Attewell Manager, Technology Enhanced Learning Research Centre Learning and Skills Development Agency Tel +44 (0)20 7297 9100 [email protected]

The m-learning project and this publication were supported by the Learning and Skills Councilas part of a grant to the Learning and SkillsDevelopment Agency for a programme ofresearch and development. The m-learning projectwas supported by the European CommissionInformation Society and Media Directorate-General(IST 2000-25270).

Contents

Acknowledgements

Foreword

Introduction 1

MLEARN 2004 conference papers

Using learning theories to design instruction for mobile learning devices 5Mohamed Ally

JELD, the Java Environment for Learning Design 9Marco Arrigo, Manuel Gentile and Davide Taibi

mCLT: an application for collaborative learning on a mobile telephone 11Marco Arrigo, Manuel Gentile, Davide Taibi, Giorgio Chiappone and Domenico Tegolo

Engaging and supporting mobile learners 15Jill Attewell and Tamatha Webster

Mobile learning in the retail trade: TransmobiLE 21(EU-funded Leonardo da Vinci II pilot project) Artur Barth and Gottfried Kreis

Wireless learning community hub 23Russell Beale

Integrating situated interaction with mobile awareness 25Russell Beale and Matthew Jones

Multimedia m-learning using mobile phones 27Kuderna-Iulian Benta, Marcel Cremene and Razvan Padurean

Finding the appropriate learning objects: 29human, mobility, and community issues Andrew Brasher

Development of a research plan for use of ambient technology 33to test mobile learning theories Andrew Brasher and Josie Taylor

Interactive Logbook: the development of an application to enhance 39and facilitate collaborative working within groups in higher education Susan Bull, Larry Bridgefoot, Daniel Corlett, Paul Kiddie, Tom Marianczak, Chet Mistry, Neil Sandle, Mike Sharples and Daniel Williams

Determining location in context-aware mobile learning 43Will Byrne, Peter Lonsdale, Mike Sharples, Chris Baber, Theodoros N Arvanitis, Pat Brundell and Russell Beale

The impact of innovation in medical and nursing training: 47a Hospital Information System for Students (HISS) accessible through mobile devices Filippo Cacace, Maria Cinque, Michele Crudele, Giulio Iannello and Marco Venditti

A system for adaptive platform-independent mobile learning 53Nicola Capuano, Matteo Gaeta, Sergio Miranda and Laura Pappacena

Mobile learning = collaboration 57Jo Colley and Geoff Stead

Tablet technology for informal collaboration in higher education 59Dan Corlett and Mike Sharples

The role of user scenarios as the central piece of 63the development jigsaw puzzle Diane Evans and Josie Taylor

Team awareness in personalised learning environments 67Alois Ferscha, Clemens Holzmann and Stefan Oppl

M-learning: an educational perspective 73Johan B Freysen

Mobile vision for ambient learning in urban environments 77Gerald Fritz, Christin Seifert, Patrick Luley, Lucas Paletta and Alexander Almer

Mobile learning in tomorrow’s education for MBA students 81Dirk Frohberg

Learning and collaborating using mobile devices and sessions 85Michael Gardner, Hui-Na Chua and Anuroop Shahi

From virtual cooperation to distance learning: 89realising remote multimedia platforms on a QoS-enabled network Stefano Giordano, Gianluca Insolvibile, Valentina Chionsini and Pietro Polese

Requirements for mobile learning games shown on 95a mobile game prototype Christoph Göth, Urs-Peter Häss and Gerhard Schwabe

A community of practice? The Toshiba Ambassador programme 101Bob Harrison

PaperLinks – linking printouts to mobile devices 105Martin Hitz and Stefan Plattner

Adaptive mobile learning systems – 109the essential issues from the design perspective Anu Jäppinen, Mikko Ahonen, Teija Vainio and Erika Tanhua-Piiroinen

Building a wireless learning management system 113Ann Jeffery and Keith Webb

Context awareness for MOBIlearn: creating an 115engaging learning experience in an art museum Peter Lonsdale, Chris Baber, Mike Sharples, Will Byrne, Theodoros N Arvanitis, Pat Brundell and Russell Beale

THE MOTFAL project: mobile technologies for ad hoc learning 119Eleni Malliou, Stavros Savvas, Sofoklis Sotiriou, Antonis Miliarakis and Manolis Stratakis

Providing quality of service (QoS) guarantees for 123e-learning applications over IP networks: the MOICANE project Riccardo Mazza, Enrico Roddolo, Gianmarco Panza and Enrico Balatti

eduSource Canada: learning object repositories and m-learning 127Rory McGreal

Interactivity in a large class using wireless devices 129Fabrice Mercier, Bertrand David, René Chalon and Jean-Pierre Berthet

mLab: handheld assisted laboratory 133Jelena Mitic, Markus Feisst and Andreas Christ

Collaborative m-learning using agents and virtual reality 135Teresa Monahan, Gavin McArdle, John Kilbride, Eleni Mangina and Michela Bertolotto

A new model for the m-learning emergency scenario in risk contexts: 139the emergency operator Anna Moreno and Sergio Grande

MOBIlearn user trials on the museum scenario: 143the first results of preparatory work Elena Murelli and Giorgio Da Bormida

How can one effectively assess students working in 149a collaborative mobile environment on an individual basis? Caoimhin O’Nuallain and Attracta Brennan

A case study on the future options for mobile workplace learning 153Marika Pehkonen and Hanne Murto (née Turunen)

Approaches to just-in-time learning with mobile phones: 157a case study of support for tourists’ language needs Anita Pincas

Mobile lessons: concept and applications for 163‘on-site’ geo-referenced lessons Antonio Pintus, Davide Carboni, Gavino Paddeu, Andrea Piras and Stefano Sanna

m-learning via the web: the challenge of size 167Kris Popat and Geoff Stead

E-learning to m-learning: an investigation into the potential 171for content conversion Leon Rodin

Educational content anytime, anywhere on any device 177Gerard Smyth

Blended learning, mobility and retention: 183supporting first-year university students with appropriate technology Andy Stone

HistoBrick: utilising J2ME to enable students to use mobile phones 187for game-based learning on descriptive statistics Georg Ströhlein

Evaluation of the technical and pedagogical mobile usability 191Antti Syvänen and Petri Nokelainen

A mobile interactive guide to the Museo Diocesano di Catania: 197how to ‘open’ the Museo using PDAs Dr Carlo Tognoni

The shift to seamless augmentation and ‘humane’ applications 201via mobile/wireless devices: a view to a future for lifelong learning David C Traub

Mobile learning – the ethical and legal challenges 203John Traxler and Nell Bridges

Implementing mobile technology as part of the 209multimode (blended) teaching and learning academic strategy @ the Tshwane University of Technology Prof H J van der Merwe

A study of mobile learning as part of everyday learning 211Giasemi N Vavoula, Mike Sharples, Claire O’Malley and Josie Taylor

Using mobile phones and pocket PCs for creating, delivering 213and retrieving content on the fly, based on mobile web browsers Daniel Weiss

International Workshop on Mobile Learning for Emergency Management(MLEM) papers

A mobile first aid training system for training doctors 217in stabilising casualties on site Verner Andersen

Management of novel situations through mobile learning resources 221Claudio Balducelli

Logistics and medical support for isolated, mobile 227and disaster situations: the I–DISCARE concept Eli Larsen, Antonio Bove and Bernard Comet

Author index 231

Acknowledgements

MLEARN 2004 conference chairs

Giorgio Da Bormida MOBIlearn project coordinator

Jill Attewell m-learning project coordinator

Professor Mike Sharples University of Birmingham

The Conference Programme Committee

Dr Theodoros N Arvanitis MOBIlearn project, University of Birmingham, UK

Jill Attewell m-learning project coordinator, Learning and SkillsDevelopment Agency (LSDA), UK

Dr Russell Beale MOBIlearn project, University of Birmingham, UK

Dr Roberto Bernazzani MOBIlearn Project, Università Cattolica del Sacro Cuore, Italy

Dr Ing Michael Boronowsky Technologie Zentrum Informatik (TZI) University of Bremen, Germany

Prof Tak-Wai Chan National Central University, Taiwan

Giorgio Da Bormida MOBIlearn project coordinator GIUNTI Interactive Labs, Italy

Dr Ingo Dahn Department of Computer Science, University ofKoblenz-Landau, Germany

Kevin Donovan development advisor, Learning and SkillsDevelopment Agency (LSDA), UK

Dr Michele Fanelli Hewlett Packard Italiana, Italy

Dr Adam Maria Gadomski Italian National Agency for New Technology(ENEA) and Interuniversity Center for Research on Cognitive Processing in Natural and Artificial Systems (ECONA), Italy

Dr Michael Gardner Institute for Socio-Technical Innovation and Research (Chimera), University of Essex, UK

Terry Keefe ICT Research and Development Ufi/learndirect, UK

Prof Michael Kelleher Learning Futures Ltd, UK

Prof Gabriele Kotsis Johannes Kepler Universität Linz, Austria

Dr Eelco Kruizinga CIBIT Consultants, the Netherlands

Dr Paul Lefrere executive director, eLearning Microsoft Education Solutions Group EMEA, UK

Prof Zakaria Maamar Zayed University, United Arab Emirates

Dr Jon Mason education.au, Australia

Prof Jeff McDonell Lingnan University, Hong Kong

Dr Elena Murelli MOBIlearn Project, Università Cattolica del Sacro Cuore, Italy (and acting local chair for the conference)

John O’Donoghue Centre for Learning and Teaching, University ofWolverhampton, UK

Dr Jeremy Roschelle SRI International, USA

Dr Carol Savill-Smith m-learning project, Learning and SkillsDevelopment Agency (LSDA), UK

Prof Mike Sharples MOBIlearn project, University of Birmingham, UK

Prof Josie Taylor MOBIlearn project, Open University, UK

David Traub Epiphany Partners, USA

Dr Gustaf Ulander Skería Utveckling, Sweden

Foreword

We are delighted that our two mobile learning research and developmentprojects supported by the Information Society and Media Directorate-General –m-learning and MOBIlearn – were able to collaborate to organise the very successful third annual MLEARN conference, MLEARN 2004.

The vision of the Directorate-General’s Information Society Technologies (IST)priority is ‘anywhere anytime natural access to IST services for all’ and mobile learning is starting to contribute to the realisation of this vision.Projects such as m-learning and MOBIlearn are improving our knowledge of mobile learning and helping us to investigate how this type of technology-enhanced learning can facilitate learning process in different learning situations for diverse groups of learners.

The MLEARN international conference series has become the primary forumfor mobile learning developers, researchers and enthusiasts to meet andexchange experiences, results, findings, tools and ideas thus contributing to knowledge and growth in this new field.

The venue for MLEARN 2004 – the beautiful and historic Odescalchi Castle in peaceful Bracciano outside Rome – provided a dramatic setting for the very lively proceedings including demonstrations of the latest technologies,systems and learning materials as well as exciting new ideas about teaching and learning.

This book contains the papers presented at the conference which we hopeyou will find interesting, informative and inspiring.

Giorgio da Bormida MOBIlearn project coordinator and MLEARN 2004 conference chair

Jill Attewell m-learning project coordinator and MLEARN 2004 conference chair

Mike Sharples MLEARN 2004 conference chair

Introduction

The authors who have contributed to this book are researchers, developersand practitioners in both educational and commercial organisations from a number of different countries. They have in common an interest in the still new and quickly evolving field of mobile learning.

The papers (including full papers, short papers and posters) are based on presentations given at the very successful and enjoyable internationalconference, MLEARN 2004, which had as its theme ‘learning anytimeeverywhere’. The conference was hosted at the Odescalchi Castle inBracciano, Rome and was organised by the MOBIlearn project, in collaboration with its sister project, m-learning.

Some of the papers detail the findings of mobile learning projects; some are based on desk research and attempts by the author(s) to identify andfurther develop theory relevant in mobile learning; and many report on work in progress. Work in progress includes both research and the development of mobile learning materials and systems.

The conference also hosted the First International Workshop on Mobile Learningfor Emergency Management (MLEM), and the papers relating to this workshopare grouped together in a dedicated section of this publication.

Many of the papers have been written by colleagues who are partners in one of the two large mobile learning projects supported by the EuropeanCommission’s (EC) Information Society Technologies’ (IST) programme –m-learning and MOBIlearn. Further information about the projects can befound on their websites at www.m-learning.org and www.mobilearn.org

MLEARN 2004 built on the success of MLEARN 2003, which was alsoorganised by the m-learning and MOBIlearn projects. A book based on the papers presented at MLEARN 2003 – on the theme of ‘learning with mobile devices: research and development’ – is available from theLearning and Skills Development Agency (LSDA) and can be downloaded in pdf format from the LSDA website (www.LSDA.org.uk).

MLEARN is now well established as an annual international event and MLEARN 2005 will be held in October 2005 in Cape Town, South Africa.MLEARN2005 is being hosted by the Tshwane University of Technology, the University of South Africa and the University of Pretoria. The conferencewebsite can be found at www.mlearn.org.za/

Jill Attewell Dr Carol Savill-Smith m-learning project coordinator m-learning project researcher

1

3

MLEARN2004conference papers

AbstractCurrent instructional design models and methodswere developed to design instruction for deliveryon personal desktop computers that have largescreens and are located in learners’ homes orworkplaces. However, there is a trend towardsthe use of mobile devices to deliver learningmaterials, and for students to learn anytime andanywhere. The use of mobile devices for learninghas implications as to how learning materials aredesigned using learning theories and instructionaldesign principles. This paper will describe learningtheories and instructional design principles forthe design of learning materials for mobile devicesand the use of intelligent agents to customiselearning materials for individual learners.

Keywordsinstructional design, learning theories, mobilelearning, mobile device, m-learning

1 IntroductionThe growing use of wireless technology andmobile devices suggests that training andeducation cannot ignore the use of mobiledevices in the learning/training process. Anincreasing number of workers are workingoutside the office and they will requirejust-in-time training wherever they are located.In the next two years, there will be more than1bn subscribers to the wireless internet(Tirri 2003). Trainers and educators need todesign learning materials for the growing use ofmobile devices; however, the design of thematerials must be based on sound learningtheories and instructional design principles.

Sharples (2002) proposed a theory based onconversation for the design of mobile technologyfor learning. Roschelle and Pea (2002) suggestedthat wireless internet learning devices (WILD)could be used for computer-supported

Using learning theories to design instruction for mobile learning devicesMohamed [email protected] of Computing and Information SystemsAthabasca University1 University DriveAthabascaAlberta T9S 3A3Canada

collaborative learning (CSCL) where the teacherbecomes a guide or coach and learners take theinitiative in their learning. They claimed thatlearning will become more learner-centered,rather than teacher-centered, which hasimplications for how instruction is designed formobile devices. According to Tella (2003), mobiledevices can be used to increase cognitive growthat the individual level, and an individual’smotivation is enhanced when he or she is able todevelop based on needs and context. Knowledgeis information in context and knowledge creationis location-dependent and situation-dependent.Mobile learning devices allow learners to learnwherever they are located and in their personalcontext so that the learning is meaningful(Sharples 2000).

The use of mobile devices in learning is referredto as mobile learning (m-learning): this is thedelivery of electronic learning (e-learning)materials on mobile devices such as personaldigital assistants (PDAs), mobile phones, TabletPCs, Pocket PCs, palmtop computers, etc. Quinn(2000) defined mobile learning as ‘theintersection of mobile computing and E-learning:accessible resources wherever you are, strongsearch capabilities, rich interaction, powerfulsupport for effective learning, andperformance-based assessment. E-learningindependent of location in time or space’. Whendesigning learning materials for mobile devices,proper learning theories and instructional designprinciples must be used to meet learners’ needs,and at the same time, help learners to achievethe desired learning outcomes. However, whendesigning materials for mobile devices, certainlimitations must be kept in mind. Ahonen et al.(2003) reported on a study where a mobiledevice was used to deliver an electronic course.When asked about limitations related to thedelivery, students mentioned the small screensize and problems in navigating within the lesson.

Ally 5

Because of the small screen size in mobiledevices, the interface should be built in such away to convey the message using the smallestamount of text, and proper navigation must bebuilt into the system to allow learners to movebetween screens and sections of the lesson.Also, the interface must be appropriate forindividual learners and the system should be ableto customise the interface based on an individuallearner’s characteristics. The next sections of thispaper suggest how to design learning materialsfor delivery on mobile devices.

2 Learning theoriesCognitive psychologists claim that learninginvolves the use of memory, motivation andthinking; and that reflection plays an importantpart in learning. They see learning as an internalprocess and suggest that the amount learneddepends on the processing capacity of thelearner, the amount of effort expended during thelearning process, the depth of the processing(Craik and Lockhart 1972) and the learner’sexisting knowledge structure (Ausubel 1974).Paivio’s theory of dual coding (1986) states thatmemory is enhanced when information ispresented both in the verbal and visual form.According to constructivism, learners interpretthe information and the world according to theirpersonal reality; they learn by observation,processing and interpretation, and thenpersonalise the information into personalknowledge (Cooper 1993). Learners learn bestwhen they can contextualise what they learn,both for immediate application and to acquirepersonal meaning. Mobile learning facilitatespersonalised learning because learning (andcollaboration) from any place and at any timeallows the learning to be contextualised.

2.1 Learning principles for designingm-learning materialsRather than presenting all of the availablematerials to learners, intelligent systems must bebuilt to develop an initial profile of the learner andpresent materials that will benefit the specificlearner. As the learning system interacts with thelearners, it ‘learns’ about each student andadapts the interface and navigation patternaccording to that learner’s style and needs.

Because of the limited display capacityof m-learning devices, designers mustuse presentation strategies to enablelearners to process the materials efficiently.

Since working memory has limited capacity,information should be organised or ‘chunked’ intopieces of an appropriate size to facilitateprocessing. According to Miller (1956), becausehumans have limited short-term memorycapacity, information should be grouped intomeaningful sequences. Information to bedisplayed on m-learning devices should bechunked into between five and nine meaningfulunits to compensate for the limited capacity ofshort-term memory and the limitation of thedisplay device.

Rather than being presented in a textual format,the information should be organised in the formof a concept map or a network that shows theimportant concepts and the relationship betweenthe concepts. Networks can be used torepresent information spatially so that studentscan see the main ideas and their relationships.Novak, Gowin and Johanse (1983) used aconcept map to show the hierarchic structure ofsubject matter, similar to the way in whichinformation is stored in long-term memory. Inaddition, as a learning strategy, learners can beasked to generate their own concept map on themobile device. According to Stoyanova andKommers (2002), concept map generationrequires critical reflection and is a way ofexternalising the cognitive structure of learners’brains to facilitate deep processing. As ahigh-level collaborative activity, learners can beasked to work in virtual small groups to constructconcept maps and review each other’s maps.

Pre-instructional strategies should be used toallow learners to store the framework of thelesson: this will help them to incorporate thedetails of the lesson and to prevent informationoverload. This is critical for m-learning sinceinformation will be presented in pieces andlearners will need to use the general frameworkto integrate the pieces. Mobile learning materialsshould use expository advance organisers toallow learners to store the general framework;and comparative advance organisers to allowthem to use existing knowledge to make senseof and take in the new materials (Ausubel 1974;Ally 2004a).

The interface of the m-learning device mustbe designed properly to compensate forthe small screen size of the display. Theinterface must be graphical and mustpresent between five and nine chunks ofinformation on the screen to prevent informationoverload in short-term memory. The interfaceis required to coordinate the interactionbetween the learner and the learning materials.

6 Mobile learning anytime everywhere

The system should contain intelligent agents todetermine what the learner did in the past andadapt the interface for future interaction with thelearning materials. The m-learning system mustbe proactive, anticipating what the learner will donext and providing the most appropriate interfacefor the interaction to enhance motivation andlearning. For learning sessions that areinformation-intense, the system must adjust theinterface to prevent information overload. Waysof doing this include presenting fewer conceptson one screen; or organising the information inthe form of concept maps to give the overallstructure of the information and then presentingthe details. The interface must also use goodnavigational strategies to allow learners to moveback and forth between displays. Navigation onm-learning devices should be automatic, basedon the intelligence gathered on the learner’scurrent progress and needs.

Learning materials for m-learning devices shouldtake the form of learning objects which are in anelectronic format and reusable (McGreal 2004).Learning materials should also be designed in theform of learning objects to accommodatedifferent learning styles and characteristics. Theobjects are then tested and placed in anelectronic repository for just-in-time access fromanywhere using mobile devices. A course orlesson will comprise a number of learningobjects which are sequenced to form aninstructional event for a lesson or learningsession (Ally 2004b). The use of learning objectsallows for instant assembly of learning materialsby learners, intelligent agents and instructors,which facilitates just-in-time learning and training.

3 ConclusionMobile learning devices can be used to deliverlearning materials to students, but the materialsmust be designed properly to compensate forthe small screen size of the devices. Learningmaterials need to use multimedia strategies thatare information-rich rather than textual strategies.As a result, the writing style of coursedevelopers has to change from textual writing toa greater use of visuals, photographs, videos andaudio. Intelligent agents should be used inm-learning systems so that most of the work isdone behind the scene, minimising the inputneeded from learners and the amount ofinformation presented on the display of themobile devices.

ReferencesAhonen M, Joyce B, Leino M and Turunen H(2003). Mobile learning: a different viewpoint. InH Kynäslahti and P Seppälä (eds) Professionalmobile learning. Helsinki: IT–Press, 29–39.

Ally M (2004a). Foundations of educationaltheory for online learning. In F Elloumi andT Anderson (eds) Online learning handbook.Athabasca: Athabasca University, 3–32.

Ally M (2004b). Designing effective learningobjects for distance education. In R McGreal (ed.)Online education using learning objects. London:Routledge/Falmer, 87–97.

Ausubel DP (1974). Educational psychology:a cognitive view. New York: Holt, Rinehartand Winston.

Cooper PA (1993). Paradigm shifts in designinginstruction: from behaviorism to cognitivism toconstructivism. Educational Technology,33(5), 12–19.

Craik FIM and Lockhart RS (1972). Levels ofprocessing: a framework for memory research.Journal of Verbal Learning and Verbal Behavior,11, 671–684.

McGreal R (2004). Online education usinglearning objects. London: Routledge/Falmer.

Miller GA (1956). The magical number seven, plusor minus two: some limits on our capacity forprocessing information. Psychological Review,63, 81–97.

Novak JD, Gowin DB and Johanse GT (1983). Theuse of concept mapping and knowledge Veemapping with junior high school sciencestudents. Science Education, 67, 625–645.

Paivio A (1986). Mental representations: a dualcoding approach. Oxford: Oxford University Press.

Quinn C (2000). mLearning: mobile, wireless,in-your-pocket learning. LineZine. Atwww.linezine.com/2.1/features/cqmmwiyp.htm,accessed 19 June 2004.

Roschelle J and Pea R (2002). A walk on theWILD side: how wireless handhelds may changecomputer-supported collaborative learning. Paperpresented at the International Conference onComputer-Supported Collaborative Learning(CSCL–02), Boulder, Colorado, 7–11 January.

Sharples M (2000). The design of personalmobile technologies for lifelong learning.Computers and Education, 34, 177–193.

Sharples M (2002). Disruptive devices: mobiletechnology for conversational learning.International Journal of Continuing EngineeringEducation and Lifelong Learning, 12(5/6), 504–520.

Ally 7

Stoyanova N and Kommers P (2002). Conceptmapping as a medium of shared cognition incomputer-supported collaborativeproblem-solving. Journal of Interactive LearningResearch, 13(1/2), 111–133.

Tella S (2003). M-learning – cybertextual travellingor a herald of post-modern education. In HKynäslahti and P Seppälä (eds) Mobile learning.Helsinki: IT–Press, 7–21.

Thornton P and Houser C (2004). Using mobilephones in education. In Proceedings of the 2ndIEEE International Workshop on Wireless andMobile Technologies in Education, Taiwan,March 2004, 3–10.

Tirri H (2003). Promises and challenges of mobilelearning. In H Kynäslahti and P Seppälä (eds)Mobile learning. Helsinki: IT–Press, 23–28.


AbstractIn this paper we present JELD, the Java Environmentfor Learning Design, an online distance learningsystem designed to implement different kinds of approach to learning. The system is a Java 2Enterprise Edition that supports pedagogicaldiversity and innovation. JELD allows users to define teaching strategies, pedagogicalapproaches and educational goals using the IMSLearning Design Specification. At the moment,the system uses a subset of IMS Learning DesignSpecification that allows a knowledge-buildingprocess to be run in a collaborative learningenvironment. Moreover, JELD can adapt learning activities to the client device; in particular,we have developed modules for PCs, personaldigital assistants (PDAs) and mobile telephones.

Keywordslearning design, learning methodologies,collaborative learning, standards, mobile learning

1 IntroductionOver the last few years, many research studiesabout learning standards have been carried out,most of them concerning metadata definitions oflearning content such as IEEE WG12 (2002) and ADLTechnical Team (2001), but few of them have beenabout learning processes. Except for a few rarecases, the learning platforms developed conformedonly to the content standards. Besides, in mostcases, a learning platform dealt with a singlepedagogical approach without considering theinteroperability of learning processes amongplatforms. Recently, some studies have been carriedout regarding learning processes (Koper 2001).

In this context, learning design (IMS 2003) representsone of the most significant developments ine-learning (Dalziel 2003). Learning design (LD) aimsto define all the features of the learning process,such as teaching strategies, pedagogical approaches,educational goals, activities and services. Our system aims to contribute to this topic.

JELD, the Java Environment for Learning DesignMarco ArrigoManuel GentileDavide [email protected]@[email protected] for Education TechnologyItalian National Research CouncilVia Ugo la Malfa153–90146 PalermoItaly

2 Framework architectureAs shown in Figure 1 (overleaf), the frameworkconsists of a core engine for running a learningprocess, a number of service modules and someconnection modules.

The core engine receives an Extensible MarkupLanguage (XML) LD document as input andproduces a runtime instance of a learningenvironment ready to be delivered to users. Thisprocess follows three main steps: validation,environment set-up and runtime creation. In thevalidation step, according to the IMS guidelines,the core engine checks the references,semantics and completeness of the inputdocument. Then a set of objects (Enterprise JavaBeans) are instanced to set up the environment.Finally, the runtime environment is created, withthe services, learning objects and user profilesdefined in the input LD document. At themoment, the system uses a subset of IMSLearning Design Specification that allows aknowledge-building process to be run in acollaborative learning environment. A set ofmodules is used to provide all the servicesneeded in runtime; for example, the messageexchange module is used to send e-mails and/orShort Message System (SMS) between users.

Moreover, the interface of the learning runtimeenvironment is adapted to the device using theconnection modules. We have developed PC,PDA and mobile telephone modules. A difficulty inmaking content available for use by several typesof device derives from the available hardwareresources (the devices’ limited display capability,network bandwidth, etc). Technologies used byJELD provide Extensible HyperText MarkupLanguage (XHTML) and XML Forms (XFORMS)interfaces to adapt information to the studentclient. Using XHTML, we share basic contentacross desktops, PDAs and mobile phones.

Arrigo Gentile Taibi 9

Dalziel J (2003). Implementing learning design:the Learning Activity Management System(LAMS). In Proceedings of the 20th AnnualConference of the Australasian Society forComputers in Learning in Tertiary Education(ASCILITE), Adelaide, 7–10 December, 593–596.

IEEE WG12 (2002). Standard for learning object metadata. At http://ltsc.ieee.org/wg12,accessed June 2002.

IMS (2003). IMS Learning Design Specificationv1.0. At www.imsglobal.org/learningdesign/index.cfm, accessed 30 July 2003.

Koper R (2001). Modelling units of study from a pedagogical perspective: the pedagogicalmeta-model behind EML. Educational TechnologyExpertise Centre, Open University of theNetherlands. At http://eml.ou.nl/introduction/articles.htm, accessed October 2001.

Leinonen T, Virtanen O, Hakkarainen K and Kligyte G(2002). Collaborative discovering of key ideas in knowledge building. In Proceedings of theConference on Computer Support for CollaborativeLearning, Boulder, Colorado, 7–11 January, 529–530.

Scardamalia M and Bereiter C (1994). Computersupport for knowledge-building communities.The Journal of the Learning Sciences, 3, 265–283.


Figure 1 JELD architecture

3 A learning design case study:collaborative knowledge buildingAt present, we have implemented a collaborativeknowledge-building case study. In this case-studyscenario, the actors involved are a teacher and agroup of students. The teacher can define aknowledge type set that is the list of artefactsused by the students in the knowledge-buildingprocess (Scardamalia and Bereiter 1994).According to Leinonen et al. (2002), theknowledge type set of progress enquiry uses thefollowing artefact: discovery of key ideas inknowledge building. After this phase, the teachermust define the logical dependences among theartefacts that guide the building path. The teacherstarts the activities by defining a context coursein which the users are invited to insert theirartefacts according to the context schema.

ReferencesADL Technical Team (2001). SCORM Specification v1.2. At www.adlnet.org/Scorm,accessed October 2001.

AbstractMobile technologies offer new opportunitiesfor distance learning and enable people tocollaborate anywhere. This paper introducesan innovative mobile platform forcomputer-supported collaborative learning inwhich traditional methodologies of collaborationhave been improved, based on 3rd-generationmobile telephones. Students can collect andshare live data immediately, anywhere and at anytime. This enables them to play an active role inthe knowledge-building process. The mCLT ispotentially an application for ubiquitouscollaborative learning.

Keywordscollaborative learning, mobile computing

1 IntroductionCooperation between an individual and a group,between people and an organisation, betweentheoretical and practical education are only someexamples of collaborative learning. Most studieson collaborative learning have been carried out inthe last decade and they have proposed differentmeanings of ‘learning’. Watanabe and Kojiri havesuggested different kinds of educational supportsystem, such as computer-assisted instruction(CAI), computer-aided learning (CAL), intelligenttutor system (ITS) and computer-supportedcollaborative learning (CSCL), in which learnersare able to freely change their learning style(Kojiri and Watanabe 2001; Watanabe 2001).Fischer’s Lifelong Learning system (2000) allowslearning space to be rearranged using informationtechnology (IT); in this sense, learning activitiescan be seen as a practical workspace changingover a long period. Other studies look at thestandardisation of collaborative learning.

mCLT: an application for collaborative learning on a mobile telephone

Okamoto, Kayama and Cristea (2001) haveplanned a standardisation of collaborative learningbased on six essential structural elements:

p collaborative learning environment expression

p collaborative workspace expression

p collaborative learning resources expression

p collaborative workplace expression

p learner group model in collaborative learning

p collaborative memory structure expression.

Our paper focuses attention on the collaborativeworkspace expression because one of the majoradvantages of using mobile devices for onlinelearning is the benefit they offer to cooperation –working together to accomplish shared goals.Within cooperative activities, individuals seekoutcomes that are beneficial to themselves andbeneficial to all other group members. Cooperativelearning is the use of small groups forinstructional purposes so that participants worktogether to maximise their own and each others’learning (Johnson, Johnson and Smith 1991).

The mobile Collaborative Learning Tool (mCLT) canbe seen as an extension of the Java Environmentfor Learning Design (JELD) system developed inthe Italian National Research Council’s Institutefor Education Technology in Palermo (JELD 2005).The JELD system allows the teacher to design thelearning process using a direct graph model. Thenodes represent the knowledge types (eg problem,comment, working theory) and the edges arerelationships between them. The studentsparticipate in the learning activities, insertingnotes according to the learning graph rules.

The mCLT is potentially an application forubiquitous collaborative learning. The mostinnovative aspect of the mCLT application is thatthe students, using a mobile device, can play anactive role in the knowledge-building process,anywhere and at any time.

Arrigo Gentile Taibi Chiappone Tegolo 11

Marco ArrigoManuel GentileDavide [email protected]@[email protected] for Education TechnologyItalian National Research CouncilVia Ugo la Malfa153–90146 PalermoItaly

Giorgio ChiapponeDomenico [email protected]@math.unipa.itDepartment of Mathematics and ApplicationsUniversity of PalermoVia Archirafi34–90123 PalermoItaly

2 Application architectureThe mCLT system is a Java™ Mobile InformationDevice Profile (MIDP) client for mobiletelephones, designed for collaborative learningand knowledge sharing among online educationalcommunities. Figure 1 shows the functionalmodules of the system. It is characterised by aset of web services hosted on the server sideand six mobile client modules. Each module usesa Simple Object Access Protocol (SOAP) tointeract with the web services; and to managethe logon phases, a Security Managementmodule has been inserted to administer thesecurity aspect of the system. Automatic logon isperformed using the latest access informationstored in the local repository of the learners’mobile devices. Security Management alsoguarantees that communication with the serverthrough the web services is encrypted, using alightweight implementation of the AdvancedEncryption Standard (AES) algorithm.

Figure 1 mCLT use case view

The Navigation Course module was developed tomanage a number of course activities. Thisallows students to browse the list of courses,and to look for detailed information about eachcourse, the participants and the list of contexts(areas for online discussion). The Course Notemodule is used to visualise, select and publishthe students’ notes, and it allows users to replyto notes or add new notes with images alreadystored in the local repository. To create andmodify notes, a student uses the Editing Module.With the current version of the system, it is onlypossible to insert static images into notes.Communication between students is performedby the Peer-to-Peer Communication modulewhich uses a message-system web service tosend e-mails and Short Message System (SMS).Finally, a Manage Local Repository module hasbeen defined in the system; this module allowsusers to work in offline modality and to arrangeand store notes and images. The images can beacquired from a photo camera embedded in thelearner’s mobile device. All of the objects storedin the local repository can be used to create newmessages or notes in the online sessions.


3 ScenarioThe first step in using the mCLT system is theinstallation phase. The MIDLet suite can bedelivered to Java to Mobile Edition (J2ME)devices by the Over-the-Air (OTA) system of theJELD platform. Following the installation, a usercan manage the local repository which can beseen as a memory area in the mobile device. Theuser can store offline notes, photos and othersets of knowledge as in a personal desktop.

When a user has logged onto the system, the listof subscribed courses will be shown and he/shecan select a course and view the contexts of thecourse. Each context is a discussion on a topicwith a specific learning graph model, which isopened by the course author. In many respects,the context is like a thread in a discussion forum,but the use of metacognition identifiers (eg in theprogressive enquiry process: problem, comment,working theory, etc) makes it easy for the studentsto distinguish the key concepts in the discussion.In this manner, the system increases theparticipants’ awareness of their fellow students’thinking (Leinonen et al. 2002) and allows themto play an active role in the knowledge-buildingprocess, replying to any note of interest with a newnote. The user could add a new note according tothe learning graph rules of the specific context; infact, a note is a portion of knowledge such as atext and/or photos marked with a metacognitionidentifier called ‘knowledge type’.

The users collaborate in building knowledge andnew theories (Scardamalia and Bereiter 1994)anywhere and at any time, using the features of3rd-generation mobile telephones. Indeed, theusers can insert contents such as photos andvideos, directly acquired through the mobile phone,into the discussion threads. In this way, thediscussion is enriched by the multimedia andmobility functions of 3rd-generation mobile phones.

4 Conclusion and ongoing workThe application described in this paper should beconsidered as a contribution to ubiquitouscollaborative learning. The mCLT allows studentsto participate in a knowledge-building process.The students interact among themselves,following the model defined by the teacher.

A phase of testing prior to deployment has beenstarted, using an application prototype. The mCLThas been tested with five users from the CNRInstitute for Education Technology in Palermo,using three Java-compliant mobile telephones(MIDP 1.0 with Mobile Media ApplicationProgramming Interface) and two personal digitalassistants (PDAs) with the Java Personal Edition.

The results collected in this phase have beenused to improve the functionality of the systemand to correct some application bugs. When thealpha version of the mCLT is ready, the plan is tostart the online experimentation in a realuniversity course. Fifty students will be involvedin this phase to test the application using severaltypes of mobile device.

ReferencesFischer G (2000). Lifelong learning – more thantraining. Journal of Interactive Learning Research,11(3/4), 265–294. (Special issue on intelligentsystems/tools in training and lifelong learning,edited by R Mizoguchi and PAM Kommers).

JELD (2005). Website. At www.pa.itd.cnr.it/jeld/

Johnson DW, Johnson RT and Smith K (1991).Active learning: cooperation in the collegeclassroom. Edina, MN: Interaction Book Company.

Kojiri T and Watanabe T (2001). HARMONY:web-based adaptive collaborative learningenvironment. In Proceedings of the NinthInternational Conference on Computers inEducation, Seoul, 12–15 November, 559–566.

Leinonen T, Virtanen O, Hakkarainen K and KligyteG (2002). Collaborative discovering of key ideasin knowledge building. In Proceedings of theComputer Support for Collaborative LearningConference. Boulder, Colorado, 7–11 January,529–530.

Okamoto T, Kayama M and Cristea A (2001).Proposal of a collaborative learningstandardization. In Proceedings of the IEEEInternational Conference on Advanced LearningTechnologies, Madison, Wisconsin, 6–8 August,267–268.

Scardamalia M and Bereiter C (1994). Computersupport for knowledge-building communities.The Journal of the Learning Sciences, 3, 265–283.

Watanabe T (2001). Knowledge managementarchitecture of integrated educational support.In Proceedings of the Ninth InternationalConference on Computers in Education, Seoul,12–15 November, 1138–1141.

Arrigo Gentile Taibi Chiappone Tegolo 13

AbstractThis paper describes the learner research andsystems trials phase of the m-learning projectand the emerging findings. This phase of theproject took place between February and July2004 and involved nearly 300 learners usingSmartphones and PDA/phone hybrids to accessthe learning materials and systems developedwithin the project.

The research and trials explored learners’ andmentors’ reactions to and experiences with thedevices, learning materials and systems; andexplored whether m-learning can: engagereluctant young learners in education or training;and deliver learning and/or more enthusiasm forlearning. Appropriate and effective ways ofproviding pedagogic support and scaffoldingwere also explored.

Keywordsmobile learning, inclusion, widening participation,literacy, numeracy, basic skills, mentoring

1 The m-learning projectThe m-learning project is a 3-year pan-Europeancollaborative research and development (R&D)programme supported by the European Union(EU). The consortium is a partnership oforganisations combining skills in pedagogy andtechnology. It includes two university-basedresearch units [Ultralab at Anglia PolytechnicUniversity (APU) in the UK and Centro di Ricercain Matematica Pura et Applicata (CRMPA) at theUniversity of Salerno in Italy]; two commercialcompanies [Cambridge Training andDevelopment (CTAD) in the UK and Lecando inSweden]; and an independent not-for-profitnational educational research and development(R&D) agency [the Learning and SkillsDevelopment Agency (LSDA)] in the UK.

Engaging and supporting mobile learnersJill AttewellTamatha [email protected]@LSDA.org.ukLearning and Skills Development AgencyRegent Arcade House19–25 Argyll StreetLondon W1F 7LSUnited Kingdomwww.m-learning.org

The project has developed prototype productsand innovative approaches designed to supportlearning – particularly literacy, numeracy and lifeand survival skills – using handheld devices suchas mobile phones and palmtop computers orpocket computers. A key objective is to engagewith and motivate young adults who arenot taking part in education or training,including those who are unemployed,homeless or disadvantaged.

2 The learner research andsystem trials

2.1 Guiding principles andresearch questionsThe work of the m-learning project is guidedby the partners’ shared belief in thefollowing propositions.

p The fact that many young adults have poorliteracy and numeracy skills (Moser 1999) andlittle or no interest in education and training(DfEE 2001) is both a personal tragedy and awaste of potentially valuable national andEuropean resources.

p Learning is a natural human activity thatmost people will engage in if given theright encouragement.

p Many young people tend to be excited by, andinterested in, new technologies.

p Learning mediated by technology can provide aconvenient, personalised and non-judgementalalternative to traditional education.

In this phase of the m-learning project wehave been addressing three overarchingresearch questions.

Attewell Webster 15

Organisation Types of learner

Project Milford, Wales Disadvantaged, including unemployed

Foyer Federation Homeless

Hartpury College Modern Apprentices

Yaa Asantewaa Arts and Community Centre Afro-Caribbean young people not interested in learning

ContinYou Family learning including young fathers

Plymouth Learning and Work Partnership Mostly unemployed, disadvantaged

Kent Adult Education Travellers, homeless, young offenders, young mothers

Avon Consortium Traveller Education Service Travellers

City College Norwich Reluctant students

Highland Council, Scotland Unemployed, homeless, travellers, young offenders

Weston College Work-based learners/Modern Apprentices

City of Stockholm Schools School pupils including potential drop-outs

Mentoring USA/Italia, Salerno Disadvantaged youth

Albatros, Napoli Immigrants and dialect speakers

p Can the enthusiasm of many young adultsfor mobile phones and other portablecommunications and entertainment devices beharnessed to encourage those not currentlyengaged in education or training to take part inlearning experiences?

p Can m-learning result in improved literacy ornumeracy skills or in changes in attitude orbehaviour, including greater enthusiasm forlearning and/or progression to further learning?

p What kind of pedagogical support and scaffoldingdo m-learners need? How can this be provided?

2.2 Collaborating organisationsand learnersThe m-learning partners have recruited 14organisations to collaborate in this phase of theproject. They all have existing support andmentoring relationships with groups of youngadults in our target audience. Together they haverecruited nearly 300 young people to take part inour research. The table in Figure 1 summarisesthe collaborating organisations and the type ofyoung people they support.

Figure 1 Collaborating organisations


2.3 Selecting the mobile devices

Figure 2 The mobile devices

P800 P900 XDA II

In selecting the equipment, we tried to avoid theproject findings becoming out of date too soonand to anticipate the technologies that would bereadily available to the target audience after theend of R&D – that is, from 2005. The projectpartners developed a ‘road map’ charting ouranalysis of current and predicted technologiesand functionality throughout the project andprojecting their likely future relevance to theproject. Consideration was given to:

p delivery options (eg WAP, e-mail, SMS,MMS, HTTP)

p platform options (eg Pocket PC, Symbian,Windows CE, J2ME)

p media options (eg video, audio files, phone calls,teleconferencing, voice recognition)

p development languages (eg Flash, C#, WML,VoiceXML, HTML, XHTML)

p transport options (eg GPRS, 3G, downloading,infrared, Bluetooth).

As the start of the learner research and systemstrials approached, we selected and acquirednearly 100 devices including SonyEricsson P800and P900 Smartphones and O2 XDAII PDA/phonehybrids (see Figure 2 opposite).

2.4 The learning materialsand systemsThe learning materials developed include:

p smaller-than-bite-sized literacy, numeracy and lifeskills modules available both offline (downloadedto the XDA II) or online via the web browser andthe learning management system (LMS)

p driving theory test quizzes, which are Java-basedand can be downloaded to mobile phones whichsupport JAVA (including the P800 and P900)

p SMS quizzes linked to text-based materials anda mini-SMS language course (basic Italian)

p a mini web-page builder.

The ‘back office’ systems include a portal, a LMSand an intelligent tutor providing services includingaccess to relevant links and materials, coursebuilding, chat, discussion and web logs (blogs).

2.5 The research approachOur research explores the general hypothesesthat handheld devices can be used for learningand that m-learning can attract young people whodo not enjoy traditional education. As both theconcept and the practice of m-learning are stillvery new, our research largely focuses on theexploration of information gathered and furtherdevelopment of hypotheses, rather than onproving pre-existing theories. However, wheresufficient common positive messages arise fromseveral small groups, we hope this may provideenough evidence to persuade funding bodiesto support roll-out of m-learning services tolarger populations.

When we have analysed more of the data wehave collected, we would like to be able to reportsome concrete outcomes; for example, that theliteracy or numeracy of some of the learners hasimproved. However, this is difficult as none of thelearners will be m-learning for a very long period.The period of involvement for collaboratingorganisations varies from 3 to 9 weeks. Also, ithas been observed that ‘It is often both unlikelyand inappropriate for many projects to expect toachieve “hard” outcomes from target groupsthat are socially excluded and facing multiplebarriers to employment.’ (Dewson et al. 2000).We are therefore seeking ‘soft’ as well as ‘hard’outcomes from our research, includingindications of a change in attitude – learnersexhibiting apparently increased levels ofenthusiasm or confidence. We are alsoparticularly interested to find out whether someof the young people who have taken part in ourm-learning research later register to take part infurther e-learning.

We have collected information from severalsources to allow triangulation. Sourcesinclude information on the young adults viaintermediaries who have an existing relationshipand credibility with them; and electronicservices. Data collection is summarised inFigure 3 (overleaf). We felt it was important thatthe young people involved did not see theactivities they were involved in as just anotherlesson, or themselves as simply ‘guinea pigs’.Therefore we recommended to the mentors thatwhen they introduced the project, theyemphasised that it is about helping us to try outthe mobile devices, find out how good they are,discover any problems and find out if they mightbe useful to help other young people to learn.For this reason, we referred to the young peopleas research assistants (RAs) rather than learnersor users.

Attewell Webster 17

Figure 3 Data collection

Source: coordinator

p Project plan including objectives for the organisation, mentors and researchassistants (RAs)

p End-of-project form inviting reflection on experiences and results

Source: research assistants (RAs)

p Start-of-project interview with mentor

p End-of-project interview with mentor

p Direct messages to the project team posted onto mediaboard or via microportal

2.6 Emerging findingsAlthough we are at a very early stage of datacollection and analysis, there are some reportsemerging of positive outcomes for young adultsinvolved. One mentor has been working with ahomeless young adult who regularly truantedwhile at school and subsequently left schoolwithout any qualifications. The mentor reportsthat as a result of participation in the m-learningproject, her client has not only developed agreater confidence in his current reading andwriting abilities, but he has also been inspired toseek help to improve his mathematical skills fromthe local Adult Basic Education Centre.

Another mentor, working with a group ofdisplaced young adults studying ESOL (Englishfor Speakers of Other Languages), reports that anumber of learners who had previously avoidedusing PCs seem much more confident aboutusing technology since taking part in them-learning project and have begun to use PCs towork on tasks such as writing letters. Onementor noticed that learners were more focusedfor longer periods when involved in m-learning,compared with traditional classroom lessons. Thementor told our researcher: ‘The group wereobserved to be remarkably focused and calmduring the session when [they were] given thedevices, in contrast to their normal behaviour inthe sessions. They were far more focused andgave up to 2 hours of time to the devices when itis normally difficult to focus them for 15 minutes.’

Source: mentors

p Start-of-project questionnaire including prior experience, expectations and objectives

p End-of-project questionnaire inviting reflection on experiences and results

p Face-to-face or telephone interview exploringexperiences, issues, outcomes, future plans

p Start- and end-of-project assessment formsdescribing each RA

Source: systems

p Information about time and duration of access to online materials and systems

p Number and duration/size of phone calls, SMS and MMS messages

Another mentor believed that ‘the devices aregood tools to engage non-traditional learners,they remove the formality, which can be themost frightening aspect for those who have notengaged with learning’ and reported that the useof the mobiles has improved retention of learnerson their course.

It is possible that some of these effects are dueto the novelty of using mobile devices andwhether this is the case or not will becomeclearer over time.

Some positive outcomes not directly related tolearning have also been reported. Some of thelearners were surprised and proud to be trustedwith such expensive and sophisticatedtechnology; for example, one project mentornoted: ‘He took really good care of it. He pointedout that because of his background, no one elsewould have ever trusted him with a mobile. Thishas meant more to him than the actual deviceitself as he feels respected.’

This finding is consistent with earlier researchprojects involving laptop computers whereresearchers have found that the loan of portabletechnology causes positive effects, such asincreased motivation, to be felt also in children’slives outside school, resulting in improvedconfidence, motivation and self-esteem fordisaffected pupils and those with learningdifficulties or disabilities (eg Passey 1999).


When we first discussed with collaboratingorganisations the plan of lending the mobiledevices to learners and allowing them to takethese away from a classroom or centre, someexpressed concerns about possible damage ortheft, excessive or inappropriate use and/orincreased likelihood of mugging. The projectteam felt, based on the encouraging evidencefrom previous projects, that with some warningsand small safeguards, we should take the risk.Experience to date suggests that this was theright approach. We have had very little loss ordamage – one XDA has been damaged (someonesat on it!) and two XDAs have been lost or stolen,representing only 3% of our devices. There hasbeen some excessive use for non-projectactivities, causing us temporarily to block phonesand issue warnings, which have resulted inimproved behaviour; and only one reportedincident of inappropriate use of the internet.

2.7 Equipment and service issuesAlthough our experience was not quite as bad asthat of Luckin et al. ( 2004) who found that ‘anenormous amount of time was spent maintainingthe devices (older model XDAs) in full workingorder’, providing first-line support to our mentorsand learners has not been a simple task.Providing support has been complex, as learnersand mentors are inexperienced users of suchsophisticated mobiles and because problemsexperienced can be connected with one or moreof the following: device hardware or nativesoftware; the signal (or lack of signal) in the areaof use; third-party software, such as Flash, whichwe have loaded onto the devices; softwaredeveloped within the project by three differentpartners; learner misuse – for example, changingsettings or deleting software.

To date, handheld devices – and particularlymobile phones – have been mostly sold forindividual rather than project or institutional use:as a result, support from suppliers and serviceproviders is underdeveloped. For example, thewebsite which allows us to monitor the use ofthe mobiles is not updated in real time and hasto be searched by individual mobile phonenumber – a very time-consuming task. It has alsoproved much more difficult (and slower) thanexpected to implement restrictions on useractivities; for example, blocking of premium-ratetelephone numbers.

ReferencesMoser C (1999). Improving literacy andnumeracy: a fresh start, Sudbury: Department forEducation and Employment.

Dewson S, Eccles J, Tackey ND and Jackson A(2000). Measuring soft outcomes and distancetravelled: a review of current practice. London:Department for Education and Employment.

DfEE (2001). Skills for life: the national strategyfor improving adult literacy and numeracy skills.Nottingham: Department for Educationand Employment.

Luckin R, Brewster D, Pearce D, Siddons-Corby Rand du Boulay B (2004). SMILE: the creation ofspace for interaction through blended digitaltechnology. In J Attewell and C Savill-Smith (eds)Learning with mobile devices: research anddevelopment. London: Learning and SkillsDevelopment Agency, 87.

Passey D (1999). Anytime, anywhere learningproject evaluation. Lancaster: LancasterUniversity/AAL.

Attewell Webster 19

AbstractThis paper introduces an adaptive learning andteaching environment for the use of mobilehandheld computers in the retail trade. Studentsand employees learn with personal digitalassistants (PDAs) how to offer, for example,‘cross-selling’ and ‘up-selling’ opportunities tothe customer or perform a convenient checkout.A collaborative mobile learning environment atthe point of activity – on the sales floor – enableslearners to improve their competences on anindividual basis. Learner-centred teaching andlearning arrangements (content) are beingdeveloped for food and lifestyle articles.

Keywordsretail trade, point of activity,teaching materials, PDA

1 IntroductionTransmobiLE is an EU-funded transnational pilotproject for the integration of mobile learning intovocational education and continuing training inthe retail trade.

This project focuses on didactical-methodicalinnovative teaching and learning arrangements(content) (Achtenhagen et al. 1992). It startedon 1 October 2003 and will be finished on31 March 2006.

2 Project targetsOne project target is the design oflearner-centred learning/teaching arrangementsfor mobile terminals within the mobile broadbandnetwork, customised for the needs of individuallearners in the retail trade according toacademically based construction criteria(see the teaching/learning matrix in Barthand Warneke 2004).

Mobile learning in the retail trade: TransmobiLE(EU-funded Leonardo da Vinci II pilot project)Artur BarthGottfried [email protected]@t-online.deCarl–Severing–BerufskollegBleichstraße 12BielefeldGermanywww.transmobile.info

Another project target is the development of acertification concept for the retail trade, to enablethe identification, assessment and recognition ofcompetences acquired in formal, non-formal andinformal ways.

3 Technical standards in the projectStudents and employees learn with PDAs andmobile computers of industrial standard for theretail trade. These mobile devices, expanded witha 512MB secure digital (SD) card, are used in awireless local area network (WLAN) with internetconnection in the store.

The multimedia content of the learning andteaching arrangements is designed using MSMobile Office, Portable Document Format (PDF)and HyperText Markup Language (HTML) filesand Flash format.

4 Teaching and learning architectureThe TransmobiLE teaching and learningarchitecture (Rosenberg 2001) is divided intoperiods of learning and training at differentlocations. The first period starts in the classroomof a college or a training centre where thelearners are made familiar with the use ofthe mobile devices and the content of thelearning arrangements.

The next period consists of mobile learning withthese arrangements on the selling floor, at thepoint of activity (POA).

Barth Kreis 21

Figure 1 Mobile learning at the point of activity (POA)

5 Mobile learning at the pointof activityLearners can get the information they requirewirelessly and use it at the POA (see Figure 1).It enables them to offer ‘cross-selling’ and‘up-selling’ opportunities to the customer,while at the same time they improvetheir competences.

An example of the teaching/learning arrangements(content) is the development of a user’s manualfor the POS (point of sale), based on the retailsystem in use, that can be run on an iPAQ 5550Series or similar device. This guarantees instantaccess to learning content and provides thelearner with a product that complementse-learning and classroom-based training.

Figure 2 Online help for the POS system

ReferencesAchtenhagen F et al. (1992). Lernhandeln inkomplexen Situationen. Neue Konzepte derbetriebswirtschaftlichen Ausbildung.Wiesbaden: Gabler.

Barth A and Warneke P (2004). Fit fore-commerce. Rinteln: Merkur Verlag Rinteln.

Rosenberg MJ (2001). e-learning strategies fordelivering knowledge in the digital age. New York:McGraw–Hill.


Extended abstractKeywordscommunity, internet, mobile, shared spaces

Community and communicationin the 21st centuryIt is often argued that in everyday society oursense of community is being eroded. This isparticularly relevant for learners, who may findthat they have fewer opportunities to interactwith colleagues and so become isolated in theirlearning activities. While it may be true that fewerpeople spend time chatting on street corners orin coffee rooms, the decline of human interactionis overstated since they have often shifted theircommunication to digitally enabled routes. Manypeople make extensive use of the internet, withthe use of e-mail and instant messaging growingall the time, while mobile phone usage for bothvoice and text is also on the increase. Wirelessnetworks are enabling people to read and replyto their e-mails while sitting in the bar or a coffeeshop, and you can check on your team’s latestresult on your mobile phone.

The decline of physical communities iscomplemented by the increase ofinternet-enabled communities (Preece 2000),using functionalities ranging from chat rooms andbulletin boards to blogs (web logs) and sharedphotographs. Although Nielsen (1997) has arguedthat web users, in particular, have different goals,and thus do not often produce worthwhilecontent, some of these communities areexceptionally popular. Slashdot (2005) is awell-known computer and technology newsforum, finding articles and new ideas that interestits 330,000 daily readers. It is well known for itsability to completely overwhelm other websiteswith the traffic it can create – but its real strengthis in the commenting features that it has.

Wireless learning community hubRussell [email protected] of Computer ScienceUniversity of BirminghamEdgbastonBirmingham B15 2TTUnited Kingdom

Each story begins as the nub of a thread as on adiscussion board, to which its army of users canoffer their opinion. To try to eliminate off-topicconversations, Slashdot lets users moderatethemselves – points are assigned on variouscriteria, and only the highest scores are shown.These communities show that sharing opinionsand viewpoints online is exceptionally popular.

Internet learning communitiesMany learning opportunities are guided ormanaged through the internet – e-learningplatforms, blogs, wikis (software that allowsweb-page creation/editing without use of abrowser) and websites – and so there is agrowing body of learners who are familiar withthese approaches. However, most of thesesystems use only one technological method –the Web or e-mail or SMS – to develop theircommunities. For many people, what is requiredis the digital equivalent of the street corner –where people can come and go; requiring littleknowledge to participate in; and where peoplecan learn and gain support and advice fromtheir colleagues.

Wireless learning community hubThis project ties these issues and technologiestogether to create a community system so thatpeople can continue to communicate whereverand however they want to, without fear of beingbombarded with useless or offensive messages.An easy-to-use and clear interface is provided sonew users can instantly get involved. The coresystem collates the messages and presents theinformation in a variety of ways. For a communalspace, a message board containing the currentstories, comments, notes and images isprojected onto a wall, providing a large sharedarea with which co-located users can react.

Beale 23

With an uncluttered design, it is easy to scan andread updated information that is being submittedby other users, especially as the system learnsexactly what interests the community. Bymonitoring the messages that are commentedon, and analysing the contents of othermessages, it builds up a statistical model of therelevance and interest each message has (Beale2004). It uses this information to order themessages, and promotes the most interestingones to the top. By promoting these stories to aprominent position, the user becomes moreproductive since less time is wasted trying todetermine what is important. This also has theadvantage that spam and unwanted messagesquickly disappear off the display, actioned fordismissal by community consensus, not byadministrative intervention. Interesting stories orcomments can be commented on, increasing thesense of community interaction. Files such aspictures or documents can be attached to amessage, making it possible to collaborateon work.

As well as a central display, the system supportsother displays, so that being co-located is not anecessity. Users can receive information on theirmobile phones, on Tablet PCs and laptops, andon personal digital assistants (PDAs).Contributions to the system can be made via arange of devices and technologies: users canpost information to the system via their internetbrowser, via a wireless connection, or throughSMS, and can be physically co-located orgeographically separated. The questions,messages and stories are shared over a networkof computers, allowing quick informationexchange between users. Using the SimpleObject Access Protocol (SOAP), which is areliable and compatible communication protocol,it enables other implementations or evencompletely different products to join the networkand share information. By using an extensibleformat, new features can easily be added withoutdestroying backwards compatibility.

The system has been trialled within a universityenvironment and informally evaluated by third-yearand second-year students. Assessing it in terms ofaccessibility and usability, it scored highly, owingto the clean design and the ability to access it viawhatever device came to hand, be it mobilephone, laptop or while sitting in the lab in front ofa computer screen. The automatic ranking ofstories was felt to be useful, though a largercommunity of users would give a less eclecticprofile of interesting stories and would tend toproduce a more consistent and comprehensibleranking. Users felt the system would be useful tosupport supplementary questions in theirlearning process and to expand their horizons, sothat they could be made more aware ofreal-world examples of whatever issues werecurrently topical in lectures. However, withoutsome form of structure and specific educationalsupport, many felt that there was less use for thesystem in supporting formal learning than therewas in supporting informal learning – where itoffered an interesting forum in which to askquestions, tapping into the wide experience andbackgrounds of the student community.

In the future, we hope to trial the system for along period and analyse both the frequency andstyles of use, to validate the informal reactionsand findings reported above.

ReferencesBeale R (2004). Improving internet interaction:from theory to practice. JASIST (Special issueon best practices and future visions for searchuser Interfaces).

Nielsen J (1997). Community is dead;long live mega-collaboration. Atwww.useit.com/alertbox/9708b.html

Preece J (2000). Online communities: designingusability and supporting sociabilty, 1st edn.NewYork: John Wiley and Sons, Inc.

Slashdot (2005). Website. At http://slashdot.org/


AbstractThis paper describes the design andimplementation of an intelligent messagingsystem to allow students and lecturers tocommunicate better with each other whenoffice-based interactions fail.

Keywordsintelligent messaging, Short Messaging System(SMS), situated interaction

OverviewA common problem for many learningorganisations is the lack of any formal method tocontact staff members rapidly without thedisclosure of potentially personal information,such as mobile phone numbers. One ‘traditional’method of interaction existing between staff andstudents involves the sticking of notes on officedoors. Lecturers wanting to leave a message ontheir office door from a remote location oftenrequire the help of another member of staff, suchas a receptionist, who might put up a note ontheir behalf. This method has worked for a longtime, but there are intrinsic issues relating to alack of security and privacy, and the practicalproblems of notes falling off doors or a lack oftimeliness in posting the note.

To address these issues, a system has beendesigned and implemented that uses mobile andother technologies to provide an increased levelof interaction between staff and students. Theprimary focus was on creating a system basedon the concept of ‘situated interaction’(Streitz et al. 2003) in an attempt to bridge thegap that currently exists between the learner andthe instructor, and to allow an increased degreeof mobility and remote accessibility necessaryfor modern learning situations.

Integrating situated interaction with mobile awarenessRussell BealeMatthew [email protected] of Computer ScienceUniversity of BirminghamEdgbastonBirmingham B15 2TTUnited Kingdom

The system, named IMMS (Intelligent MobileMessaging System), involves the placement of anumber of display units on the office doors ofvarious members of staff in the department,acting as information and messaging terminalsfor students. These units are iPAQs or similarhandheld devices. Members of staff may write amessage and have a picture displayed on theunit; the content displayed on the units is set bythe owner of the unit via a remote-accessweb-based management system. For staff userswho do not have the time or opportunity to sitdown at a computer and log into themanagement system, an SMS-based system isavailable, allowing users to update their displayunit content by sending a text message fromtheir mobile phone to the IMMS server. This maybe of significant use to staff members who havea great many department-based commitments,but are frequently away from the department forwhatever reason. For example, staff memberscan update the display to inform scheduledvisitors that they will be late because they arestuck in traffic.

Student members of the department are not onlyable to view the image and the textual messageset by the owner of the unit, but are also able tosend messages to the owner via the display-unitinterface. A student calling in to see a lecturerand finding an empty office may use theweb-based interface on the display unit tocompose a short message and mark it urgent ornon-urgent. In the case of an urgent message,composed by students, other departmentalmembers or visitors, IMMS will either send theinformation via text message to the unit owner’smobile phone or store the message in themanagement system for display next timethe IMMS user logs into the system. Therouting of the messages sent via the displayunit depends on the configuration settingsprescribed in the personal profile of the user.

Beale Jones 25

This intelligent routing of messages extendssimilar work on situated door displays byCheverst et al. (2002) which had non-intelligentSMS notification.

The web-based components of the system offerconfiguration and message management, andalso allow remote access to the informationpresented on the screen, allowing users on theinternet to look up the status of lecturers withouthaving to go to their door.

The system was evaluated against its designgoals, meeting the main criteria for providingeasy-to-use, current, accessible information onthe door; for providing a way of managingmessages from the student to the lecturer viathe Web and/or SMS; and for enabling thelecturer to manage and update the informationdisplayed from a remote location.

ReferencesCheverst K, Fitton D, Dix A and Rouncefield M(2002). Exploring situated interaction withubiquitous office door display. In Proceedings ofthe International Workshop on ‘SituatedInteraction’ at the ACM Conference on ComputerSupported Cooperative Work, 16–20 November,New Orleans.

Streitz NA, Röcker C, Prante Th, Stenzel R andvan Alphen D (2003). Situated interaction withambient information: facilitating awareness andcommunication in ubiquitous workenvironments. In Proceedings of the TenthInternational Conference on Human–ComputerInteraction, 22–27 June, Crete.


AbstractAs mobile phones are getting smarter and morewidely used, they have become a familiar tool inour environment. Using them as a learning toolcalls for services that are compatible with theconcept of communication tools. Here wepropose a multimedia m-learning application thatcombines the advantages of text, image andsound using a combination of Wireless MarkupLanguage (WML) pages and MultimediaMessaging Service (MMS).

KeywordsWML, MMS, collaboration

1 IntroductionMobile phones are communication tools thatmay establish voice (sound), text and image linksbetween people in order to exchange informationand in this way can give support to the teachingand learning process. This implies gettinginformation from a source; putting questions toand getting answers from a tutor; and exchangingideas between students. All this must be in asimple form (Ring 2001); for example, a taskdefinition or a motivational message, verifyingthe information gathered.

1.1 ScenariosThe scenario is a biology lesson in which a classof students (or pupils) have to recognise a plantfollowing a WML text description and amultimedia message (MMS) provided by thetutor. The system provides a multimedia chatplatform to encourage discussions and pictureexchange with fellow students and with the tutor.

Multimedia m-learning using mobile phones

Accessing an encyclopaedia: the users caneventually send the plant picture to an artificialintelligence (AI) image recognition service thatwill detect the plant species. This service will belocated on the server due to the lowcomputational power of the mobile device.

Location-adapted information: a location featuremay also help students to do a mapping of howthe plant spread in the area and to research thehistorical evolution of the plant’s distribution inthat region.

2 The architectureThe students are using Wireless ApplicationProtocol (WAP) and Multimedia MessagingService (MMS)-enabled phones to access theinformation (Stead 2003). The mobile phone’sWAP browser reads the WML pages anddisplays the text within the link. If there is a needto listen to a sound or to see an image, thestudent may request an MMS with that contentby clicking on a link. The teacher composes his orher lessons by adding text, links between pages,images and sounds related to a line of text,using a HyperText Markup Language(HTML)-based interface.

Figure 1 (overleaf) depicts the servicearchitecture. The service responds with WMLpages and MMSes, putting a response togetheron the spot. The content is retrieved from theencyclopaedia database and sustains themultimedia information exchange between thestudents and the tutor.

Benta Cremene Padurean 27

Kuderna-Iulian BentaMarcel [email protected]@com.utcluj.roTechnical University of Cluj-NapocaStr. Constantin Daicovicu nr 15400020 Cluj-NapocaRomania

Razvan [email protected] to Technical University of Cluj-NapocaRomania

Figure 1 The service architecture

The architecture contains:

p the front-end components: MMS gateway (nowSMS) that sends and receives MMS, the WMLservlet that composes and responds with WMLpages, and the HTML servlet – responsible forthe tutor’s interface

p the MMS composer which assembles an MMSmessage from given ‘ingredients’

p the MMS receiver, which takes apart an MMSmessage and extracts the information (images,text, sound)

p the encyclopaedia database – which contains allthe multimedia information necessary for thelearning process

p the AI search engine, which is able to find in theencyclopaedia images or sounds similar to agiven image or sound, using a classical patternrecognition algorithm

p the service manager which implements theservice logic, the operation sequence,menus, etc

p the location server, an external and optionalcomponent which is useful to relate the serviceresponse to the location.

3 ResultsThe application has been tested on a group of 10students. Their opinions about this applicationdiffer … Some are concerned about the content’srelevancy, but most people think that it would bean interesting learning tool. The multimedia trendand the interactions with other people capturedtheir attention very quickly.The students used theMMS to send non-didactical content too.

If it were implemented as a service, the costswould not be too high, assuming that packagesof services at a special rate could be negotiatedfor MMS and given that costs per transfer forGPRS (General Packet Radio Service) are nowlower than they used to be. In terms of itsusefulness, the students thought that thisexample was good, but other implementationscould be done. We intend to develop alocation-based mobile-accessible encyclopaedia.

4 ConclusionThe paper presents a new m-learning multimediaservice allowing interaction between theparticipants and access to information sourcesvia WML and MMS. As the project develops,artificial intelligence (AI) and location features willbe considered and an extension of the databasecontent as well.

ReferencesFreeman JA and Skapura DM (1991). Neuralnetworks: algorithms, applications, andprogramming techniques. Addison WesleyPublishing Company.

Ring G (2001). Case study: combining Web andWAP to deliver e-learning. In Learning Circuits,June issue. Available via www.learningcircuits.org

Stead G (2003). Early footsteps and next steps:‘m-learning’ with disengaged young people.Paper presented at the MLEARN Conference2003, London, 19–20 May.


AbstractThis paper describes an approach to thedescription of the educational characteristics oflearning objects which aims to reflect thedemands of mobile use; it also addresseshuman factors related to the creation of therelevant metadata.

Keywordsmetadata, ontology, human–computerinteraction (HCI)

1 IntroductionIn the education domain there are severalspecifications (eg McKell and Thropp 2001) andstandards (IEEE 2002) for metadata descriptionsof learning objects. These are intended to enablecomputer systems to identify instances relevantto a user’s needs. These specifications andstandards break the description down intoseveral categories such as ‘general’, ‘technical’and ‘educational’. Unlike many of the othercategories of descriptors, most of those in theeducational category cannot be determined byanalysing the learning object itself, becausethese descriptors describe characteristics of theexpected use of the object, rather than of theobject itself (Marshall 1998). Examples includedescriptors of the difficulty of a learning object,and of the typical age range of the intendedusers. Mobility of learners can imply furtherdegrees of uncertainty about the expected useof the learning object, and this paper discussesan approach to the creation of the educationalcategory of descriptors which takes account ofthe need for human production of most of thedescriptors in this category.

Finding the appropriate learning objects: human, mobility and community issuesAndrew [email protected] Open UniversityWalton HallMilton Keynes MK7 6AAUnited Kingdom

2 Research issuesConsider an organisation which produces avariety of learning objects with a specific targetaudience (ie specific groups of learners) in mind.This organisation could be a university or acommercial publisher, but for the sake of thisexample, assume that it is a university producinglearning objects for its students. Each learningobject includes IEEE LOM (Institute of Electricaland Electronics Engineers Learning ObjectMetadata) standard metadata, includingmetadata in the educational category.

Chan et al. (2004) considered the suitability ofthe IEEE LOM for the mobile learning scenariosproduced by the MOBIlearn project, andconcluded that the only changes necessary werein respect of rights management. Accepting thatthe educational category is suitable for use inmobile learning scenarios, consider one elementfrom that category – the ‘difficulty’ element – asan example.

With respect to the ‘difficulty’ element, theexplanatory note in the IEEE standard states(IEEE 2002): ‘How hard it is to work with orthrough this learning object for the typicalintended target audience.’ It should be noted thatwith respect to both the ‘typical learning time’and ‘difficulty’ elements, the standard states:‘NOTE – The “typical target audience”can be characterized by data elements5.6:Educational.Context and5.7:Educational.TypicalAgeRange’.

Now consider that the university in questionhas made collections of its objects availableto third parties – for example, via an initiativesimilar to MIT OpenCourseWare (MIT 2003);or via commercial relationships with individualorganisations – for example, institutionssuch as museums or art galleries.

Brasher 29

In the former case (OpenCourseWare-typeinitiative), the objects will be freely available onthe internet to any learner; and in the latter case(museum or art gallery), the institution inquestion wishes to enable visitors to theinstitution to learn from appropriate learningobjects and thus makes the objects available viaa wireless system within its building. Note that inthis paper, only issues concerned with themetadata description of the objects is of interest,so the issues of transforming and renderingobjects on a variety of mobile devices – personaldigital assistants (PDAs), mobile phones, etc –are not discussed.

Next we consider the situation described byLonsdale et al. (2004), one in which a contextawareness system filters and selects objectsappropriate to a learner’s context from adatabase of available objects. The selection andfiltering operations described by Lonsdale et al.require that the metadata descriptors of theeducational characteristics of the learning objectscan be correctly interpreted by the contextawareness system. If learning objects containeducational metadata which is appropriate toonly one educational context (which ischaracterised by the Educational.Context andEducational.TypicalAgeRange elements in theIEEE LOM), then that is not sufficient for thecontext awareness system to select and filterobjects from a variety of contexts.

For instance, a specific learning object from anundergraduate art history course is described asof ‘medium’ difficulty for art historyundergraduates studying the course that theobject is part of. In other words, the communityof practice which produced the learning object(ie lecturers at the university) have decided thatthis learning object is of ‘medium’ difficulty forthe community it is intended for – art historyundergraduates at the university. If it is to becorrectly filtered and/or selected on the basis ofits difficulty for other contexts, then ‘difficulty’and ‘typical target audience’ metadata must becreated for those contexts; or a mechanismwhich allows the interpretation of relationshipsbetween contexts must be introduced. As statedearlier, ‘difficulty’ and ‘typical target audience’metadata is metadata which describes theexpected use of an object, and thus cannot beautomatically generated – it must be generatedby people. Experience at the Open Universityand elsewhere (eg Chan 1989) has shown thathuman-produced metadata is often neithercomplete nor consistent: thus it was decidedto focus on other approaches to achieve thedesired result.

3 The approachThe approach that has been developed andrealised as a working prototype in OWL (WebOntology Language) (McGuinness and vanHarmelen 2003) is an ontology which can act as amechanism for relating perceptions of difficultyappropriate to an individual educational contextto perceptions of difficulty in another educationalcontext. For example, one educational contextcould be characterised as undergraduatesstudying for an art history degree; and the othercould be characterised as people with a higherlevel of education, but not in art history, who arevisiting an art gallery.

The ontology enables the establishment of arelationship between descriptions of difficultyappropriate to one context and an expectation ofperceptions of difficulty in the other context. Forinstance, for two contexts (context1 andcontext2), a set of useful descriptors for thisrelationship is as follows.

p Assignments of difficulty made in this context(context1) are perceived as ‘more difficult’ thanassignments of difficulty made in this context(context2).

p Assignments of difficulty made in this context(context1) are perceived as ‘less difficult’ thanassignments of difficulty made in this context(context2).

p Assignments of difficulty made in this context(context1) are perceived to be ‘as difficult’ asassignments of difficulty made in this context(context2).

It should be noted that in this case, ‘assignment’should be interpreted as ‘the action of assigning’,not as ‘a task allocated to somebody as part of acourse of study’.

4 ConclusionsThis ontology enables metadata descriptions of‘difficulty’ appropriate for one context to becorrectly interpreted for learning that is occurringin other contexts, provided that the relationshipbetween the two contexts has been instantiated.This instantiation does require humanintervention, but as it describes relationshipsbetween contexts, only one instance is requiredfor each pair of contexts (compared withadditional metadata for every learning object inthe alternative approach discussed above).


Furthermore, the ‘difficulty’ metadata that isappropriate to one context (eg an undergraduateart history degree) can be created by expertsfrom the relevant community of practice (ie thelecturers on the course), for whom there is themotivation to ensure the creation of accurate andconsistent metadata. Moreover, the instantiationof the context relationship can be achieved via adiscussion between people who have endemicmotivation to ensure that the relationship iscorrectly captured; for example, senior managersor lecturers at the relevant organisations.

ReferencesCampbell LM (2003). Engaging with the learningobject economy. In A Littlejohn (ed.) Reusingonline resources: a sustainable approach toe-learning. London: Kogan Page Ltd, 35–45.

Chan LM (1989). Inter-indexer consistency insubject cataloguing. Information Technology andLibraries, 8, 349–358.

Chan T, Sharples M, Vavoula G and Lonsdale P(2004). Educational metadata for mobile learning.In J Roschelle, T Chan, Y Kinshuk and SJH Yand(eds) Proceedings of the 2nd IEEE InternationalWorkshop on Wireless and Mobile Technologiesin Education (WMTE). Computer Society Press.At http://csdl2.computer.org/comp/proceedings/wmte/2004/1989/00/19890197.pdf

IEEE (2002). IEEE Standard for Learning ObjectMetadata. IEEE Std 1484.12.1–2002.NewYork:The Institute of Electrical andElectronics Engineers.

IMS (2001). IMS learning information packagespecification. At www.imsglobal.org/profiles/,accessed 12 May 2004.

Lonsdale P, Barber C, Sharples M and Arvantis TN(2004). A context-awareness architecture forfacilitating mobile learning. In J Attewell andC Savill-Smith (eds) Learning with mobiledevices. London: Learning and SkillsDevelopment Agency, 167–171.

Marshall C (1998). Making metadata: a study ofmetadata creation for a mixed physical–digitalcollection. In Proceedings of the ACM DigitalLibraries ‘98 Conference. New York: ACM Press,162–171.

McGuinness DL and van Hamelen F (2003).OWL Web Ontology Language overview.At www.w3.org/TR/owl-features/, accessed21 May 2004.

McKell M and Thropp S (eds) (2001). IMSLearning Resource Meta-data Information Model.Version 1.2.1: final specification. Burlington,MA: IMS Global Learning Consortium. At www.imsglobal.org/metadata/imsmdv1p2p1/imsmd_infov1p2p1.html,accessed 5 September 2005.

MIT (2003). MIT OpenCourseWare.At http://ocw.mit.edu/, accessed 10 June 2004.

Brasher 31

AbstractThis paper reports on the progress that has madebeen towards the development of a researchplan to enable ambient technology to beemployed as a tool which researchers can usewhen testing theories of learning with mobiledevices, and developers can use when evaluatingmobile learning systems. It describes theapproach used, informed by the T-Plan approachdeveloped by Phaal, Farrukh and Probert(2004), and the progress made by examiningsources including:

p existing academic research which tests theoriesof learning (eg Moreno 2004)

p for evaluating educational software(eg Jones et al. 1999)

p information about relevant existing and emergingtechnologies such as those for identification ofphysical objects (eg RFID); data communication(eg GSM, GPRS, WLAN, UMTS or Bluetooth);outdoor location (eg GPS or GNSS); indoorlocation (eg infrared or optical); and video andaudio data capture.

Keywordsambient technology, evaluation, theory of learning

1 IntroductionThis paper focuses on the potential of ambienttechnology to obtain quantitative data aboutlearners’ interactions with all kinds of learningmaterials in such a way that the learners areunaware of the process of the data capture. Itdescribes some preliminary work towards thedevelopment of a research plan which, iffollowed, aims to enable ambient technology tobe employed as a tool which researchers can usewhen testing theories of learning with mobiledevices, and which developers can use whenevaluating mobile learning systems.

Development of a research plan for use of ambient technology to test mobile learning theoriesAndrew BrasherJosie [email protected]@open.ac.ukUserLabThe Open UniversityWalton HallMilton Keynes MK7 6AAUnited Kingdomwww.mobilearn.org

The need for this research (and hence itsinclusion in the development of the MOBIlearnroad map) was initially prompted by papersdescribing methodological difficulties with mediacomparison studies (eg Joy and Garcia 2000).

It describes the method used, informed by theT-Plan approach developed by Phaal, Farrukh andProbert (2004), and the progress made byexamining sources including:

p existing academic research which tests theoriesof learning (eg Moreno 2004)

p frameworks for evaluating educational software(eg Jones et al. 1999)

p information about relevant existing and emergingtechnologies such as those for identification ofphysical objects [eg RFID (Radio FrequencyIdentification)]; data communication [eg GSM(Global System for Mobile Communications)GPRS (General Packet Radio Service), WLAN(Wireless Local Area Network), UMTS (UniversalMobile Telecommunications System) orBluetooth]; outdoor location [eg GPS (GlobalPositioning Systems) or GNSS (Global NavigationSatellite Systems)]; indoor location (eg infrared,optical); and video and audio data capture.

We conclude by summarising some issues whichare critical to the success of such a research plan.

2 Framework: the T-Plan approachThe aim of a science and technology research‘road map’ is to provide a consensus view orvision of the future landscape available todecision makers. The road-mapping processprovides a way to identify, evaluate and selectstrategic alternatives that can be used toachieve a desired science and technologyobjective (Kostoff and Schaller 2001).

Brasher Taylor 33

This paper describes one aspect of work towardsthe development of a ‘road map for furtherresearch into the theory and practice of learningin a mobile environment supported by newtechnologies’ being carried out within theMOBIlearn project (MOBIlearn 2002). Theprocess adopted by MOBIlearn is based on theT-plan approach (Phaal, Farrukh and Probert2004), which has been designed to generateroad maps for product planning anddevelopment. Key reasons for considering theT-Plan approach within the short timescalepermitted in the collaborative MOBIlearn projectare as follows.

p It has been developed to support the rapidinitiation of road mapping.

p It provides a customised approach whichincludes guidance on the broader application ofthe method beyond its primary purpose ofsupporting product planning.

p It was developed during a 3-year programme inwhich more than 20 road maps were developed,and it has been applied more than 40 times.

p The general principles of the approach have beenused to develop multi-organisation (or collaborative)road maps (Phaal, Farrukh and Probert 2004).

The T-Plan approach entails four facilitatedworkshops, the first three of which focus on thethree main layers of the road map, with the finalworkshop bringing the themes together on atimescale-basis to construct the road map.A schema of the process as adapted for theMOBIlearn project is shown in Figure 1. Phaal,Farrukh and Probert (2004) make the point that‘Often a considerable part of the initialroadmapping effort will be directed at definingthe layers and sublayers that will form theroadmap’; and this paper is one output of thatinitial effort. The work described in this paper isintended to expose aspects of our currentthinking about two layers: the ‘theoreticalfoundations’ layer and the ‘organisation andinfrastructure’ layer. With reference to thetheoretical foundations layer, we propose thatthe road map should include developments toexisting models for evaluation and validation; andwith reference to the organisation andinfrastructure layer, we propose the necessarytechnological developments as part of an‘ambient technology’ sub-layer.


Figure 1 Adaptation of the T-Plan process steps for the MOBIlearn projectAdapted from Phaal, Farrukh and Probert (2004)

Figure 2 The CIAO framework (context, interactions, attitudes and outcomes)

Context Interactions Attitudes and outcomes

Rationale

Data

Methods

The aims and the context of use

Designers’ and courseteam’s aims. Policy documents and meeting records

Interview CAL (computer-assisted learning)designers and course team members; analyse policy documents

Data about students’interactions with thesoftware allows someanalysis of the process

Records of studentinteractions; products of students’ work;student diaries; online logs

Observation; diaries;video/audio andcomputer recording

It is important to try toassess learning outcomes,but also to considereffective outcomes; eg a change in perceptionsand attitudes

Measures of learning;changes in students’attitudes and perceptions

Interviews; questionnaires;tests

With respect to the sources identified insection 1, the first two – existing academicresearch which examines theories of learning;and frameworks for evaluating educationalsoftware – reflect the current state of the artwith respect to testing theories of learning, andhence feed into the development of the‘theoretical foundations’ layer. The third source –information about relevant existing and emergingtechnologies – reflects the state of the art andpotential future evolution of ambient technology,and thus is crucial for the ‘organisation andinfrastructure’ layer. An analysis of these sourceshas yielded topics and issues which we believeneed to be discussed at the relevant workshops forthese two layers. This analysis is discussed in moredetail in section 3 and section 4 respectively.

3 Theoretical foundations:evaluation and validationOur first step is to define what we mean by‘mobile learning’, and what this implies in termsof techniques for evaluation and validation.Rather than adopting a definition of ‘mobilelearning’ which takes account only of uses ofcertain technologies, the MOBIlearn project hasadopted a broader definition: ‘Any sort of learningthat happens when the learner is not at a fixed,predetermined location, or learning that happenswhen the learner takes advantage of the learningopportunities offered by mobile technologies.’(O’Malley et al. 2003).

In terms of evaluation and validation, Taylor(2004) promotes analysis of activities and designof mobile systems to support the learningactivities, stating that:

From the evaluator’s point of view, then, the taskis to evaluate the effectiveness with whichlearners are able to achieve their goals, andcomplete learning activities, irrespective of thespecific devices that might have been used indoing so. Indeed, the same or similar activitiescould be instantiated in a variety of different waysdepending on availability of technical support (egaccess to wireless Local Area Network, LAN) anduser preferences. In so doing, we will necessarilybe evaluating the validity of the tasks themselvesas vehicles for learning.

Therefore, it is the system plus the environmentthat must be evaluated, using a task-centredapproach. The work described here is thus a firststep towards investigating the potential ofambient technology to obtain quantitative dataabout learners’ interaction with materials in sucha way that the context of the learner’s tasks isnaturalistic, and the learners themselves areunaware of the processes of data capture.

Referring back to the definition of mobile learningstated above, we initially focus our attention onlearning opportunities offered by mobiletechnologies, and thus consider an existingframework for evaluation of educationalsoftware. Figure 2 shows a version of the CIAO(context, interactions, attitudes and outcomes)framework (Jones et al. 1999) which has beenamended to identify modifications that make itappropriate for our purposes – that is, to exploitambient technology. Components to whichambient technology could hypotheticallycontribute are shown in bold text in the cells.

Brasher Taylor 35

The CIAO framework encompasses bothqualitative and quantitative techniques. Thispaper is focusing on quantitative techniques andthus we must consider the ability of quantitativetechniques to generate reliable and valid datawith respect to learning. Many researchers havereported difficulties with quantitative techniques,including validity of interpretation of observations(because similar observed behaviours canrepresent very varied mental processes andintentions) and problems with moving fromstatistical to causal relationships betweenobserved events (see eg Wegerif and Mercer1997). We recognise that similar difficulties willoccur if the observations are a result of ambienttechnology rather than of a human observer.

3.1 Previous studiesAs a starting point we take our first source:studying existing peer-reviewed academicresearch which has devised and utilised methodsto gather and record quantitative observationaldata. First, we hypothesise as to how ambienttechnology could improve the recording ofobservational data. We then hypothesise howthe theoretical extensions to the CIAO evaluationframework discussed above could be realisedusing ambient technology. In section 4, weconsider the capabilities of the ambienttechnology that exists today, and we hypothesisedevelopments and improvements to thistechnology that would enable the theoreticalextensions to the CIAO framework to be realised.

So far, our analysis of published studies of mobilelearning indicates that with respect to evaluationand validation of mobile learning, ambienttechnology can be used to:

p identify gaps between what people say they doand what they actually do (Smørdal and Gregory2003); for example, to validate information gainedvia questionnaires and interviews

p generate additional information about learnerbehaviour during longitudinal studies oftemporally and spatially distributed learners(eg to contribute to a study such as is currentlyunderway within the MOBIlearn project(MOBIlearn 2002)

p record data about social processes that occurduring learning episodes that may not otherwisebe accessible to researchers (eg the collaborativelearning episodes reported in Lundin andMagnusson 2003; and in Zurita andNussbaum 2004).

A consideration of technological andinfrastructural issues related to thesepropositions is presented in section 4.

4 Organisation and infrastructure:ambient technologiesSection 3.1 described potential applications ofambient technology for evaluating and validatingtheories of learning. With respect to thesepotential applications, we can postulate that at thevery least, we need to consider in a workshopthe generation of positional data (related to bothpeople and objects); temporal data (related toboth people and objects); and audio and/or videodata (related to social interactions).

We now focus our analysis on technological andinfrastructural issues which need to be researchedand developed as part of the road map.

Hill et al. (2004) describe a range of platformclasses of nodes (hardware and softwarecombinations) which can be used to createwireless sensor networks. The classificationconsists of four classes of nodes(special-purpose sensors; generic sensors;high-bandwidth sensors; gateway), wheregateway nodes interface to existing networks(eg the internet); high-bandwidth nodes capture,for example, audio and video data; generic nodesdetect, for example, motion; and special-purposenodes are low-power cubic-millimetre scaledevices which are designed and programmed toperform specific functions. The authors state:‘While the capabilities, cost, and size of eachclass of device will change with technologicaladvances, these four fundamental classes ofdevice will likely remain for the foreseeable future.’

Thus this classification appears to be a usefulstarting point for discussions, at workshops andother forums, of these kinds of powered sensor.In addition, no matter what the nature of thesensor, there will be wireless security andinterference issues which will have to beconsidered, in addition to a variety of potentialmechanisms for locating people and objects intime and space.

Specific examples of the utilisation of suchdevices include the technique of Back et al.(2001) for monitoring page movements usingRFIDs – that is, a method to gather data relatedto people’s interaction with books. This was usedsuccessfully in the Listen Reader project (Backand Cohen 2000), which also used sensors todetect the motion of a reader’s hands.


5 ConclusionsThis paper has considered a variety of issueswhich affect the creation of a research road mapto explore the potential of ambient technology foruse as a tool to enable researchers to validatetheories of learning. It has identified ways inwhich ambient technology could be used withrespect to mobile learning; and has put forwardframeworks for discussion of relevant issueswhich should enable fruitful road-map workshopsto occur.

As stated in section 3, there are problemsconcerning the interpretation of observationaldata, and in moving from statistical to causalrelationships between observed events, that willaffect the usefulness of any data that is generatedby ambient technology. These are perhaps themost difficult and challenging issues that theresearch plan discussed here will have to face.

We have not considered how to integratequantitative data with qualitative data in thispaper, and we recognise that consideration ofsuch integration should be included in thediscussion that will occur during thedevelopment of the road map.

Finally, with respect to technologicaldevelopments, it is apparent that the rate ofadvance of technologies which can be used inor as ambient devices is continuing and willcontinue over the next decade. This being so,social and legal issues such as security andprivacy will have an impact on the utilisation ofsuch technology, and thus must also beconsidered and discussed.

ReferencesBack M, Cohen J, Gold R, Harrison S andMinneman S (2001). Listen reader: anelectronically augmented paper-based book.In Proceedings of the SIGCHI Conference onHuman Factors in Computing Systems.NewYork: ACM Press, 23–29.

Back MJ and Cohen J (2000). Page detectionusing embedded tags. In Proceedings of the13th Annual ACM Symposium on User InterfaceSoftware and Technology. New York: ACM Press,159–160.

Hill J, Horton M, Kling R and Krishnamurthy L(2004). The platforms enabling wireless sensornetworks. Communications of the ACM, 47, 41–46.

Jones A, Scanlon E, Tosunoglu C, Morris E,Ross S, Butcher P and Greenberg J (1999).Contexts for evaluating educational software.Interacting with Computers, 11, 499–516.

Joy EH and Garcia FE (2000). Measuringlearning effectiveness: a new look atno-significant-difference findings. Journal ofAsynchronous Learning Networks, 4, 33–39.

Kostoff RN and Schaller RR (2001). Science andtechnology roadmaps. IEEE Transactions onEngineering Management, 48, 132–143.

Lundin J and Magnusson M (2003). Collaborativelearning in mobile work. Journal of ComputerAssisted Learning, 19, 273–283.

MOBIlearn (2002). Next-generation paradigmsand interfaces for technology supported learningin a mobile environment. Exploring the potential ofambient intelligence. Project No IST–2001–37187.Technical Annex. MOBIlearn Consortium.

Moreno R (2004). Decreasing cognitive load fornovice students: effects of explanatory versuscorrective feedback in discovery-basedmultimedia. Instructional Science, 32, 99–113.

O’Malley C, Vavoula G, Glew JP, Taylor J,Sharples M and Lefrere P (2003). Guidelinesfor learning/teaching/tutoring in a mobileenvironment. MOBIlearn deliverable D4.1. At www.mobilearn.org/download/results/guidelines.pdf, accessed 15 June 2004.

Phaal R, Farrukh CJP and Probert DR (2004).Technology roadmapping – a planning frameworkfor evolution and revolution. TechnologicalForecasting and Social Change, 71, 5–26.

Smørdal O and Gregory J (2003). Personal digitalassistants in medical education and practice.Journal of Computer Assisted Learning,19, 320–329.

Taylor J (2004). A task-centered approach toevaluating a mobile learning environment forpedagogical soundness. In J Attewell andC Savill-Smith (eds) Learning with mobiledevices. London: Learning and SkillsDevelopment Agency, 167–171.

Wegerif R and Mercer N (1997). Usingcomputer-based text analysis to integratequalitative and quantitative methods in researchon collaborative learning. Language andEducation, 11, 271–286.

Zurita G and Nussbaum M (2004). Computersupported collaborative learning using wirelesslyconnected handheld computers. Computers andEducation, 42, 289–314.

Brasher Taylor 37

AbstractThis paper describes the development ofInteractive Logbook, an application designedprimarily for university students, to facilitate andenhance individual and collaborative learning inhigher education. Interactive Logbook exploitsthe Tablet PC architecture to incorporate wirelesscommunication, natural handwriting input andcontext awareness, and provides a seamlessintegration of tools for learning, collaboration, andtime management. The development follows closelythe socio-cognitive engineering methodology(Sharples et al. 2002) for the human-centreddesign of educational learning technology. Inaddition, usability evaluation techniques such ascard sorting, heuristic evaluations and field studieshave been used to ensure that the systeminterface is intuitive, uncluttered and easy to use.

KeywordsTablet PC, collaborative, computer-supportedcollaborative work, higher education, contextawareness, groupware

1 IntroductionA recent development within education,particularly higher education, is support for groupand collaborative learning. Previous research hasshown that peer collaboration is an effective meansof learning (Collier 1980; Johnson, Johnson andSmith 1991). Although the results of this researchhave been available for over 20 years, it has onlyrecently had a widespread influence within highereducation. It is now accepted that a significant partof a degree should include working within smallgroups towards the successful completion of aproject. To benefit from group work, the studentsneed to communicate and share ideas quicklyand efficiently. New communications technologycan offer tools for distributed team working, butthese have mostly been designed for work inbusiness teams rather than student learning.

Interactive Logbook: the development of anapplication to enhance and facilitate collaborativeworking within groups in higher education

The aim of Interactive Logbook is to provide anintegrated suite of tools to facilitate and enhancecollaborative learning by small groups within ahigher education (HE) setting. In conjunction withthis aim, the emergence of Tablet PCs as a viablealternative to desktops, laptops and personaldigital assistants (PDAs) was recognised, and theincreased mobility and accessibility of this formof tool was harnessed to improve distributedgroup working.

A collaborative tool is not a new idea, nor is theability to work wirelessly while mobile. Wirelessand mobile technology for use within educationhas been available for some 10 years or more(Wayne 1993), but the barriers toimplementation, such as cost and technologylimitations, have only recently been overcome.

A similar system is the CoVis CollaboratoryNotebook (Edelson and O’Neill 1994), whichprovided a knowledge base to enable remotecollaboration between students, academic staffand professionals. However, this is now verysimilar to an internet message board and istherefore somewhat limited and outdated. Therehas also been developments in software toolsthat aid students to organise their studies moreeffectively. These are not necessarilycollaborative applications, but they indirectlyincrease peer collaboration through greaterpersonal organisation and local access toresources. A recent example of such anapplication is the Student Learning Organiser(SLO) (Holme and Sharples 2002). This wasdesigned for the Pocket PC and provided amobile platform for students to organise theirstudies, including tools for management of timeand resources. The authors took several keypoints from this project, including the necessityfor such organisational tools and the need for amobile device to provide anywhere, anytimeaccess to required information.

Bull Bridgefoot Corlett Kiddie Marianczak Mistry Sandle Sharples Williams 39

Susan BullLarry BridgefootDaniel CorlettPaul KiddieTom MarianczakChet MistryNeil SandleMike SharplesDaniel Williams

Department of Electronic, Electrical and Computer EngineeringUniversity of BirminghamEdgbastonBirmingham B15 2TTUnited Kingdomwww.interactivelogbook.tk

[email protected][email protected]@[email protected]@[email protected]@[email protected]@gmail.com

Figure 1 Socio-cognitive engineering methodologySource: Sharples et al. 2002

2 Development procedureTo complete the complex task of building such anapplication, a socio-cognitive engineeringmethodology (Sharples et al. 2002; see Figure 1)was adopted. This allowed an analytic approachto the design of human-centred technology.Through consideration and combination ofseveral different viewpoints, it facilitated thedevelopment of the system, but ensured thatthe main focus throughout the development wason the user.

Socio-cognitive engineering breaks systemdesign into building blocks, each of which isnecessary to deliver a successful application, asoutlined in Figure 1. The left-hand side, includinggeneral requirements, field studies and theory ofuse, provides a user-orientated approach to theanalysis of activity and theory, related tocollaborative learning in small groups. The taskmodel integrates these into a coherent depictionof how students learn together with their currenttechnologies. The task model also indicateslimitations of current learning and teamworkpractices that can inform the design of newtechnology. It enables an iterative process ofsystem design and evaluation, which isrepresented by the right-hand side.

3 Aims and general requirementsInteractive Logbook is intended to assiststudents in coordination and collaborativeproblem solving, through previously gainedknowledge, but not to support basic learning orknowledge gathering.

The initial general requirements were identifiedby undertaking brainstorming sessions withpotential users from different academicbackgrounds. The requirements include the abilityto communicate effectively, to share resourcesand information and to create collaborativedocuments. In addition, the brainstormingsessions showed the necessity for personal andgroup organisational tools, and the possibility ofcontextually aware devices was highlighted.

The theory of use (see Figure 1) wasconstructed through research into cognitiveprocesses and the social interactions involvedwhen students collaborate to reach a commongoal. The findings of this research showed thatthe advantages of collaborative learning overindependent study lie in the processes ofarticulation, conflict and co-construction of ideasbetween collaborating peers (Crook 1996).Dillenbourg et al. (1996) suggest that whenconsidering such processes, it is necessary todifferentiate between learning and problemsolving; and also between collaboration andcooperation.

Consideration of the differences betweencollaboration and cooperation suggests thatInteractive Logbook should be designed toensure a high level of mutual contributions toattaining a common goal. This is in contrast tosimply dividing work between group members.


Figure 2 The Interactive Logbook user interface

Facilities should be provided to ensure thatcoordination is simple and effective betweengroup members, and that contributing to a taskwhile interacting with group members is feasibleand simple.

In combination with the theory of use, the fieldstudies attempted to discover the requirementsof potential users. The field studies took the formof questionnaires, scenarios and focus groups,which pointed to a core set of requirements tobe included in the final application. Theserequirements included natural handwriting input,peer messaging and time-management facilities.

4 Ensuring usabilityUsability issues were highlighted during designand developments through the use of fieldstudies undertaken with several groups ofpotential users and experienced usability experts.

Possible interface designs were drafted andshown to groups of students so that they couldselect and modify the designs. In addition,15 students from a cross-section of courses tookpart in a card sort. Students grouped relatedfunctions, such as Calendar,To Do’s andAppointments into a common group which wasnamed appropriately, Diary.

This consultation process ensured ease of use,as the target users were consulted throughoutthe design process.

The result was the identification of four groups:Programs; Modules; Meetings; and Diary.

5 Development toolsand functionalityInteractive Logbook v1.1 is the first usablerelease compiled by the authors. The applicationemploys the .NET architecture and utilises theTablet PC SDK. Several user interface (UI)components were provided by Divil (2004).Figure 2 gives a typical overview of theInteractive Logbook’s interface.

The main source of user interaction is the LaunchPanel to the right-hand side of the interface, fromwhich four separate sections can be accessed.

p Programs: provides easy access to ‘miniapplications’ from within Interactive Logbook,such as the internet, text messaging to peers,and freehand notes entry.

p Modules: provides access to teaching material atthe user’s request by selecting the name of acourse module. It also displays relevant lecturematerial if the calendar shows a lecture inprogress at the current date and time.

Bull Bridgefoot Corlett Kiddie Marianczak Mistry Sandle Sharples Williams 41

p Meeting: provides collaborative tools such asaccess to previous group minutes and initiationof a peer-to-peer whiteboard session.

p Diary: provides time-management features.These include To Do’s and Appointments,represented in a Calendar view.

Furthermore, a Tab Panel located towards thebottom of the interface provides the student withaccess to shared group resources and privateresources. It also allows simple file manipulation.

Location sensing of other group members isprovided by the Ekahau (2004) positioningengine. Some functionality is provided for userswho belong to a number of project groups. Thecurrent group environment may be swapped topermit the access of resources related to anothergroup to which the current user belongs.

Finally, administration, though still in its infancy, ispresent in this release of Interactive Logbook. Itallows the use of Interactive Logbook on anyshared network space and the use of the systemby lecturers, who may add teaching material to aparticular lecture slot in the calendar, therebyallowing that material to be displayed when thedate and time for that lecture arises. Lastly,administrators may create year timetables thatcan be offered to all students, providingreminders and related teaching material from theInteractive Logbook system.

6 Further developmentsInteractive Logbook is still a prototype system.The authors are committed to using feedbackgained from the initial release to improve theusefulness of the system.

The collaborative features will be developed toimprove its usefulness in group-workingscenarios, team whiteboard sessions, inkmessaging and administration of teams by theInteractive Logbook server. In addition, theauthors will ensure that existing features are ascomprehensive as possible – for example, theadaptability of the software depending on theenvironment of the student; the addition ofmultimedia elements to notes and administrativetools. Finally, the reliability of Interactive Logbookwill be evaluated and improvements made toensure more sophisticated data handling inshared environments.

The authors intend to continue development withthe aim of establishing Interactive Logbook as anindispensable tool for students using Tablet PCs.

ReferencesCollier KG (1980). Peer-group learning in highereducation: the development of higher-order skills.Studies in Higher Education, 5(1), 55–62.

Crook C (1996). Computers and the collaborativeexperience of learning. New York: Routledge.

Dillenbourg P, Baker M, Blaye A and O’Malley C(1996). The evolution of research on collaborativelearning. In H Spada and P Reimann (eds)Learning in humans and machines. Oxford:Pergamon Press, 189–211.

Divil (2004). Website. At www.divil.co.uk/net/,accessed January 2004.

Edelson DC and O’Neill DK (1994). The CoVisCollaboratory Notebook: supporting collaborativescientific enquiry. In A Best (ed.) Proceedings ofthe 1994 National Educational ComputingConference, 146–152.

Ekahau (Jan 2004). Website. At www.ekahau.com,accessed January 2004.

Holme O and Sharples M (2002). Implementing astudent learning organiser on the Pocket PCplatform. In Proceedings of the EuropeanWorkshop on Mobile and Contextual Learning,Birmingham, 41–44.

Johnson DW, Johnson RT and Smith KA (1991).Cooperative learning: increasing college facultyinstructional productivity. ASHE–ERIC HigherEducation Report No 4. Washington DC: TheGeorge Washington University, School ofEducation and Human Development. (ERICDocument Reproduction Service No ED 343465)

Sharples M, Jeffrey N, du Boulay JBH, Teather D,Teather B and du Boulay GH (2002).Socio-cognitive engineering: a methodology forthe design of human centred technology.In European Journal of Operational Research,136(2), 310–323.

Smith R, O’Shea T, O’Malley C, Scanlon E andTaylor J (1991). Preliminary experiments witha distributed multimedia problem solvingenvironment. In J Bowers and S Benford (eds)Studies in computer supported cooperative work:theory, practice and design. Amsterdam:Elsevier, 31–48.

Wayne C (1993). Wireless coyote:a computer-supported field trip.Communications of the ACM, 36(5), 57–59.


AbstractContext-aware mobile learning systems rely onthe availability of information about a mobile user.One of the most obvious aspects of context islocation (Small, Smailagic and Siewiorek 2000),and several methods exist to determine a user’sposition, varying not just in the accuracy of theinformation they provide, but also in the extent towhich they intrude on the user’s experience.Issues with intrusiveness also arise when thisinformation is used. For instance, a user maynot necessarily want content in their browserautomatically updated whenever theychange position, or when they are busy withother activities.

This paper describes the ongoing developmentof a testbed to explore some of these methodsand issues by modelling a mobile learningscenario involving a visit to an art gallery. A userwith a mobile device will have the most relevantcontent made available as he or she movesaround the gallery.

Keywordscontext, location, mobile learning, wireless

1 Approaches todetermining locationAn important aspect of many context-awaremobile systems is the need to know where auser is, or what they are doing at a given time. Inaddition, we may be interested in a history oftheir movements; for example, if a user in agallery is studying a series of paintings by oneartist, we may be interested in which ones theyhave already visited, and perhaps in which order.

Determining location in context-aware mobile learning

Given reliable and accurate location information,we are also interested in how to use it in the bestand most unobtrusive way. Users may not wantto have to constantly request new content asthey move around the gallery, but they mayquickly become irritated if content isautomatically updated whenever the systemdecides they have moved enough to changetheir current context (Jones and Brown 2002).They may also be engaged in some other activity,which limits the amount of information they canaccept, something that may also be takeninto account by a context-aware system(Lonsdale et al. 2003).

1.1 ArchitectureThe location-sensing component of acontext-aware system should ideally beindependent of other components, and its outputshould be in a general, portable form that can beconsumed by a range of applications or othercomponents in the same system (Bauer, Beckerand Rothermel 2001). A location service may alsomanage information from more than one type ofpositioning system; for example, an ultrasoundpositioning system could be used in conjunctionwith Radio Frequency Identification (RFID) tags.Our testbed will use a web services architecturein which the location service periodically updatesthe context-aware system with the positions ofall the users it is tracking. Location information ispassed as Extensible Markup Language (XML),described by a simple schema, allowing location‘objects’ to be exchanged between componentsor services.

1.2 Types of position informationThe location information provided by a locationservice can be broadly divided into two maintypes: absolute and proximity.

Byrne Lonsdale Sharples Baber Arvanitis Brundell Beale 43

Will ByrnePeter LonsdaleMike SharplesChris BaberTheodoros N ArvanitisPat [email protected] of EngineeringUniversity of BirminghamBirmingham B15 2TTUnited Kingdom

Russell [email protected] of Computer ScienceUniversity of BirminghamBirmingham B15 2TTUnited Kingdom

With an absolute location, we have a position forthe user within the area or space covered by thelocation service. This could take the form of XYcoordinates, or perhaps a floor of a building or alogical area. This kind of position information ismore flexible than proximity position informationand can be more independent of other parts of acontext-aware system.

A proximity position only tells the system whichitem of interest a user is closest to, and includesno information about their position in a widersense. The position information could simply be,for example, the name of the picture or exhibit ina gallery that the user is interested in; however,this information is not very portable and couldonly be understood by specific applications orcomponents of a context-aware system.

1.3 Types of location serviceThere are also two main approaches todetermining position, which can be described ascontinuous and intermittent.

In a continuous location service, a user’s positionis known at all times as they move through thespace covered by the location service. Theservice is transparent to the user – that is, theydo not need to be concerned with how it is doneor provide any input into the system. A goodexample of this type of tracking is the Ekahau(2005) positioning engine, which uses the signalstrengths received by mobile devices on awireless local area network (WLAN) to calculatetheir position. The advantage of this kind ofsystem is that it is unobtrusive and providesmore information to a context-aware system,enabling it to offer a wider range of services. Acontinuous service would probably returnabsolute position information, varying in accuracyaccording to the system being used.

In an intermittent service, the user must activelyprovide the system with information about theirposition; for example, by passing an RFID readerover a tag, or perhaps lining up their device withan infrared (IR) beacon. This type of system ismore suited to an ‘object-based’ scenario like agallery or museum, in which individual objectscan be identified or tagged. The system has noknowledge of a user’s location until they (or theirdevice) interact with a tagged object and in effect,are explicitly requesting information about thatitem. In this case, the location service is moreobtrusive. This type of service might typicallyprovide proximity location information, although itcould also provide an absolute position.

The testbed will include location sensing ofboth types.

2 Using location informationeffectivelyIn general, a continuous service can providemore information to a context-aware system,enabling it to offer a wider range of services orbuild a more complete history of the user’smovements. For example, a user in our artgallery scenario might use their device to read anRFID tag, notifying an intermittent service thatthey are in front of a painting by Botticelli, and thecontext-aware system could then provideappropriate content for that picture. But withinformation from a continuous location service,the system could also point out works by otherartists that may be of interest when the userpasses them as he or she moves throughthe gallery.

However, the more obtrusive intermittentapproach to finding location does have asignificant advantage: we know a user isinterested in an object or service. A continuous,unobtrusive service can tell us that a user is nearan object, but this does not necessarily meanthat they are interested in it or in any associatedservices or content.

3 Aims of the researchSeveral methods and systems for locationsensing will be explored and assessed throughuser trials, with regard to three main criteria:

p the ability to deliver accurate and reliable locationinformation to the context awareness system

p the impact on the user’s experience as theymove through the gallery

p the need for the user’s device to have othercapabilities beyond wireless network access.

In addition, we will explore other issues arisingfrom the use of location-aware systems,in particular:

p how to distinguish interest from simple proximity

p how to incorporate a ‘location history’ into context

p how to infer a change of position in thecontextual sense from changes in physicallocation reported by a location service

p how location-driven changes to content orservices can best be presented to a mobile user.

Our testbed aims to simulate a visit to an artgallery; however, most of these points will beapplicable to other context-aware servicesand scenarios.


ReferencesBauer M, Becker C and Rothermel K (2001).Location models from the perspective ofcontext-aware applications and mobile ad hocnetworks. In Proceedings of the Workshop onLocation Modelling for Ubiquitous Computing,Atlanta, Georgia, 35–40.

Ekahau (2005). Website. At www.ekahau.com

Jones GJF and Brown PJ (2002). Challengesand opportunities for context-aware retrieval onmobile devices. In Proceedings of the SIGIRWorkshop on Mobile Personal InformationRetrieval, Tampere, Finland, 47–56.

Lonsdale P, Baber C, Sharples M and Arvanitis T(2003). A context-awareness architecture forfacilitating mobile learning. In Proceedings of theMLEARN Conference, London, 19–20 May.London: Learning Skills and DevelopmentAgency, 36–37.

Small J, Smailagic A and Siewiorek D (2000).Determining user location for context awarecomputing through the use of a wireless LANinfrastructure. Pittsburgh, PA: Institute forComplex Engineered Systems, CarnegieMellon University.

Byrne Lonsdale Sharples Baber Arvanitis Brundell Beale 45

AbstractIn the HISS (Hospital Information System forStudents) project, run by the CampusBio-Medico, a university in Rome, students ofmedicine, nursing and dietetics practising in thewards were trained to use handheld devicesconnected through a wireless local area network(WLAN) to record patients’ data. Besides learningthis new technology and applying it to accessfreely teaching resources from any place on thecampus, the students were able to design newuser interfaces for accomplishing daily tasks. Thework done by dietetics students was a goodbasis for the development and implementation ofa real-life solution at the University Hospital.

Keywordsmobile training, ubiquitous computing, adaptiveuser interface, impact of innovation

1 IntroductionLearning on the job has always been considereda fundamental approach for medical-relatedprofessions. An important part of the teaching isaccomplished on the wards, while visitingpatients. The typical way of memorising what issaid or done by the teachers, nurses or physiciansis to take a written note in an exercise book. Thisleads to unstructured data and makes it difficultto access specific information quickly. Furtherwork is usually needed to reorganise the notes ina practical way for easy recovery of any part of it.

At the Campus Bio-Medico, a university in Rome,we introduced wireless networks and portabledevices at the beginning of 2003, starting anumber of projects for assessing the use of this technology. One of them emphasises the use of handheld computers (Palmtop and Tablet PCs) for training students of medicine, nursing and dietetics. The project wasfinanced by an HP Applied Mobile TechnologySolutions in Learning Environments grant.

The impact of innovation in medical and nursingtraining: a Hospital Information System for Students(HISS) accessible through mobile devices

This was the only Italian project in a list of40 projects funded worldwide. It was presentedat the HP Labs in Palo Alto, California inSeptember 2003 to all the other institutions thatreceived a grant.

We installed wireless devices so that from everyroom of every ward a connection could be madeto a separate LAN – different, for securityreasons, from the HISS wireless local areanetwork (WLAN). Each student participating inthe project was equipped during trainingsessions with a WLAN-enabled HP iPAQ 5500capable of fingerprint authentication. Some ofthem, especially the teachers (nurses andphysicians), used an HP Tablet PC.

Attention was paid to feedback: a system wasset up for collecting comments, bug notices,proposals and other information from the users.Experienced tutors closely followed the monitoringactivity so that quantitative data was integratedwith qualitative evaluations directly gatheredfrom observation of student and teacher activity.

We had many goals: first of all, we wanted toteach our students the new technologies theywill encounter in the future while working inhospitals. We also wanted to give them a bettertool for learning the medical topics they weredealing with on the wards. We also wanted themto define the user interface for medicalapplications on handheld computers. This provedpopular with students and medical staff. Insteadof involving in the interface design actual nursesand doctors, whose time is expensive and whoare always busy in their daily tasks, we usedfeedback from the students to develop an actualHospital Information System (HIS) accessiblethrough mobile devices.

From the learning point of view, we wereinterested in examining whether the studentsusing handheld computers were achieving betterresults in their examinations than those takingnotes in the traditional way.

Cacace Cinque Crudele Iannello Venditti 47

Filippo CacaceMaria CinqueMichele CrudeleGiulio IannelloMarco Venditti

Campus Bio-MedicoVia E Longoni 8300155 RomaItalyhttp://research.unicampus.it/HISS

[email protected]@[email protected]@[email protected]

We soon realised, however, that this last goalwas too complex to accomplish because toomany factors are involved in the learning phase,and comparing groups with and withouthandhelds requires many students and anenormous amount of data.

2 Project development

2.1 A task-oriented solutionOur first chore was to convert the written noteinto an Electronic Patient Record (EPR) suitablefor handheld computers. We developed thestructure and contents of the EPR by studyingsome existing models like Ward-in-Hand(Ancona et al. 2000) and Bedside Florence(Policlinico Gemelli 2002) and by addressing thespecific needs of our 120-bed University Hospital.We held regular meetings with the teachers,tutors, physicians, nurses and dieticians in chargeof their departments. We also followed the dailyhospital activity of the three main groups ofactors: physicians, nurses and dieticians. Weconcentrated on the tasks that might require amobile device: we selected all the informationthat they usually wrote at the bedside, excludingtherefore other longer data, such as the patient’sadmittance and discharge letters which can bebetter written on a desktop computer.Furthermore, we based our analysis on theexisting paper models in order to reduce theimpact of innovation and achieve a higher degreeof acceptance by both teachers and students.

2.2 An adaptive users’ interfaceAfter defining the main roles of the peopleinvolved and the tasks that they normally carryout, the main problem was content adaptation,mainly due to challenges posed by the featuresof the mobile devices and the frequent changesin the interface definition. We were aware thatthe development process in writing the softwarehad to be adaptive, because the students woulddiscover new solutions and re-define thestructure many times.

Since we were not bound to a real productionschedule and we had no strict deadlines, wewere free to try different solutions [online andoffline; XML and RDBMS (Relational DatabaseManagement System); access trough WLAN,GPRS, UMTS; interface adaptation for Pocketand desktop PCs]. Eventually we based oursystem on ASP.NET, C#, XML and SQL Server.

The web-based application, accessible through awireless LAN (WLAN), included both generic andspecific data-entry masks. By using XML, webuilt 30 different masks by simply combininga few tags: <Section>, <Title>, <Voice>,<VoiceName>, <Value> for the differentstructural parts; <SmallText>, <MediumText>and <BroadText> for data input; <Drop>corresponding to the object DropDownListin ASP.NET and <Check> corresponding toCheckBoxList in ASP.NET (see Figure 1 opposite).

The possibility of rapidly changing the contentsvia XML schema, without varying the code,allowed us to improve the application day by day:for example, in 10 days of work with thephysicians, the patient history module changedfour times and the general examination modulefive times.

Apart from content adaptation, we carefullystudied the way in which to present it to theusers. The design involved ergonomic andtechnical factors. The interface, very simple atthe beginning, was enriched thanks to users’feedback. The main innovations were as follows:

p multiple sections masks to avoid overlongHTML pages

p substitution of text boxes with drop-down lists orchecklists of options

p distinction between frequently and lessfrequently used options

p text boxes of different sizes (small, medium andlarge) at the end of the lists to insert additionalinformation (values, observations, descriptions)

p different modules to modify or read therecorded information

p an automatic view of the clinical informationrequested by a specific department.

In the final implementation, there are three waysof collecting information from a patient: writeinto a free text box; search a (complete or partial)string in the database; answer a multiplesections mask with drop-down lists, checklists ortext boxes based on XML schema language.

Our next goals will be: full adaptation capabilityand context awareness – that is, the contentserver could transparently deliver information in aformat suitable for a specific device and user;and standardisation of the recorded clinicalinformation, beginning from the classification ofpathologies – which is now based on the Italianversion of ICD9–CM (US Public Health Servicediagnostic code) – in order to apply HL7 (theHealth Level Seven standard). We are aware thatthe final implementation of any record systemmust be based on actual tasks and attitudes inthe particular healthcare environment(Rossi Mori, Consorti and Ricci 2000).


Figure 1 XML data entry visualised on the Pocket PC

2.3 Ubiquitous accessTo improve network performance and availability,we tested HISS access through heterogeneousnetworks. Our approach was based on theemerging Mobile IPv6 protocol, which providesthe following benefits: management of mobilityat the network layer, allowing networkapplications to be unaware of changes in thenetwork; ease of use and configuration so thatthe user does not need to change anything whenmoving from local to cellular wireless networks.

Our Mobile IPv6-enabled clients can nowseamlessly access HISS through both wirelessaccess points and GPRS or UMTS VPN (VirtualPrivate Networks) without changing the sessionor relogging. A network interface monitoringmodule in the client device traces the signal levelof wireless interfaces, as well as the presence ofwired connections, and triggers the handoffsfrom one interface to the other when needed. Inorder to minimise changes to the existinghospital network and to allow access to IPv4hosts, we developed a NAPT–PT (NetworkAddress Port Translation + Protocol Translation)IPv6–IPv4 transition mechanism that translatesIPv6 data packets into IPv4 packets and vice versa.

This allows a transparent communicationbetween IPv6 nodes (mobile clients of the HISSenvironment) and the IPv4 nodes (legacyplatforms of the existing Hospital InformationSystem). The main benefit of this approach issimplicity: the mechanism is easy to configure,since clients need only to use a DNS server thatsupports IPv6 addresses.

We also evaluated transport protocolperformance during handoffs amongheterogeneous networks as well as usability(user perception of network performance in areal environment, wide-area access to HISS sodoctors can follow up on patients from home,etc). So far, we have implemented mobile accesson Tablet PCs running Linux, since WindowsMobile does not yet completely support allmobility protocols. In the future, we plan toextend it to other operating systems, devices andinterconnection technologies (eg Bluetooth).


2.4 Formal training andinformal educationWe monitored students using palmtop devicesduring their training in the hospital, which is aquite different activity from learning in aclassroom. As previously stated, we did notmonitor their improvement in learning, but wedid analyse how they managed the new devicesand what effect this training had on activitieswithin in the hospital.

Since the system is not a mere simulation, but adatabase with real cases, access to the data wasrestricted to students and their teachers, strictlyfollowing the Italian privacy regulations. At first,the students were trained to convert paperrecordings into electronic recordings, startingfrom data retrieved from the actual HospitalInformation System. In the second phase, theyoperated directly at the bedside, whileadmitting patients.

We tried different ways of collecting data (onlineand offline) and we used many iPAQ 5500s andsome Symbol PPT 2800s. We found that userstend to prefer a smaller device, with colourdisplay, to a faster one (usability seems moreimportant than performance). The comparisonbetween offline and online activity offeredinteresting results. In the offline case, thesoftware on the handheld device requiredsynchronisation with a desktop PC. Thus, the‘back office’ work was much harder. But from thepoint of view of the users, the offline solutionavoided all the problems related to networkperformance. Therefore we envisage a newscheme for the future: data is saved offlineanywhere and automatically synchronised assoon as the device is on an active cell.

Apart from the modules created for the project,which were accessible online, the studentslearned to use all the software installed on theiPAQ 5500 (including recording verbal notes) andother trial applications (HanDBase, MedicalPocket PC) which were selected and installed bybioengineering students. We also acquiredPocket PC versions of medicine manuals, like theWashington or Harrison manuals of medicaltherapeutics. An external company wascommissioned to create a specific databaselisting the commercial names of drugs and howthey work/should be prescribed, and this waspersonalised according to the users’ needs.These resources were very useful, not only forthe students, but also for their teachers: it was anapplication of continuing medical education.

In recent years we have been producingstreaming video-based lessons for nurses inorder to give them access to university-levelresources and the capability of reaching theLaurea level (Campus Bio-Medico was the firstItalian university to establish a university-levelcourse for nurses). We later adapted thesevideos in order to stream them on handheldcomputers so that a nurse in a ward, during arelatively calm period (such as a night shift) caneasily access the video-on-demand library oflessons. Campus Bio-Medico has also producedvideo-based tutorials for the VAKHUM project,funded by the European Commission (EC), whichdeals with 3D kinematics (study of motion) of thehuman skeleton (Van Sint Jan, Crudele andGashegu 2003).

3 Users’ feedback onwireless trainingThe evaluation methodology was based on users’observations, interviews with them andquestionnaires filled in by students and theirtutors. Furthermore, we tried to monitor theusers’ feedback using categories created byEverett Rogers, a theorist who spent over 30years studying the diffusion of innovations, fromqwerty keyboards to new agricultural methods indeveloping countries. According to Rogers(1995), the characteristics of an innovation, asperceived by the members of a social system,determine its rate of adoption. The five attributesof innovations are: (1) relative advantage;(2) compatibility; (3) complexity; (4) trialability;(5) observability.

Interesting results came from a test submitted toall the first-year dietetics students. The studentswere asked to indicate the main advantages anddrawbacks of using a Pocket PC instead of apaper questionnaire. The majority of themindicated two main advantages: speed in findingthe answers; and time spared in the transcriptionof data from paper to PC. Usability was alsoindicated as one of the factors in favour of ahandheld device: the palmtop computer doesnot need a stand and it is not ‘uncomfortable’(ie not heavy to carry in comparison to anormal PC). Furthermore, 80% of the studentsinterviewed thought that the presence of akeyboard would not help them in data entry.Thethree main requirements for acceptanceof an innovation (advantage, compatibility andacceptable complexity) were therefore fulfilled.


The other two indicators: trialability (the degreeto which an innovation may be experimentedwith on a limited basis before making anadoption/rejection decision); and observability(the degree to which the results of an innovationare visible to others) – were considered lessimportant by the students interviewed.

Among the nursing students, great importancewas given to the ‘relative advantage’ factor. Inthe first phase, only 60% of them felt that thethe use of PDAs was advantageous to them.Since the attitude of some tutors towards theHISS project was negative, the students’ feelingsmay have been biased. In this case, we had towork on the tutors to achieve good results withthe students. Only when the tutors began toperceive the innovation as being better than theactivity it superseded (writing on a piece of paperand then rewriting it on a PC, or delivering it tothe physicians), did they motivate the students touse the new system.

A crucial indicator for PDA acceptance was‘accuracy’. We demonstrated to the tutors thatstudents using handheld devices for data entry instructured masks were more accurate than thosewriting on a blank piece of paper; the first groupnoticed more things (having different questionsto answer) and were more precise.

The most critical response came from themedical students. The complexity of the tasksand the different approach to data entrycontributed to a very low degree of acceptanceof the new technology. Therefore, after anineffective extensive phase (involving all thestudents in all the wards), we tried an intensiveapproach. A pilot project started in thedepartments which had shown a more positiveattitude during the first phase – general surgeryand cardiology. All the medical staff wereinvolved, not only the students and their tutors.This phase showed that the presence of aleading authority figure is a key element for theacceptance of innovation: the fact that thedirectors of both departments were keen onusing the devices motivated all the staff.

4 ConclusionsThe HISS project, which was carried out in ouruniversity campus during the academic year2003/04, achieved the following goals:

p enhancing the level of technology in the hospitalby improving accessibility to the informationsystem at different levels (students, nurses,physicians) through mobile technologies

p improving teaching and learning in the wardsthrough a faster access to clinical data

p designing new interfaces for small devices forcollecting and examining data at the bedside

p achieving a deeper comprehension of patientdata security issues

p facilitating the requirement analysis ofmobile applications

p helping with performance evaluation.

The iPAQs were used by 110 students (indifferent departments), recording data relating tomore than 1500 patients to accomplish 30different tasks. The medicine students entered495 records; the nursing students 243; thedietetics students 193; while their tutorsentered 919.

The positive effects of the project went beyondour expectations. Two companion projectsstarted, driven by the enthusiasm of somemembers of staff: the first one, for the surgerydepartment physicians, has completely changedthe way that rapid data entry is done at thebedside – previously this was done on a sheet ofpaper placed on a wooden tablet. The secondproject, in collaboration with the CampusInformation System software developers, iscarrying out the complete conversion of therecord of all dieticians’ activities (such asbedside–kitchen communication) to anelectronic version.

We feel that other institutions may benefit fromour experience by being encouraged to use theirstudents as generators of specifications forreal-life software applications.

Besides being the acronym of a system forstudents, HISS became a powerful metaphor:the ‘hiss’ – ie the whisper, the buzz – spreadand involved more people and more tasksthan expected.

5 AcknowledgmentsWe thank HP for providing all the hardware forthe project. Special thanks are due to AndreaRiccio, who wrote most of the software for HISSand produced his Laurea thesis with the resultsof the project. Thanks also to all the tutors whopatiently followed their students while workingwith the handheld devices.


ReferencesAjzen I (1980). Understanding attitudes andpredicting social behavior. Englewood Cliffs (NJ):Prentice–Hall.

Ancona M, Dodero G, Gianuzzi V, Minuto F andGuida M (2000). Mobile computing in a hospital:the WARD-IN-HAND project. In Proceedings ofthe ACM Symposium for Applied Computing,Como, 19–21 March, 554–556.

Bernaschi M, Cacace F, Cinque M, Crudele M,Iannello G and Venditti M (2004). Interface designand mobility in ubiquitous access to HIS. Paperpresented at the Medicon Conference, Ischia,31 July–5 August.

Karahanna E, Straub D and Chervany N (1999).Information technology adoption across time: across-sectional comparison of pre-adoption andpost-adoption beliefs. MIS Quarterly, 23, 183–207.

Munoz MA, Rodriguez M, Martinez-Garcia AI andGonzalez VM (2003). Context-aware mobilecommunications in hospitals. IEEE Computer,36(9), 38–45.

Policlinico Gemelli (2002). Wireless ward system allows nurses to record and retrievedata instantly at patients’ bedside. Atwww.microsoft.com/resources/casestudies/CaseStudy.asp?CaseStudyID=12245, accessed31 January 2002.

Rogers E (1995). Diffusion of innovations.NewYork: Free Press.

Rossi Mori A, Consorti F and Ricci F (2001).Sharing clinical information. From local EPR totask oriented solution. In Proceedings of the4th European Working Conference on ElectronicHealth Care Records, Aix-en-Provence,5–6 November.

Van Sint Jan S, Crudele M and. Gashegu J(2003). Development of multimedia learningmodules for teaching human anatomy:application to osteology and functional anatomy.In The Anatomical Record Part B: The NewAnatomist, 272B(1), 98–106.


AbstractIn this paper we present our work on theextension of a state-of-the-art e-learningsystem – the Intelligent Web Teacher (IWT) – tosupport multimodal mobile access in order tooffer a complete set of learning experiences,services and models that are able to fit thecomplex and variegated mobile world. Theextended platform can offer customisede-learning experiences depending on the typeand capabilities of the user’s mobile device. Aftera brief overview of the IWT e-learning platform,the paper describes how we approached theextension of IWT to provide browser-, SMS-and voice-based interactions. Some firstexperimental results are then given in the lastsection. The work described here was realised inthe context of the EC-funded m-learning project.

Keywordse-learning, ITS (Intelligent TransportationSystems), mobile technologies, SMS, IVR

1 IntroductionThe spread of the use of mobile devices andtechnologies offers great opportunities for thee-learning community. New scenarios, didacticalmodels, learning experiences and services haveto be designed for this new category of user.Learning may now become, as never before,‘when you want’; but also, and this is thepotential of mobile technologies, ‘where youwant’ – in the street, on the underground train,when it is necessary or when there is enoughtime. Moreover, mobile technologies permitprovision of continuing education/training tothose users who are not taking part in traditionaleducation, but have a mobile phone and use itevery day.

A system for adaptive platform-independentmobile learningNicola CapuanoMatteo GaetaSergio MirandaLaura [email protected]@[email protected]@crmpa.unisa.it

Centro di Ricerca in Matematica Pura ed Applicata (CRMPA)Via Ponte don Melillo84084 Fisciano SAItalywww.m-learning.org

The mobile users’ community is a complex andvariegated world, characterised by differentdevices and technologies which are constantlychanging. Technologies range from highlydiffused SMS (Short Message Service) to MMS(Multimedia Messaging Service); from WAP(Wireless Application Protocol) to HTTP (HyperText Transfer Protocol); and from GSM (GlobalSystem for Mobile Communication) throughGPRS (General Packet Radio Services) to UMTS(Universal Mobile Telecommunications System).Devices vary from mobile phones to personaldigital assistants (PDAs), from new-generationSmartphones to laptop computers.

Our aim is to extend an existing e-learningplatform, the Intelligent Web Teacher (IWT), inorder to allow the design and delivery ofsignificant learning experiences via a huge set ofmobile technologies. This was done in thecontext of the EC-funded m-learning project.

2 The IWT e-learning platformEvery didactic/formative context requires its ownspecific e-learning solution: not only with regardto the content, but also the didactical model,formative modules, applications and tools. Thesenecessities usually imply hard work for analysts,engineers and developers. IWT (Capuano, Gaetaand Micarelli 2003) solves the problem, thanks toan extensible solution which can become thefoundation of a virtually infinite set of applicationsfor either traditional or innovative e-learning.

Conceptual and technical choices in design anddevelopment have brought about an architecturethat can be extended both in terms of servicesand learning object typologies through theinstallation of added components. Indeed, IWTallows the user to add to the platform new serviceswritten ad hoc for the system by previousimplementations of suitable plug-in software thatis compliant with system specifications.

Capuano Gaeta Miranda Pappacena 53

As illustrated in Figure 1, the general IWTplatform offers a set of generic managementservices for content, courses, collaboration andadministration. Some of those services are fixed(general base platform) while others are optional(general added resources).

Figure 1General IWT architecture

IWT allows the user to manage contents likelearning objects, tests, business games, etc.Furthermore, given the extensibility of theplatform, it is possible to add new typologies ofresource at any time, without altering thesystem. IWT also allows the users to managecourses. Two kinds of courses are supported:simple and intelligent. The intelligent course(Capuano et al. 2002) can be customisedaccording to the real needs and preferences ofthe students and fully applies the principles ofstudent-centred learning.

Given a set of didactic objectives chosen by theteacher on the domain ontology, the intelligentcourse is able to generate the best learning pathfor each student starting from its student model,based on acquired knowledge about the studentand his/her learning preferences. Differentstudents with the same didactic objective willthus have different courses generated bythe system.

IWT also allows the users to collaborate andcommunicate with each other via messages,discussion forums, chat, content sharing,videoconferencing, etc. IWT applies threedifferent models: the knowledge model, thestudent model, and the didactical model and isparadigm-based on a pattern adapter to allowextensibility. IWT is built completely usingMicrosoft .NET technology.

3 Browser-based interactionUsers who have HTML-enabled mobile devicesmay access a subset of IWT functionalities.Content and layout (see Figure 2) areautomatically adapted by the IWT engine whichrecognises what kind of device users areemploying for the connection. This was achievedby adopting the Microsoft Mobile Internet Toolkit(Lee 2002) which extends the Microsoft .NETFramework with server-side technology, allowingthe delivery of content to a wide variety ofmobile devices including mobile phonessupporting Wireless Markup Language (WML)and compact HTML (cHTML), HTML pagers andPDAs like the Pocket PC.

Figure 2IWT interfaces for mobile devices

Supported IWT functionalities on the mobilebrowser include course subscription, coursedelivery and basic collaboration features(like chat, forums and messaging). Suchfunctionalities are synchronised with thoseavailable on the PC web browser. It means thatpeople are able to communicate with each othereven if they are using different kinds of device.

4 SMS-based interactionSMS is the most common and frequently usedmobile service: it is present in every kind ofmobile device and offers the possibility ofreaching all mobile users. Our idea was to buildcourses composed of SMS ‘pills’ – short textuallearning objects – together with multiple-choicetests delivered by SMS. The learner can answertests by simply replying to the test SMS questionwith an SMS containing the answer.The systemtracks the answer received from every singlelearner, verifies the results and sends him or hera new SMS containing the test results andsuggesting improvements.


Figure 4 Adoption of voice-based interaction in IWT

Figure 3 Extended IWT architecture for SMS support

To allow the management of these new types oflearning object, we created several new IWTresources (see Figure 3): SMS Test, SMS Textualand SMS Course. To allow the learner tosend/receive SMS, we have extended the IWTservices with an SMS router module thatinteracts with an SMS gateway. An SMSgateway is a simple SMS-enabled mobile phoneor an SMS server accessible as a web service.

5 Voice-based interactionLast but not least, the system is also able toprovide voice-based interaction. The learner canaccess the system with a normal phone call andcan navigate the IWT portal using IVR (InteractiveVoice Response). A synthesised human voicereads a simplified version of the portal contentand the user can browse by dialling on thephone keyboard.

This access method was achieved by connectingto the IWT general base platform asoftware-based IVR facility linked to thetelephone network through an H323 gateway(see Figure 4). A new version of the IWT portalwas then provided in the form of a navigationtree. Each node of such a tree represents thecurrent position of the learner, while each branchis a possible choice at that point. Someinformation features (bulletin and news reading,course lists) and collaboration features(messaging, voice chat with a tutor) were thenimplemented using voice-based interaction.

Capuano Gaeta Miranda Pappacena 55

6 Preliminary experimentationresultsThe mobile browser-based interaction describedin section 3 is under trial, with around 200 usersfrom the UK, Italy and Sweden taking a set ofcourses on literacy and numeracy.

In collaboration with Albatros, an organisationfrom South Italy that works for the socialintegration of foreign and dialect-speakingpeople, we are currently starting a new trialbased on an SMS first-level Italian languagecourse, again using voice-based interaction forcollaboration activities.

The final results of both trials are not availableyet. The first impression is that users aregenerally highly motivated when using mobiletechnologies for learning, with a particularemphasis on SMS technology, which seemsto be more effective and simple thanbrowser-based interaction which still presentsconnection problems.

ReferencesCapuano N, Gaeta M and Micarelli A (2003).IWT: una piattaforma innovativa per la didatticaintelligente su Web. AI*IA Notizie, XVI(1), 57–61.

Capuano N, Gaeta M, Micarelli A and SanginetoE (2002). An integrated architecture for automaticcourse generation. In Proceedings of theInternational Conference on Advanced LearningTechnologies (ICALT), Kazan, Russia, 322–326.

Lee W (2002). Developing mobile applicationsusing the Microsoft Mobile Internet Toolkit.At http://devx.com/wireless/Article/10148,accessed 2 December 2002.


AbstractCommunicating is what mobile technologiesdo best!

Feedback from learners and tutors in them-learning project trials has taught us that usersenjoy the content, but they love the collaboration.

One of the key challenges that the m-learningproject addressed has been the development ofa collaborative learning tool called mediaBoard.This allows learners to collaborate using text andpicture messaging, contributing to a sharedwebsite that everyone has helped to make.

In this paper, we will review howmediaBoard works.

Keywordsmobile learning, mediaBoard, MMS, SMS,picture messaging, mobile phone, collaboration,communities of practice, P900, XDA2, PDA

1 The m-learning projectThe m-learning project is a 3-year pan-Europeancollaborative research and development (R&D)programme supported by the European Union(EU). The consortium is a partnership oforganisations – the Learning and SkillsDevelopment Agency (LSDA), Cambridge Trainingand Development (CTAD) and Ultralab from theUK; Lecando from Sweden and the Centro diRicerca in Matematica Pura ed Applicata(CRMPA) from Italy – combining skills inpedagogy and technology.

The project has developed prototype productsand innovative approaches designed to supportlearning, particularly literacy, numeracy and lifeand survival skills, using handheld devices suchas mobile phones and palmtop or pocketcomputers. A key objective is to engage with andmotivate young adults who are not engaged ineducation or training, particularly those who areunemployed or homeless.

Mobile learning = collaborationJo ColleyGeoff [email protected]@ctad.co.ukCambridge Training and DevelopmentLincoln HouseThe Paddocks347 Cherry Hinton RoadCambridge CB1 8DHUnited Kingdomwww.m-learning.org

The project has funded the purchase of 100mobile devices (P800s, P900s, XDA2s), whichwere used by 300 learners in the threepartner countries.

2 Learning with the mediaBoard

2.1 ContextThe mediaBoard (see Figure 1 overleaf)encourages learners to use a mix oftechnologies – mobile phones (with and withoutcameras), SMS, MMS, mobile and regular webaccess – to contribute to online projects. Tutorscan set up their own mobile event or project.These projects can be competitive (treasure-huntstyle), collaborative (with differentiated tasks) orreflective [using the mediaBoard as a diary orweb log (blog)]. At the time of development,there were no commercial alternatives. Recentlyseveral web logging (blogging) websites havesprung up which accept text and picturemessages. These deal with the same technologychallenges, though they use a different ‘play’model, and there is less emphasis on the notionof a community of practice. This makes for aninteresting comparison with our thoughts on themediaBoard.

Current users include travellers, asylum seekersand inner-city learners from the UK and Sweden.

Colley Stead 57

Figure 1Sample mediaBoard

2.2 Learning‘Content isn’t king – community is sovereign’(Heppell 2003). The learning developed by theproject falls broadly into two categories:individual (tests, quizzes, self-assessments) andcollaborative (the mediaBoard). Collaborativelearning allows for a more cohesive, socialconstructivist style of learning.The mediaBoardsupports a ‘community of practice’ approach, inwhich a group of learners feels involved and isable to use the technology as a facilitator for itsown creative ideas (see Lave and Wenger 1991;Wenger 1998). The learning can be made morefruitful as it is extended over time, with sessionsfor preparation followed by sessions that use theinformation gathered for further activities.Learners’ affective and social needs areaddressed by engaging the imagination andpromoting exploration and creativity within anon-threatening environment. This concept andpractice of learning is offered as acounterbalance to the current UK focus on amore target-oriented approach to learning; andalso as a means of engaging learners for whommore traditional methods have failed.

2.3 The challengesChallenges are both technical – some very ruralareas do not have sufficient connection to anetwork/supplier to make use of the mediaBoardfeasible; and practical – there is an expectation inlearning of this kind that both tutors and learnerswill be enthusiastic users of new technologies,but this is not always the case. Training of tutorsand learners is a key element in this style oflearning. There is also the challenge of developinga community of practice: learners need todevelop their own agendas for use of themediaBoard, so that they are genuinely involvedin their own knowledge creation.

3 Case studiesCase studies that have been conducted using themediaBoard include:

p The Southampton Timeline: a mediaBoard wasused to explore the history of immigrationto Southampton

p College Open Day: this case-study used amediaBoard as part of an induction week ina college

p An Italian Park: a mediaBoard was used forlanguage learning.

4 ReferencesHeppell S (2003). Keynote speech. MLearnConference, 19–20 May, London, UK.

Kowch E and Schwier R (1997). Characteristics oftechnology-based learning communities. Collegeof Education, University of Saskatchewan. At www.usask.ca/education/coursework/802papers/communities/communities.htm

Lave J and Wenger E (1991). Situated learning.Legitimate peripheral participation. Cambridge:University of Cambridge Press.

Rogoff B, Turkanis CG and Bartlett L (eds) (2001).Learning together: children and adults in a schoolcommunity. New York: Oxford University Press.

Wenger E (1998). Communities of practice:learning as a social system. Systems Thinker,June issue. At www.co-i-l.com/coil/knowledge-garden/cop/lss.shtml


AbstractThis paper describes a trial of Tablet PCs withstudents in higher education. The participantswere provided with a set of collaboration andknowledge-management tools, and encouragedto explore the Tablet PC as support for informalaspects of their learning, particularly incollaborative project work.

KeywordsTablet PC, informal collaboration, higher education

1 IntroductionA study was undertaken to explore thepossibilities afforded by wireless Tablet PCs inhigher education, with particular regard to theinformal and collaborative uses of the device.The work is complementary to, but separatefrom, other research into using technology informal learning situations, such as ActiveClass(Ratto et al. 2003). Building on the results of anearlier project using Pocket PCs (Corlett et al. 2004),this trial set out to create an environment in whichbest use of the tablet technology could be achieved.Students recruited to the trial were given a rangeof software and hardware tools for an open-endedinvestigation of their learning activities.

2 Method

2.1 ParticipantsSeventeen students were recruited to the trialfrom a third-year engineering course in which, aspart of the course, they form teams of six orseven to conduct a year-long design project. The project requires effective collaboration and a high level of cooperation on the part of thestudents. Three teams opted to join this study,each member receiving a Tablet PC to use astheir own until the end of the academic year.

Tablet technology for informal collaboration in higher educationDan CorlettMike [email protected]@bham.ac.ukEducational Technology Research GroupElectronic, Electrical and Computer Engineering DepartmentUniversity of BirminghamEdgbaston B15 2TTUnited Kingdom

In addition, the teams’ supervisors and mentorsalso received a Tablet PC and could use this tocommunicate and work with their teams. A furthertwo students from the same cohort joined thisstudy, but without the rest of their team.

2.2 System specification andsupporting infrastructureThe Tablet PCs used in the study were a mixtureof clamshell and slate-only form factors. Allstudents within a specific team were equippedwith the same model. External keyboards wereprovided with the slate-only tablets. Additionally,one team was equipped with audio headsets anda USB ‘webcam on a stick’. One team startedwith the slate-only form and then changed toclamshell for comparison.

Software provided to the students included theentire Microsoft Office 2003 Professional suite,plus OneNote (tablet-optimised note-taking andorganising tool). Mindjet MindManager for TabletPC was installed to provide concept-mappingfeatures that integrate with the Office products.Students were free to install any furtherapplications of their choice.

In response to the outcomes of the earlier trial, acentral portal (Microsoft SharePoint) was set up toprovide technical support, course information andcommunication tools. Each team also had a privateportal site with discussion, document-sharingand tracking, time-management and publishingtools. The private portal helped to providecohesion and a persistent workspace to providethe teams with greater mobility.

A wireless network (802.11b) is availablethroughout the department and across some ofthe open spaces on campus. The departmentoperates an Exchange Server for personal andshared information management. Each participanthad an Outlook client to connect to this.

Corlett Sharples 59

2.3 Data collection and analysisData was collected from students viaquestionnaires issued before, during and afterthe trial. Questionnaires aimed at particular usersor activities (eg people who additionally use aPDA) were issued on an ad hoc basis. Studentswere also encouraged to provide feedback at anytime. Observations, particularly of teamsessions, were conducted and focus groupswere used after the trial to expand on problems,observations and suggestions raised in thequestionnaires or feedback.

3 Outcomes and resultsInitially, participants were harder to recruit to thetrial than had been anticipated. Concernsincluded taking responsibility for expensiveequipment, time needed to be spent incustomising the Tablet PC and learning to usenew software, and simply not being able tovisualise the benefits. All participants ownedeither a desktop or laptop PC, in which they hadalready invested time and money, and so somesaw no need for another computer.

Learning support for the project was describedas good by all participants – the only suggestedimprovement was more tutorials for help in usingunfamiliar software.

The teams that took part varied in theirenthusiasm and willingness to try new activities.However, the team that initially showed mostreluctance about taking part became the mostspontaneous innovators. This group used internettext and audio messaging to liaise with theirsupervisor and communicate with one another.They developed technology-enhanced teammeetings in which members used their Tablet PCas a shared whiteboard and collected audio andvideo that could be embedded in their notes.

3.1 Generic usability issues andform factorThe study revealed the obvious usability issuesrelating to hardware and software, ranging fromweight and battery life to slow performance andunreliable synchronisation of files to thedepartmental file server. Several studentsintended to use their device out of doors, withinthe wireless network, but found that screenbrightness made this impossible. Users weredisappointed not to have a full choice ofcustomisation, though found it useful onoccasions to have the same look and feel on thedepartmental desktops as on the Tablet PC.

There were considerable differences betweenthe slate-only and clamshell designs. All studentsreported that a keyboard is fundamental tomaking full use of the device. Five of the sixstudents who switched to a clamshell part of theway through claimed that they used it more andcarried out a wider variety of tasks than with theslate-only device. The clamshell offered betterfeatures, greater comfort and more flexibility.

Six students were equipped with a wireless Pocket PC in advance of the Tablet PC trial andwere allowed to keep it throughout to make acomparison. They all stopped using the Pocket PCcompletely after receiving the Tablet PC, citingthe relative usefulness of the tablet as the mainreason, despite the advantages of the smaller formfactor with the Pocket PC in certain situations.

3.2 Patterns of use and collaborationBy the end of the project, four of the 17 studentshad stopped using their Tablet PC. Three of thesehad only done so once all timetabled teachingwas complete, with only one giving up beforethat. All four had the slate-only model. Tenstudents were still using their Tablet PC at leastonce a day until they handed it in.

The most popular activities were: (1) e-mail;(2=) creating documents, browsing, listening tomusic; (3=) reading, taking lecture notes, takingmeeting notes; (4) watching video; (5) programming;(6=) internet text messaging, annotating slides;(7=) keeping a record of work, managing time.

Prior to the study, all teams had set up aninternet group account for discussion anddocument sharing. All teams switched to usingthe SharePoint Portal, since it offered greaterdata capacity and better customisation. Sharedfiles could also be created and accessed directlyfrom the productivity suite on the Tablet PC.While discussion was also popular, other featuresof the portal such as calendar, contacts andannouncements were not widely used. Bothfocus groups unanimously agreed on theusefulness of such a team workspace. Instantinternet messaging was used by all participantsfor learning and leisure. Multi-user chat was usedby seven participants.

Team meetings were observed and followed upwith a short interview. Activities introduced tothe meetings as a result of the technology were:instant minute-taking and sharing; making anaudio recording of the meeting (especially ofvalue to those working in a second language);using a shared whiteboard; show-and-telldemonstration of project deliverables; filetransfer; and immediate follow-up of requests fore-mails and other information.


3.3 Analysis of applications/and toolsDespite frequent complaints that the PCs wereslow and lacked memory, a number of studentsinstalled a range of powerful programming andimaging tools that they used on a regular basis.Only one student claimed to have installed anygames, though nine out of 17 placed a mediaplayer in their five most frequently used pieces ofsoftware. Many watched videos downloadedfrom the internet, with two admitting to doing soduring lectures.

OneNote, the note-taking tool, was listed amongthe five most frequently used tools by 16 of the17 respondents, with seven claiming it to be themost useful individual tool.

Participants were asked to list the top threebenefits of having had a Tablet PC, and thesebenefits were then grouped into categories.Wireless communication, anytime/anywhereaccess and note-taking were rated in the topthree by 13, 12 and 11 of the 17 participants,respectively. All other benefits were suggestedby only one person, with the exception ofstyle/novelty claimed as a top benefit bythree students.

The concept-mapping tool was not used by anystudents for any purpose other than to try it. Inthe focus group, the consensus was that it wouldbe a very useful tool, but it appeared to beparticularly difficult to use. However, none of thestudents took up the offer of a demonstrationsession early in the project. Most weresurprised, when shown, how little effort it took tocreate basic concept maps, and suggested thatthe software needed redesign to make this clear.

4 Conclusions and commentOur study has found no evidence that use of abasic Tablet PC will in itself radically change orenhance student learning, although the peninterface can facilitate shared whiteboards andannotation of lecture slides among other things.The main observation from the data, backed upby feedback from the focus groups, is that themain advances are made with e-mail, instantmessaging, sharing of electronic resources, etc,none of which are exclusive to the Tabletplatform. Rather, it is the mobility, flexibility androbustness of a Tablet PC (over and above otherform factors) that makes it possible for thesepowerful uses of IT to become embedded inevery aspect of informal, collaborative learning.This in turn was considered by the teams to havea positive effect on their learning.

AcknowledgementsThe authors wish to acknowledge the financialsupport, donations in kind and involvement ofMicrosoft, Toshiba, Viglen, Mindjet Corporationand Ekahau.

ReferencesCorlett DJ, Sharples M, Chan TKS and Bull S(2004). A mobile learning organiser for universitystudents. In Proceedings of the 2nd IEEEInternational Workshop on Wireless and MobileTechnologies in Education, Taoyuan, Taiwan,23–25 March. Los Alamitos, CA: IEEE ComputerSociety, 35–42.

Ratto M, Shapiro RB, Truong TM and Griswold WG(2003). The ActiveClass project: experimentsin encouraging classroom participation. InComputer Support for Collaborative Learning.New York: Kluwer, 477–486.

Corlett Sharples 61

AbstractThis paper considers the employment of userscenarios within the context of the prototypicaldevelopment of a mobile learning architecture.Through a case study, it demonstrates howscenarios have influenced the developmentprocess by focusing on the role of teachers andlearners throughout different stages of theprototype development. It illustrates ways inwhich scenarios relate to system requirementsand evaluation, and how they offer acommunication strand between technical andnon-technical partners. During the project, keyscenarios were identified and gradually refined tobecome more concrete, so that eventually theycould be used to evaluate the final system.Scenarios used in this way also serve to resolvethe difficulty identified by Taylor (2003) of how tobring together a high-level evaluation design withimplementation issues.

Keywordsuser scenarios, mobile learning architecture, usertrials, evaluation planning

1 IntroductionThe potential domain for the use of mobiledevices to support learning is wide and varied.The major aim of the EU-funded MOBIlearnproject (IST–2001–37440) is to define andevaluate a pedagogically sound mobile learningenvironment. Taylor (2003) points out some ofthe inherent difficulties in achieving this aim,noting that there needs to be a discoursebetween the technical developers and thoseresponsible for high-level pedagogic design – adiscourse that is informative and helpful to bothparties. The decision to employ user scenarios wasmade to fulfil two purposes. First, to assist withthe process of ‘envisionment’ (Carroll 1995) –that is, to develop ideas about possible uses.

The role of user scenarios as the central piece of the development jigsaw puzzleDiane EvansJosie [email protected]@open.ac.ukIET UserlabOpen UniversityWalton HallMilton Keynes MK7 6AAUnited Kingdomwww.mobilearn.org

Second, to enable progression towards fieldstudies by providing user requirements in theuser-centred context. In the context of a largeand disparate set of partners, it also provided theopportunity to support dialogue betweentechnical developers and the non-technicalparticipants, who were responsible forrequirements gathering and evaluation.

We define scenarios, in this context, as ‘a storyfocused on a user or group of users, whichprovides information on the nature of the users,the goals they wish to achieve and the context inwhich the activities will take place’ (MOBIlearn2002). They are written in ordinary language, andare therefore understandable to variousstakeholders, including users. They may alsocontain different degrees of detail.

Three target domains were identified for studyand development. These were museums(including art galleries), health, and MBAstudents attending a university business courseon a part-time basis. It was within these threedomains that the scenarios first came to life. Thefirst stage was to capture the expertise ofpartners – who came from a variety of domainswith an extensive experience of teaching andlearning – in the creation of situations andsettings relevant to mobile learning. Pedagogicconsiderations were embedded in the resultsfrom this process, which led to the creation of alibrary of around 200 potential scenarios. It alsoprovided a substantial list of user requirementswhich were represented as Volere shells andstored within a requirements database (Haleyet al. 2004). These requirements presented therationale of what the users wanted to be able to do.

The second stage was a calibration exercise,gathering user data from learners andteachers within the context of the domain.Questionnaires and interviews were held withpeople visiting the Uffizi gallery in Florence;

Evans Taylor 63

interviews and focus groups involved students atthe University of Zurich; and future technologyworkshops (Vavoula, Sharples and Rudman 2002)were used with First Aiders at the Open Universityin the UK and students from the Universityof Birmingham. Outcomes from these andsubsequent stages provided further requirements.

The evolution of these scenarios, remaininggrounded in users and their requirements, hasbecome increasingly important to the evolutionof the system. The rest of the paper considersthe development of the agreed final scenariosand explains how they have contributed to otherproject deliverables – in particular, how thepreparation of a specific instantiation of thescenario assists the preparation of the evaluationplan for the user trials (where the term‘instantiation’ indicates that the scenario is to runin a specific place, at a given time, with realusers and real outcomes). Although threescenarios were developed, one representingeach domain, each followed the same process:this is illustrated here with examples from thehealth scenario. The work is ongoing.

2 Developing the health scenarioto focus on user trialsTo support this process, we re-defined scenariosas ‘a story focused on teachers and learners in aspecific context, identifying specific activitieswhich will be followed to achieve certainpredefined educational goals’. This definition iswell matched to educational activity in which acourse sets certain learning objectives and goalsfor the students to achieve. A tutor/trainer maythen define a set of activities, to take place over aperiod of time, which will together support thelearning objectives.

The calibration exercise used future technologyworkshops (Vavoula, Sharples and Rudman 2002)held at the Open University: the participantswere First Aiders who perform their role withinthe workplace environment in addition to theirnormal job. The workshop activity identified waysin which the use of mobile technology couldsupport their requirements for training, skillsdevelopment and practice. The resulting scenariowas written to focus on their training needs andis therefore grounded in their user requirements.It is based on constructivist pedagogy andinvolves learner interaction focused on definedactivities or learning episodes. It was planned toserve as the potential basis of the user trials,scheduled for autumn 2005. Figure 1 gives anexample of a stated requirement with relatedscenario activity.

Figure 1Example of a stated requirement with related scenario activity

RequirementFirst Aiders need to be able to store documentsfor later perusal either by themselves ortheir peers.

RationaleFirst Aiders will be able to learn from evaluationsand discussions of their prior assessments offirst aid incidents.

Scenario activity

1 Gill receives a picture of a first aid event.Thepicture is accompanied by a text message askingher to scrutinise the picture, and using hermobile device, to file an initial assessment withthe leader as to what needs to be attended to inthe environment.

2 The leader is pleased with Gill’s initialassessment, but notes that it differs fromPeter’s, another First Aider participating in theexercise. She suggests that Peter and Gillcommunicate with one another about theirinterpretations of the scene. They need to agreea plan of action.

In parallel with the development of the technicalprototypes, the next step was to refine thescenarios. The need was identified for a morespecific instantiation of the scenario in order toprovide detail for the user trials. The OpenUniversity’s first aid trainer was engaged in theprocess through discussions. The scenarioactivities were ratified and specific skills andtopics worthy of inclusion in the scenario wereidentified. Figure 2 opposite shows the structureof learning activities.


1 Log on to system

2 Select group of users

3 Enter text message

4 ‘Send’ message

MD_PDA–25

MD_PDA–26

MD_PDA–16

Login page displayed

Correct login displays main menu

List of users/groups of users is displayed

Allows selection of target group

Place to enter text is displayed

Option to send text

Figure 3 Matching scenario activities with system service calls

System Scenario activity Sub-activities service calls Expected system response

Notification of quiz:

Team leader sends message to all the team participants informing them of the imminent quiz and the procedure they should follow.

Figure 2Structure of learning activities

The refined health training scenario consists ofthree distinct episodes which support differentlearner activity and collaborative activity over aperiod of a week.

p the quiz –based on first aid procedures and processes

p the picture-/video-based activity –based on incident assessment

p the enactment –a role-play activity based on a simulated incident.

A further episode of individual activitylasting throughout the week enables the useof self-assessment content, access toexternal links and the opportunity to viewother related content.

The activities will cover procedures laid down foraction at an emergency: CPR (cardiopulmonaryresuscitation) and the recovery position.

The instantiation of the test scenarios provideddetails of users, content and the devices thatwere required. A decision was made to extendthe instantiation to satisfy the needs of theevaluation process by including details of thecontext of use, questions to be answered duringthe evaluation, and the activity or task whichwould be used to gather responses. This stageprovided an opportunity for partners who werefocusing on pedagogic validity andsocio-pedagogic usability to have an input intothe process.

3 Added benefits of scenario useThe development of the scenarios proved to be amulti-threaded activity and revealed furtherbenefits. We experienced difficulties in gettingthe non-technical partners to engage with thedatabase and the process of gathering formalrequirements. The key to overcoming thisdifficulty was the scenario refinement activity.There was also a need to bring togetherevaluators and technical partners to supportthe evaluation. The scenarios assisted withthis by providing a grounded focus innon-technical language.

The links between services and requirementsand between requirements andtesting/evaluation were facilitated by thedecomposition activities carried out as part of thefinal test instantiation. Developers were able touse the agreed scenarios to identify, plan anddevelop a range of services. Testing andevaluation of the developing prototypes could beplanned against the required functionality of thescenarios, creating a link back to the services.Figure 3 illustrates the matching of scenarioactivities with system services.

We anticipate that the process of producing thetest instantiation for the scenarios will have amajor benefit in driving the higher-levelpedagogic and socio-cognitive evaluation duringthe user trials. Issues have been identified andquestions defined before the system is completeand it is possible to engage in user activity.Finally, as we move towards the user trials, wehope that the recognition of their input into thescenario will entice First Aiders to volunteeras participants.

Evans Taylor 65

4 ConclusionWe have found that the use of scenarios in thisway has many features.

p It is multi-stranded, supporting mutuallyinforming dialogue.

p It performs a technical role in relation torequirements as specified in the database ofVolere shells.

p It supports both formative andsummative evaluation.

p It provides the missing link betweenhigher-order concepts, such as pedagogy andlearning objectives, and lower-levelimplementation aspects.

p It supports the verification of systemfunctionality against user requirements.

p It offers a mechanism to support activities linkedto user-system interaction.

p It engages users and their representatives in thedevelopment process and keeps their interestsat the centre of the project.

ReferencesCarroll JM (ed.) (1995). Scenario-based design.NewYork: John Wiley and Sons Inc.

Haley DT, Nuseibeh B, Sharp HC and Taylor J(2004). The conundrum of categorisingrequirements: managing requirements forlearning on the move. Paper presented at theRequirements Engineering Conference,6–10 September, Kyoto.

MOBIlearn (2002). Next-generation paradigmsand interfaces for technology supported learningin a mobile environment. Exploring the potential ofambient intelligence. Project No IST–2001–37187.Technical Annex. MOBIlearn Consortium.

Taylor J (2003). A task-centered approach toevaluating a mobile learning environment forpedagogic soundness. In J Attewell andC Savill-Smith (eds) Learning with mobiledevices. London: Learning and SkillsDevelopment Agency, 167–171.

Vavoula GN, Sharples M and Rudman PD (2002).Developing the future technology workshopmethod. In MM Bekker, P Markopoulos andM Kersten-Tsikalkina (eds) Proceedings of theInternational Workshop on Interaction Design andChildren (IDC2002). Shaker Verlag, 65–72.


AbstractIn most of today’s campus-like learning settings,students are frequently confronted withe-learning systems that do not only deliver anddisplay learning content, but are equipped with anumber of integrated tools for file sharing,communication, personalisation (personalannotations, text marking, etc) and cooperation.However, recent observations show that thesefunctions are rarely used. One reason for notusing the technological opportunities for informalcommunication and interaction is the poor designof such systems with respect to team awarenessin learning teams. The poor support forsynchronous and asynchronous communicationand collaboration in today’s e-learning systemshas motivated the development and prototypicalimplementation of a context-aware teaminteraction support system. This offersup-to-the-moment information about the location,state and activities of learning teams, thushelping to build up awareness of thesurroundings in which team learning processescan take place. In this paper, we present anextensible awareness system, which is capableof providing context information in apresentation-independent format. A team spaceis provided for mobile learning teams,integrating state and task awarenessfunctionality with a means of synchronousand asynchronous communication.

Keywordsmobile learning, team awareness, contextawareness, learning context

1 IntroductionUntil the last few years, learning was generallyconceived as a one-way knowledge transfer froma teacher to a learner, either directly or via media(eg books). This view has now largely beensuperseded by one where learning takes placeeverywhere and implicitly, thus becoming moreand more informal: ‘Informal learning is implicitlearning, which means it is derived from directinteraction … and a range of cues given by peersand instructors that go well beyond what isexplicitly being taught.’ (Ewell 1997).

Furthermore, it has been recognised that learningis much more effective if the learned informationcan be put into context with the learner’s existingknowledge. Hull (1993) makes the point that:

According to contextual learning theory, learningoccurs only when learners process newinformation of knowledge in such a way that itmakes sense to them in their frame of reference(their own inner world of memory, experienceand response). This approach to learning andteaching assumes that the mind naturally seeksmeaning in context – that is, in the environmentwhere the person is located – and that it does sothrough searching for relationships that makesense and appear useful.

Both informal and contextual learning lead tolearning settings, where it is no longer the casethat a teacher transfers knowledge to learners,but instead learners form teams to gainknowledge by discourse and dialogue, building asort of team knowledge that goes beyond thesum of individual understanding.

In recent years, many means to support learningprocesses on computers have been developed,as shown in Figure 1 overleaf. There is a broadrange of systems that enable or supportpresentational learning or self-study. Thesesystems are designed to present content in away that the learner can easily understand.

Ferscha Holzmann Oppl 67

Team awareness in personalised learning environmentsAlois FerschaClemens HolzmannStefan [email protected]@[email protected]

Institut fuer Pervasive ComputingJohannes Kepler University LinzAltenberger Strasse 694040 LinzAustria

Figure 1 Computer-mediated learning

There are possibilities of presentation that couldnot be realised in ‘traditional’ presentationallearning or self-study. Furthermore,instructor-initiated learning is also supported bymeans of computer-mediated communicationlike chats, forums or e-mail. At this stage,systems are not able to provide the same qualityof service that could be reached in face-to-facesettings, although there are many widely usedmeans of supporting interaction.

The most sophisticated support is needed forteam learning. Building a team learningenvironment which can provide the features thata physical meeting room provides is still an openissue in research. In particular, the lack ofpossible methods to keep the team membersaware of each other’s state has motivated thedevelopment of the system presented here.

1.1 Support for mobile learning teamsFerscha (2000) has identified the followingorganisational systems as crucial for thesuccessful support of effective mobile teams:

p information and knowledge management system(team memory)

p awareness system (team awareness)

p interaction systems (meeting support)

p mobility systems (mobile teams)

p organisational innovation systems (teamworkplace innovation).

While all of these systems are important foreffective team collaboration, we will now focuson team awareness. As it is hard to draw bordersbetween the supporting systems listed above,we will also partially cover aspects of interactionand mobility systems, as they can be enhancedutilising awareness information.

1.2 Team awareness – notionand deficitsDourish and Bellotti (1992) proposed one of thefirst and most common definitions of awarenessin CSCW (computer-supported collaborativework) settings: ‘Awareness is an understandingof the activities of others, which provides acontext for your own activity.’ This generaldefinition applies to ‘group’ or ‘team’ awarenesstools, the aim of which is to convey someinformation about the state and activities ofpeople within a team.

Supporting the learning process in distributedteams by awareness information has been atopic for research for over a decade now. Gutwin,Stark and Greenberg (1995) identify four generaltypes of awareness that a learner should have tosupport him or her in the team learning process:social awareness, task awareness, conceptawareness and workspace awareness. Whilesome of these types are possibly more importantthan others in most settings, team awareness isthe more general term, being thus a superset ofthese four types.

Jang, Steinfield and Pfaff (2000) identified fourspecific types of awareness deficiencies intoday’s virtual (distributed) teams. Generallyspeaking, there is a lack of awareness aboutothers’ activities (what are they doing?); aboutothers’ availability (when and how to reachthem?); about process (where are we in theproject?); and about the perspective (what arethe others thinking and why?).

Ferscha (2000) defined a team awarenesssystem as being responsible for facilitatingawareness of team activity by communicatingwork context, agenda and workspace informationjust-in-time to the user interfaces of relevantteam members and granting anytime access toteam memory.

2 Team awareness to enhancelearning in contextTo enable people to learn not just side by side,but together, to form teams and interact witheach other, it is crucial to use not only theinformation about the individual learner’s context,but also that on the context of the whole learningteam, supporting the members’ interaction inthose teams.


2.1 What is context?Many attempts have been made to define thenotion of context. Most definitions of context aremade by enumeration of examples or bychoosing synonyms for context. In the work thatfirst introduces the term ‘context-aware’, Schilitand Theimer (1994) refer to context as ‘location,identities of nearby people and objects, andchanges to those objects’. In a similar definition,Brown, Bovey and Chen (1997) define context as‘location, identities of the people around theuser, the time of day, season, temperature, etc’.Ryan, Pascoe and Morse (1997) define context asthe ‘user’s location, environment, identityand time’.

A widely used definition is the one given inthe PhD thesis of Dey (2000), who says thatcontext is:

any information that can be used to characterizethe situation of an entity. An entity is a person,place, or object that is considered relevant to theinteraction between a user and an application,including the user and applications themselves.

Following this definition, Dey (2000) defines asystem as being context-aware ‘if it uses contextto provide relevant information and/or services tothe user, where relevancy depends on the user’stask’. Other definitions of context-aware systemsinclude ‘systems that adapt themselves tocontext’ (Schilit and Theimer 1994) or ‘systemshaving the ability to detect and sense, interpretand respond to aspects of a user’s localenvironment and the computing devicesthemselves’ (Ryan, Pascoe and Morse 1997).

2.2 Context dimensionsIn order to design context-aware applicationssystematically, it is useful to identify categoriesof context (so-called context dimensions) thathelp designers to uncover the most likely piecesof context to be used. Although many andvarious classifications have been proposed, wewill present the categories that Dey (2000)identified, as these are the most common ones.

The number of dimensions that could beidentified is nearly uncountable, but in practice,some of them are more important than others.These are location, identity, activity and time, whichcould be seen as the primary context types forcharacterising the situation of an entity, as they canserve as indices into other sources of contextualinformation (secondary context types). If we knowa person’s identity, we can easily derive relatedinformation (such as birth date, list of friends ore-mail addresses) from several other data sources.

Knowing the location of an entity, we candetermine the nearby objects and people andwhat activity is occurring near the entity.

There are some situations in which multiplepieces of primary context are required to indexinto an information space to acquire secondarycontext information. For example, the forecastedweather is a context dimension for outdoor activityand can be obtained by querying a forecastdatabase with the desired location and time.

2.3 Mobile learning contextTailored to mobile learning settings, the notion ofcontext can be defined slightly differently. Thelearning context as defined by Wang (2004) readsas follows: ‘any information that can be used tocharacterize the situation of learning entities thatare considered relevant to the interactionsbetween a learner and an application’.

Wang (2004) identifies six dimensions that arerelevant in computer-aided mobile learning:

p identity

p spatio-temporal

p facility

p activity

p learner

p community.

Location has proved to be one of the mostimportant and effective contexts in manyapplications. Besides identifying the geographicalposition of the user, it makes sense to introducesome sort of meta-information that enables thesystem to distinguish between locations used fordifferent purposes. In learning settings, forexample, these ‘meta-locations’ could beclassroom, home or outdoors, thus enabling thesystem to adapt to the current learning situation.

In our approach, we are adding team awarenessas a seventh dimension, as the six predefinedcontext dimensions are single-user centred anddo not cover issues that can only arise whenlooking at the learning team as a whole. Thecontext of the team is distilled out of theindividual contexts of its members. Therefore,the team awareness dimension might seem tooverlap strongly with several of the dimensionsdefined by Wang, especially the communitydimension and the activity dimension. As thecontext of a team is much more than the sum ofthe contexts of the team’s members, thedifference is that team awareness focuses on theteam itself and not on the single users, enablingus to gather high-level context information aboutthe team by combining the members’ contexts.


For example, it would now be possible to notifythe user of a meeting of other team membersoccurring somewhere else.

2.4 Architectural modelTaking into account all the issues stated above,we could now construct an architectural model(see Figure 2), which shows the role of contextin a team learning setting. Context is sensed,transformed and represented in the contextmodule. This module holds the individuallearners’ contexts, but it also generates andholds the team learning context, taking intoaccount team awareness information.

Figure 2 Architectural model showing the role ofcontext in a team learning setting

The context module serves as an interface forthe learners, not only adapting content to thecurrent context of the learner, but also providingadditional information about the context of thelearner and his/her colleagues. A comparablearchitecture has been developed byLonsdale et al. (2003) as part of the MOBIlearnproject (2005). The main difference in ourapproach is that we explicitly use teamawareness as a context dimension, focusingon the team as a central part of (mobile)learning settings.

3 Team awareness andinteraction supportWith the system presented here, we havedeveloped and implemented an extensiblearchitecture that uses several dimensions ofcontext gathered from individual users tocompose team awareness information. Thisinformation is presented to the team membersand used to support interaction within teams.

For now, only a subset of team awareness issupported. We focus on parts of workspace andsocial awareness that can be used to enhancethe interaction among team members. As thewhole architecture is built upon a highlyextensible and dynamic framework (Beer et al.2003), however, it is fairly simple to add furthercontext sources, thus extending the aspects ofteam awareness that are covered.

3.1 Gathering individualcontext informationThe context dimensions that are used in oursystem at the current stage of development areidentity, time, location and activity. Informationon time could easily be obtained by querying theRTC (Real Time Clock) of the learner’s computer.Much more sophisticated is the detection of thelearner’s current location. For this purpose, wehave developed a technology-independentpositioning system, currently using WLAN(wireless local area network), RFID (radiofrequency identification) and Bluetooth sensors.As our user interface is built upon an instantmessenger that utilises an existing messagingnetwork (namely ICQ), we are using thefeatures of this network to obtain identity andactivity information.

3.2 Triggering context-aware servicesOnce information has been gathered about thecontext of the learner and his/her co-learners, wecan now trigger context-aware actions to acquireteam awareness and support team interaction. Acontext-aware action is one of the followingtypes (Dey 2000):

p context-aware presentation [passive contextawareness (Chen and Kotz 2000)]

p automatic execution of services (activecontext awareness)

p tagging of context for later retrieval.

To realise context-aware actions, we use theSiLiCon Context-Framework, an event-basedsystem, which is described by Beer et al. (2003).The open architecture of this framework allowsus easily to extend not only the context sourcesused, but also the behaviour of our applicationto improve the support for team awarenessand interaction.


3.3 Presentation of team awarenessinformation and supportof interactionUsing the available pieces of context, weidentified several functionalities that are usefulfor the support of synchronous andasynchronous interaction in teams and thepresentation of team awareness information. Asalready stated, our user interface is based on anexisting instant messenger, using the features italready provides.

Synchronous interaction for individualsand groupsWhile current messaging systems often supportonly peer-to-peer interaction, it is also necessaryto support synchronous group interaction; forexample, in cases where team members initiallydo not know each other (as may be the case insettings where the participants of a lecture forma learning team). For that reason, we havedeveloped team contacts that are displayed inthe contact list of the messenger and can beused like standard contacts.

Asynchronous interaction for individualsand groupsWhile in general, e-mail or forums cover thisaspect, we extend the concept of asynchronousinteraction by messages (‘post-its’) being boundto physical locations, thus enabling users to leavemessages for others in the context of the currentlocation. These virtual ‘post-its’ can also have anexpiry date, and visibility can be restricted tocertain teams. It is also possible to comment on‘post-its’. For example, users could leave a noteat their office when they leave, telling others (butmaybe only the members of their department)where they can be found. Another possibleapplication is the virtual blackboard, wherepeople could post requests, offers or simplyrelevant information.

Team formation supportTeams differ in their purpose, so it is necessaryto provide different types of team: closed teams(with an owner); open teams, which can bemanually joined and left by the users; anddynamic teams, which are formed automaticallyby the system using the current context ofpossible members and meta-information aboutthe dynamic team’s location (whether it makessense or not to build a team at thecurrent location).

Team meeting supportUsing the available context information, we aretrying to recognise team meetings to provideseveral supporting services. These includeregistering members joining and leaving themeeting, notifying absent members andsupporting protocol storage and distributionamong team members.

To support awareness about the activity andlocation of team members (as a subset of teamawareness), we permit automatic availabilityadoption (depending on user-defined rules usinglocation and time as triggers). We also display thecurrent location of all team members in thecontact list and (optionally) in a floor-plan-basedgraphical viewer. Several more tools to supportindividual task and awareness management areavailable, but these are not presented here.Figure 3 shows the running system with enabled2-D viewer.

Figure 3Running system with enabled 2-D viewer

The current version of our system (see Figure 3)is, in the autumn of 2005, in its internalevaluation stage. After this evaluation iscompleted, it will be made available to studentsat the University of Linz, enabling them tobecome aware of their friends’ and teammembers’ availability and position. This willenable them to extend their interactionpossibilities easily without needing to get used toa new interface (as our system is built on anexisting instant messenger). This field test willalso be a chance for us to evaluate the scalabilityand user-acceptance of our approach.


4 SummaryWe have shown the need for using contextinformation to support learning in teams. It isnecessary to make use of both the individuallearner’s context and team awareness informationto make possible content adaptation andcontext-aware presentation of information as wellas context-aware support for interaction in groups.

We believe that it is crucial to build awarenesssystems upon flexible and extensibleframeworks, thus providing a platform that caneasily be adapted to the available contextsources and to the required form of presentation.

With the implemented system, we show ways ofhaving context-aware interaction support inteams and presenting team awarenessinformation to team members, providing astarting point for a comprehensive teamawareness support system.

ReferencesBeer W, Christian V, Ferscha A and Mehrmann L(2003). Modelling context-aware behavior byinterpreted ECA rules. In Proceedings of theInternational Conference on Parallel andDistributed Computing, Klagenfurt, Austria,26–29 August. Austria: Springer Verlag,1064–1073.

Brown PJ, Bovey JD and Chen X (1997).Context-aware applications: from the laboratoryto the marketplace. IEEE PersonalCommunications, 4(5), 58–64.

Chen G and Kotz D (2000). A survey ofcontext-aware mobile computing research.Hanover, NH: Computer Science Department,Dartmouth College.

Dey A (2000). Providing architectural support forbuilding context-aware applications. PhD thesis:Georgia Institute of Technology.

Dourish P and Bellotti V (1992). Awareness andcoordination in shared workspaces. InProceedings of the 1992 ACM Conference onComputer-supported Cooperative Work, Toronto.NewYork: ACM Press, 107–114.

Ewell PT (1997). Organizing for learning: a pointof entry. Draft paper prepared for discussion atthe 1997 AAHE Summer Academy at Snowbird.National Center for Higher EducationManagement Systems (NCHEMS).At www.intime.uni.edu/model/learning/learn_summary.html

Ferscha A (2000). Workspace awareness inmobile virtual teams. In Proceedings of the IEEE9th International Workshop on EnablingTechnologies: Infrastructure for CollaborativeEnterprises, Gaithersburg, Maryland, 14–16March. IEEE Computer Society Press, 272–277.

Gutwin C, Stark G and Greenberg S (1995).Support for workspace awareness in educationalgroupware. In Proceedings of the Conference onComputer Support for Collaborative Learning,Bloomington, Indiana, 17–20 October. New York:ACM Press, 147–156.

Hull D (1993). Opening minds, opening doors:the rebirth of American education. Waco,TX: Center for Occupational Researchand Development.

Jang CY, Steinfield C and Pfaff B (2000).Supporting awareness among virtual teams ina web-based collaborative system: theTeamSCOPE System. SIGGROUP Bulletin,21(3), 28–34.

Lonsdale P, Barber C, Sharples M and Arvantis T(2003). A context awareness architecture forfacilitating mobile learning. In Proceedings ofMLEARN 2003, London, 19–23 May.

MOBIlearn (2005). Website.At www.mobilearn.org

Ryan N, Pascoe J and Morse D (1997).Enhanced reality fieldwork: the context-awarearchaeological assistant. In V Gaffney, M vanLeusen and S Exxon (eds) Computer Applicationsin Archaeology.

Schilit B, Adams N and Want R (1994).Context-aware computing applications.In Proceedings of 1st International Workshopon Mobile Computing Systems andApplications, 85–90.

Schilit B and Theimer M (1994). Disseminatingactive map information to mobile hosts.IEEE Network, 8(5), 22–32.

Wang Y (2004). Context awareness andadaptation in mobile learning. In Proceedings ofthe 2nd IEEE International Workshop on Wirelessand Mobile Technologies in Education, Jhongli,Taiwan, 23–25 March. IEEE CS Press, 154–158.


AbstractWith the exponential development in technology,there is always the danger that the equilibriumbetween technology and learning is disturbed.This paper aims to look at m-learning from aneducational perspective. The context of thisperspective is higher and lifelong education in adeveloping country.

Keywordsm-learning, e-learning, lifelong learning, adultlearning, education, higher education,media attributes

1 ConceptualisationHow would you categorise the learning takingplace in the following examples in Figure 1?Are we referring to ‘flexible’, ‘distance’, ‘e-’ or‘m-’ learning? Shouldn’t we refrain from all this‘hype’ and focus on the essence – the learning?

Figure 1 Examples of learning situations

Example 1Listening to a learning programme on an audiocassette while driving your car

Example 2Reading study material or a textbook in print,compared to reading a PDF text on yourcomputer, while travelling by train

Example 3Reading e-mail from your lecturer at home, youroffice or at a ‘hotspot’ at the cafeteria

By trying to categorise the examples into thedifferent subsets (see Figure 2), you will agreethat ‘learners … continually on the move…’(Brown 2004) cannot sufficiently serve as aunique descriptor of mobile learning (m-learning).

M-learning: an educational perspectiveJohan B [email protected] Learning and Education InnovationUniversity of Pretoria0002 PretoriaSouth Africa

Figure 2 The subsets of flexible learning

Brown (2004) pictures the relationship betweenm-learning and the other different subsets quiteclearly, as shown in Figure 2.

Shepherd’s statement (2001) that: ‘m-learning isnot just electronic, it’s mobile’, suggests thatm-learning is a subset of e-learning (see alsoBrown 2004). This, unfortunately, does notsufficiently clarify the term ‘mobile’. Quin (2001)tries to overcome this problem by providingsome examples: ‘M-learning is e-learning throughmobile computational devices: Palms, WindowsCE machines, even your digital cell phone’.

For the purpose of this discussion, I will definem-learning as the use, both synchronously andasynchronously, of mobile communicationtechnology (MCT) to achieve a learning taskor outcome.

Freysen 73

2 ContextualisationIn order to lay a foundation for further discussions,m-learning is placed in the context of formalhigher education (HE) in a developing country.

Mobile communication technology (MCT) will beused differently and for different purposes, basedon the different needs of adult learners withregard to formal, informal and non-formallifelong learning.

The consideration of m-learning as a viable optionfor the formal HE sector should be based uponfactors such as convenience, expediency,immediacy (Seppäla and Alamäki 2003), culturalpreferences, access, media literacy andaffordability from the learner’s perspective.

The per capita income, poor infrastructure,preference for oral communication and ‘digitaldivide’1 in Africa makes ‘leapfrogging’2 withregard to m-learning an attractive proposition.

3 An educational perspectiveWhen discussing m-learning per se, it isnecessary to distinguish clearly between the useof mobile technology for administrative purposesand its use for the facilitation of learning.

I am of the opinion that most of the instanceswhere MCT is being used in higher education, itis for administrative support of learning, ratherthan for learning itself. The case study at theUniversity of Pretoria (Brown 2004), wheremobile phones are used to support learning atthe Faculty of Education, as well as the project atthe Faculty of Health Sciences where personaldigital assistants (PDAs) are being used toassess medical learners during their ObjectiveStructured Clinical Examination (OSCE) are casesin point.

Apart from administrative uses, learners couldalso use MCT to contact an expert, a fellowmember from a community of practice (COP), oreven search a database in order to getjust-in-time information. This type of learningusually takes place on a less structured level.

The question is: to what extent can m-learningcontribute within the context of structured orformal higher education?

This issue could have been approached from aconstructivist, cognitive or even from adevelopmental perspective on learning –rather than from a technological one.

Instead, the adapted model of Bloom’s taxonomyby Anderson and Krathwohl (2001) is taken as apoint of departure in order to look at thepossibilities of m-learning in higher education bymerging the knowledge dimension with thecognitive process.

Although this revised taxonomy (see Figure 3below) is focusing on the development ofobjectives, it could also be applied to theimplementation of m-learning in highereducation. Understandably, the scope of thispaper will not allow a detailed discussion.

From the lecturer’s viewpoint, considering thedimension of the knowledge will ensure that thelearning task is applicable. Considering thedifferent cognitive processes will ensure a varietyin the complexity of learning level. The challengeis to identify attributes (like screen size orstorage capability) and select the MCT that isappropriate for the learning task and level. This is,however, still very much a content-basededucational paradigm!

From the learner’s viewpoint, m-learning shouldbe transparent,3 based on collaborative learningthat is just-in-time (JIT). The nature of thislearning is based on new insights arrived atthrough the ‘chunking’ of JIT information.


Figure 3 The revised taxonomy tableAdapted from the Encyclopaedia of Educational Technology (2004)

The knowledgedimension

Factual knowledge

Conceptual knowledge

Procedural knowledge

The cognitive process dimension

Remember Understand Apply Analyze Evaluate Create

For example, should the learner need tounderstand procedural knowledge (see Figure 3opposite), he or she would hardly use ShortMessaging System (SMS), but would rathermake a phone call to achieve the learning task.

To conclude, even in a formal learning situation,MCT could be used to achieve different learningtasks at different levels. By shifting the focusfrom the technology and its promising potentialto sound fundamental educational applications,m-learning will be accepted by the broadHE community.

ReferencesAnderson LW and Krathwohl D (eds) (2001).A taxonomy for learning, teaching, andassessing: a revision of Bloom’s Taxonomy ofEducational Objectives. New York: Longman.At: http://coe.sdsu.edu/eet/Articles/bloomrev/index.htm, accessed May 2004.

Brown TH (2004). The role of m-learning in thefuture of e-learning in Africa. In D Murphy, R Carr,JTaylor and W Tat-meng (eds) Distance educationand technology: issues and practice. Hong Kong:Open University of Hong Kong Press, 197–216.

Encyclopaedia of Educational Technology (2004).Bloom’s revised taxonomy. At: http://coe.sdsu.edu/eet/articles/bloomrev/index.htm, accessedMay 2004.

Quin C (2001). mLearning: mobile,wireless, in-your-pocket learning.At: http://www.linezine.com/2.1/features/cqmmwiyp.htm, accessed May 2004.

Seppäla P and Alamäki H (2003). Mobile learningin teacher training. Journal of Computer AssistedLearning, 19(3), 330–335.

Shepherd C (2001). M is for maybe.At www.fastrak-consulting.co.uk/tactix/features/mlearning.htm, accessed 29 March 2004.

Endnotes1 Digital ‘divide’ refers on the one hand to those

who have the technology and are able to use it;and those on the other hand, who either do nothave the technology and/or are not able to use it.

2 Leapfrogging’ implies that certain phases intechnological development are skipped to catchup with the rest of the world.

3 The learner should hardly notice that s/heis learning.

Freysen 75

AbstractWe describe a mobile vision system that iscapable of automated object identification usingimages captured from a personal digital assistant(PDA) or a camera phone. We present a solutionfor a technology that will enable outdoorsvision-based object recognition, which willextend state-of-the-art location andcontext aware services towards the achievementof object-based awareness in urbanenvironments. In the proposed applicationscenario, tourist pedestrians are equipped with aGlobal Positioning System (GPS), a wireless localarea network (WLAN) and a camera attached to aPDA or a camera phone. They are asked to lookand see whether their field of vision containstourist sights that would prompt them to seekmore detailed information. A mobile user who isintending to learn within the urban environmentmight want to explore multimedia-type dataabout the history, architecture or other relatedcultural context of historic or artistic relevance.Accordingly, ambient learning is achieved bypointing the device towards the sight, capturingan image and consequently getting on-siteinformation about the object within the focus ofattention – that is, the user’s current fieldof vision.

Keywordsmobile vision, object recognition, location-basedservices, learning in urban environments

1 Motivation for the projectMobile learning systems operating in urbanenvironments must take advantage of contextsarising from the spatial and situated informationat the pedestrian user’s current location.Location-based services today are, in principle,able to provide access to rich sources ofinformation and knowledge to the nomadic user.

Mobile vision for ambient learning in urban environmentsGerald FritzChristin SeifertPatrick LuleyLucas PalettaAlexander Almer

However, the kind of location awareness thatthey do provide is not intuitive, requiringreference to maps and addresses – in otherwords, the information is not directly mediatedvia the object of interest.

In contrast, the work presented here takes adecisive step towards getting in line with theuser’s current intention to relate information tohis or her current sensorial experience – theobject in his/her line of sight (Figure 1a). In thisway, the system can respond to the user’s focusof attention; for example, for the purpose oftourist information systems. A camera attachedto the mobile system (PDA or camera phone)pointing towards the object of interest (eg abuilding or a statue) will capture images on demand,automatically finding objects in the tourist user’sview. The images are then transmitted to a serverthat automatically extracts the objectinformation, associates it to m-learning contentand sends the resulting data back to the mobileuser (Figure 1b). ‘Mobile vision’ here refers tomobile visual data that is processed in anautomated way to provide additional informationto the nomadic client in real time.

State-of-the-art pattern recognition in mobilecomputer vision has advanced from indoorsprocessing (Aoki, Schiele and Pentland 1999) andtime-consuming outdoors recognition (Coors,Huch and Kretschmer 2000) to nomadic signidentification (Yang et al. 2001) and attentiveprocessing for robust recognition performance(Fritz et al. 2004). Layered mobile location-basedservices (Hightower, Brumitt and Borriello 2002)support, in addition, more accurate estimatesabout the immediate environment of the user;for example, to provide mobile tourist informationservices (Almer and Luley 2003). The goal is toannotate detected objects with characteristicdata and to provide an interface to access evenmore detailed multimedia information about theselected object.

Fritz Siefert Luley Paletta Almer 77

[email protected] of Digital Image ProcessingJoanneum ResearchWastiangasse 6A–8010 GrazAustria

Figure 1 The nomadic tourist using a vision-enhanced ambient learning system

c d e f

1a The nomadic tourist explores the city in interaction with his/her vision-enhanced ambient learning system.

1b The image capture relating to the tourist sight is transmitted via the internet to the server that performs image processing and automated object recognition.

1c–f Screenshots from the mobile learning system: (c) the user initiates sight identification via an image query;(d) the image is transmitted to the server; (e) the resulting information is displayed at the client site;(f) contextual information is provided together with a menu that links to additional information sources(movies, pictures, audio, etc).

Entries within a corresponding database ofmultimedia information of various kinds (images,video, sound, texts) are interlinked to enablecross-referencing and access to differentinformation spaces (Figure 1c–f).

2 Mobile learning using visualobject detectionThe user equipment of the ambient learningsystem (Figure 1b) consists of a camera-basedPDA or camera phone and a GPS receiverconnected via Bluetooth. The client device shouldprovide an integrated or connected camera witha resolution of at least 640x480 pixels assupported by the industrial standard today.The mobile device is linked via wirelessconnection to a server. We assume that, atthe very least, a mobile phone network canbe used to establish an internet connection;


a

b

alternatively, different data transfer technologieslike General Packet Radio Service (GPRS),Universal Mobile Telecommunications System(UMTS) or wireless local area network (WLAN)could be used as well, depending on the choiceof local network providers. Object identification isperformed at the server site and hidden to theuser who just receives the result (Figure 1e) inabout 2–3 seconds. The server receives a rawGPS-based position estimate and collects aselection of relevant sights which potentiallymight appear in the tourist’s field of vision. Theuser initiates an image capture to start the visualobject recognition (Figure 2). The system firstextracts a grey-level pixel pattern (Figure 2a)and identifies image regions that are highlydiscriminating with respect to objectidentification (Figure 2b). Larger regions oflocal object votes (Figure 2c) are integrated in asecond processing stage, resulting in a singleobject vote (eg the blue rectangle in Figure 2a).

The demonstration system contains 10 localobjects of interest from the city centre of Graz inAustria (buildings, statues, etc; see sampleresults in Figure 2e–h), but is currently beingextended to display up to 50 objects of interestto tourists. The PDA will then display associatedm-learning information from the contentdatabase corresponding to the object-searchresult (Figure 1b and f). The multimediainformation will describe the cultural contextrelating to the object in the field of vision,enabling the user to learn within the urbanenvironment. The presented experimental resultsseem to be promising and are being extended toevaluate larger object sets.

Fritz Siefert Luley Paletta Almer 79

Figure 2 Operation of the mobile vision system

e f g h

2a The mobile vision system operates on grey-level pixel patterns; for example, with reference to the Town Hall in Graz.

2b It first identifies distinctive local patterns (Fritz et al. 2004) in the complete image (d depicts the corresponding local patterns from the region in a, circumscribed by the white rectangle). Blue pixels were classified as ‘voting for’ the ‘town hall object’.

2c describes a tiled image partition, being colour coded from a more abstract voting process.

2e–h Colour coding of correct object classifications for several sights and a statue (h) found among the sights of interest to tourists in Graz.

a b c d

ReferencesAlmer A and Luley P (2003). Location-basedtourism information on mobile systems. InProceedings of the European NavigationConference, Graz, Austria, 22–25 April.

Aoki H, Schiele B and Pentland A (1999).Real-time personal positioning system forwearable computers. In Proceedings of the IEEEInternational Symposium on WearableComputing, San Francisco, California.Los Alamitos, CA: IEEE Computer Society, 37.

Coors V, Huch T and Kretschmer U (2000).Matching buildings: pose estimation in an urbanenvironment. In Proceedings of the IEEE andACM International Symposium on AugmentedReality, Munich, 5–6 October, 89–92.

Fritz G, Seifert C, Paletta L and Bischof H (2004).Rapid object recognition from discriminativeregions of interest. In Proceedings of theNational Conference on Artificial Intelligence,San Jose, California.

Hightower J, Brumitt B and Borriello G (2002).The location stack: a layered model for location inubiquitous computing. In Proceedings of theIEEE Workshop on Mobile Computing Systemsand Applications, Callicoon, New York. LosAlamitos, CA: IEEE Computer Society, 22–28.

Yang J, Gao J, Zhang Y and Chen X (2001).An automatic sign recognition and translationsystem. In Proceedings of the Workshop onPerceptive User Interfaces, Lake Buena Vista,Florida, 15–16 November.


AbstractThis paper reports on the MBA case-studyscenario strand of the MOBIlearn project,focusing on the user trials and evaluationsplanned for the autumn and winter of 2004. Itspecifies the added value expected byintegrating the MOBIlearn system into atraditional MBA-course lecture. This paper tries toindicate what kind of results might be gainedfrom the user trials and the evaluation of theMOBIlearn system.

All results presented in the framework of theMOBIlearn project are funded by the SwissFederal Office for Education and Science (BBW).

Keywordsmobile cooperative learning, case-study education

1 IntroductionMOBIlearn is about to start the final phase ofevaluating the MOBIlearn system with realusers. The MOBIlearn system has beendeveloped during the past 2 years based on threemain scenarios: museums, health and MBAstudy. More details about the scenario-drivenapproach and its impact on system design arepresented in Evans and Taylor (2005). The presentpaper reports on the MBA scenario, which is ledby the University of Zurich, with regard to theevaluation planned for the autumn and winter of2004. To put the discussion into context,section 2 describes the profile of a typical ZurichMBA class which is significantly different to thatof a class of undergraduates. Section 3 reportson the expected added value of mobiletechnology and on evaluation plans based onthree selected relevant challenges derived from aformal MBA-course lecture. Section 4 addressesevaluation issues when doing a case study duringan MBA course.

Mobile learning in tomorrow’s education for MBA studentsDirk [email protected] of InformaticsUniversity of ZurichWinterthurerstrasse 1908057 ZürichSwitzerland

2 User profile of Zurich’sMBA studentsTypical Zurich MBA students are between 35 and45 years old and already have a couple of yearsof practical management experience. They areused to working under extreme time pressureand must coordinate their participation in theMBA course with their daily business. Often theydo not work in their office, but have meetingsoutside and thus need to be very mobile. Theyare experts in their specific area of business, andwithin this area, they have practical knowledgethat is superior to that of fellow students andeven the lecturer. They expect to get somehigh-level theoretical reflection in their area ofexcellence plus up-to-date knowledge aboutother areas that are relevant to their job or career.So an MBA class of about 40 students is anextremely heterogeneous and demanding group,with both experts and lay members in any topiccoming from very different sectors, companies anddepartments. The MOBIlearn scenario for such anMBA course contains two elements: a traditionallecture (see section 2); and a case-study exerciseto be solved in small groups (see section 3).

3 The added value of mobiletechnology in a traditionalMBA-course lectureThere are some requirements for integratingmobile technology into a traditional course lecture:each student needs to be equipped with a mobiledevice, such as a personal digital assistant (PDA),which has a set of mobile learning tools installedand allows wireless local area network (WLAN)access. The classroom needs WLAN coverageand an additional electronic blackboard orbeamer. Furthermore, the lecturer needs centraltools and functionalities in order to moderate andcontrol the class activities. An assistant shouldsupport him or her in moderating and dealingwith the digital channel.

Frohberg 81

There are several challenges to be faced whenteaching an MBA course. This section addressesthree specific challenges that lead into severalresearch questions concerning what value can beadded when using mobile technology.

First, the teacher needs to present the content inaccordance with the previous knowledge of theclass. Students with no or little previousknowledge should not be swamped; andstudents with expert knowledge should not getbored by the lesson. The teacher usually does notknow the MBA students’ previous knowledge indetail, which complicates the situation and maylead him or her simply to ignore theheterogeneity of the class. One way to ensuresome minimum previous knowledge would be toask the students to prepare for the lectureindividually in advance. However, a flexibleteacher could try to get a constant feedbackabout the class performance and adopt his or herlesson to the current needs of the students.

Addressing this challenge would be easy with avery simple tool that provides just three buttons:red, yellow and green, plus a comments box oneach client. Students could indicate to theteacher whether they currently feel comfortablewith the lecture or not. As long as ‘green’ ispressed, there are no problems, but ‘yellow’ oreven ‘red’ means that either the student cannotfollow any more or conversely, feels unchallenged.The text box can be used to specify the problem.It is up to the teacher either to show the overviewfrom the server publicly or just keep it private ona separate screen for him/herself. If there appearto be too many red or yellow indications, theteacher gets a clear message to deal with it in aproactive way. Using class awareness tools ofthis kind, the lecturer would be able to dealsensitively with the needs of the class.

A tool of this sort will not be part of theintegrated MOBIlearn system, but is currentlybeing developed in Zurich to be used during thefinal user trials of MOBIlearn. The questions toconsider are first, whether students and teachersaccept and use the tool; and second, if it makes apositive difference to the teacher’s performance.Both questions will be addressed inquestionnaires and in interviews.

Second, the expert knowledge of the audience isoften not exploited, because possibilities to shareit at the right moment are very limited. Duringobservations made in MBA classes (seesection 3), students often seemed to be willingto share their experience by telling anecdoteslinked to the current content of the lesson. Dueto a tough time schedule, however, the teacherscannot allow all the students to tell their story,even though it would be very valuable from amotivational and didactical point of view.

Providing the students with a chat facility orforum so they can share their valuable opinions,questions and anecdotes with the othersaddresses this challenge. An advanced toolwould allow any entry to be linked to the relevantslide of the lecture and thus associate it with therelevant context. The typing of a whole anecdoteon a PDA, however, entails a lot of distraction forthe writer. A notebook might be a better deviceto provide this service, but some level ofdistraction will remain. We assume that mostwriters will be experts in the area presented bythe lecturer, so instead of being bored, they maycontribute personal and practical experience tothe course community.

There are a number of earlier projects that havealready developed and tested useful tools forsuch activities, but with undergraduate students;for example, ActiveClass, WILD@Mannheim orConcertStudeo (Lehner, Nösekabel and Bremen2003). The MOBIlearn system is going tointegrate similar tools to support the MBAcase-study scenario. The evaluation should givesome indication whether the heterogeneity ofstudents and the presence of expert knowledgein an MBA class will lead to different and morepositive effects than in a class of undergraduates.The quality (and quantity) of comments made willprobably answer this question to some extent.

Third, there is little or no electronic support in astandard classroom for carrying out someeffective collaborative group sessions (CSCL –computer-supported cooperative learning); forexample, a digital brainstorming session withongoing prioritisation, categorisation anddecision-making support (voting). Such a moderateddigital session has been proved to be very effectivecompared to traditional methods with pencil andpaper (Schwabe 1995). It is a valuable way ofactivating the students, making them think and talkabout their thoughts, discuss complex questionsand share their knowledge. One possible referencetool for Fixnet applications is GroupSystems (2005).Unfortunately, such tools are not heavily used ineveryday teaching practice, because the effort tobe spent in preparation outweighs the gain. Onereason would certainly be the inconvenience ofbooking and preparing a computerised room andmoving the whole class into it. Flexible andspontaneous use of such methods is simply notpossible under those circumstances. With mobiletechnology, however, the classroom is potentiallycomputerised for free, because students bring theirown devices. The evaluation of the MOBIlearnsystem, which is supposed to support digital groupwork, should show if the teacher and (MBA)students are more likely to use collaborative groupsession software when this is available quickly andon demand, making use of mobile technology.


4 The added value ofmobile technology in acase-study exerciseThe second part of the MBA course scenario inMOBIlearn is the performance of a case-studyexercise. Non-technical pre-tests with an MBAclass and a class of undergraduates have beendone in the winter of 2003/04 and in spring 2004.The MBA course, in particular, with 43 participants,has been evaluated in respect of the MOBIlearnuser tests by carrying out observations and aseries of interviews which have included nearlyall learning groups.

The technical setting would be similar to thatdescribed above (section 3), but extended.Additionally, there should be a number ofawareness tools to submit the current location ofgroup members or data on their social presence(current availability, willingness to communicate,current activity, etc) There should be WLANcoverage at the university campus, in mostcompanies, and in some cases, at home as well.A number of effects are anticipated and the planis to evaluate these when integrating mobiletechnology in order to complete an explorativecase-study exercise in group work (collocated,distributed, synchronous and asynchronous).Four of those anticipated effects will bediscussed here.

4.1 Mobile technology improvescoordination and leads to a shift fromasynchronous to more synchronouscooperation which results in abetter learning performanceDuring the observed course, Zurich’s MBAstudents started their group work with a kick-offmeeting where they were agreeing on a roughprocess by which to complete the case-studyexercise. About half of the groups of MBAstudents needed, for business reasons, to workon the case study in a distributed andasynchronous way. They usually split theexercise work into evenly sized sub-tasks, onwhich each group member worked individually.This approach seems to be most efficient forthem, but didactically it is sub-optimal.By splitting up the work, they miss out ondiscussing their ideas, do not get a sharedunderstanding of their work and can hardlycontribute to each other’s work. By integratingmobile technology, the available informationabout social presence and communication toolsallows group members to coordinate their workin a much more flexible and spontaneous way.

They can quickly make even small decisionsabout the work process as needed, availthemselves of the opportunity of meeting bychance, and have the possibility of observing andcontributing to each other’s work. Thus thecooperation may become closer and moresynchronous and overcome theaforementioned barriers.

4.2 Mobile technology supportsthe build-up of a communitynetwork that may last longer thanthe MBA courseNot only working and learning together, but alsosharing private issues might be important to buildup lasting friendships and a working communitynetwork. The more opportunities there are tohave informal conversation, the more likely it isthat a community might work. Awareness of thelocation and social presence of others providedby mobile technology may often serve as amotive to start informal conversations by chanceor to ask for help. The evaluation of theMOBIlearn system should show if the frequencyof communication increases, and if people feelthat these new methods would improve thesocial network among the MBA students.

4.3 Mobile technology enablescase-study exercises which might beintegrated into everyday businessDue to their small size and light weight, mobiledevices can always be carried with one. Thus, it isalways possible to get spontaneous access to aworkspace with all data, documents, tools orpeople. We assume there are spontaneousopportunities during everyday business lifewhere it would be possible to link them in someway to an ongoing case-study exercise – perhapsit would be the sudden chance to interview a keyperson met at the lunch table; or perhaps anunexpected incident that could be used as anexemplary anecdote from practice; or perhapssimply an idea triggered in the mind by someevent. Mobile technology provides ways to keepthe current context so that others canunderstand the situation (using integratedcamera; voice recording; saving the informationon current awareness status such as location,time or people around) and make immediatenotes that can be used as a reminder andnotification for other group members.

Frohberg 83

4.4 Mobile technology helps to reducethe cognitive load of MBA studentsMBA students are managers and usually have avery high cognitive load. They have only a verylimited capacity to keep details and issues of littleimportance in their mind long enough for them tobe handled. Mobile technology provides themwith an opportunity to deal with more of thesetasks when they are not in their office, as long asthey can be done online right away. Sending ane-mail or message or reaching the right andavailable partners quickly to make decisions,helps them to tick things off the list as soon asthey appear (Straub and Karahanna 1998).

5 Test and evaluation plans forMOBIlearn’s final user testsin ZurichAs presented in this paper, there are many facetsof mobile learning that will be covered in the usertests. The success of all evaluation plans andactivities will depend strongly on theperformance of the MOBIlearn system which willbe available at the beginning of September 2004.In order to ensure a high-quality system and toallow untroubled user tests, MOBIlearn partnerswill run a series of technical functional tests andpre-tests to check its usefulness. According tothe MOBIlearn project schedule, the user testswill basically be done with undergraduatestudents until December 2004. If those tests aresuccessful and indicate some added value asexpected, further user tests with MBA studentscan be done in spring 2005.

ReferencesEvans D and Taylor J (2005). The role ofuser scenarios as the central piece of thedevelopment jigsaw puzzle. In J Attewell andC Savill-Smith (eds) Learning anytime anywhere –a book of papers. Proceedings of the MLEARN2004 Conference, Rome, 5–6 July. London:Learning and Skills Development Agency.

GroupSystems (2005). Website.At www.groupsupport.com

Lehner F, Nösekabel H and Bremen G (2003).M-learning und M-education – Mobile unddrahtlose Anwendungen im Unterricht.Regensburg: University of Regensburg.

Schwabe G (1995). Objekte der Gruppenarbeit –ein Konzept für das Computer Aided Team.Wiesbaden: Gabler.

Straub D and Karahanna E (1998). Knowledgeworker communications and recipient availability:toward a task closure explanation of mediachoice. Organization Science, 9(2), 160–175.


AbstractThis paper presents an overview of our latestwork on developing a ubiquitous computingframework to allow seamless access toapplications from multiple devices. This paperintroduces our application session teleportation(AST) framework, designed to support themovement of application sessions betweendevices in personal area networks (PANs) andacross the internet. So far, we have used the ASTframework to allow web sessions to be sharedbetween Internet Explorer clients. In terms ofmobile learning, this raises further implicationsfor users who are collaborating through sharedweb sessions from different devices. We presentthe results from our research into design metricsfor content to be used in this type ofco-browsing environment.

Keywordsmobile learning, sessions, collaboration,device characterisation

1 IntroductionWithin a ubiquitous computing environment, auser will own and make use of many devices thatvary in functionality and power. Furthermore,these devices will most likely be interconnectedto form a personal area network (PAN). PANs aregenerally confined to a user’s personal operatingspace, which has a typical range of 10m, but maybe extended to 100m for environments such asthe home (Braley, Gifford and Heile 2000).Devices may autonomously join and leave thePAN through device detection and locationmanagement techniques in an ad hoc orseamless manner. For example, a user’s homePAN may automatically detect and add a personaldigital assistant (PDA) device, which could thenbe used to interact with other devices withinthe PAN.

Learning and collaborating using mobile devicesand sessions

In general, learning applications should supportaccess to meaningful content, provide a meansto undertake constructive activities, and allowdialogue and collaboration between learners (seeeg Fowler and Mayes 1999). Mobile learningtechnology adds further opportunities for usersto access content from different devices, formoving sessions between devices and forsupporting collaborative activities – all potentiallyfrom different types of device. This paper focuseson research addressing the technologyrequirements for mobile learning, and presentsearly results from research on how we can adaptlearning materials so that they are suitable forcollaboration and co-browsing from differenttypes of device.

2 Application session teleportationApplication session teleportation (AST) is ourcandidate framework for supporting sessionmovement in PANs and the internet. PANs maycomprise various types of device with a range ofcomputation/storage capabilities. For this reason,some PAN devices such as mobile phones,digital badges and speakers may rely on morepowerful devices to teleport application sessions.This has subsequently led us to identify anumber of classes of device and modes ofinteraction – namely, controller, listener, minimaland output nodes.

Gardner Chua Shahi 85

Hui-Na [email protected] ExactzAsian Research CentreKuala LumpurMalaysia

Anuroop [email protected] of EssexAdastral ParkIpswichSuffolk IP5 3REUnited Kingdom

Michael [email protected] of EssexAdastral ParkIpswichSuffolk IP5 3REUnited Kingdom

Figure 1 Controller node architecture

Figure 1 depicts the controller node architecture.Listener nodes will be constructed from a subsetof these components (such as session transfer,security and content adaptation). Minimal nodesare essentially thin clients, with lightweightcomponents for session transfer and security,which interact with controller nodes to transfer asession. Finally, output nodes may only processoutput streams instantiated from controller andlistener nodes. Profiling functions are containedin the Context Awareness component and will berequired to hold information regarding users,PAN devices, AST applications and theircurrent locations.

Currently, we have built an initial prototype of theAST framework by implementing the SessionTransfer and Device/Service Detectioncomponents. To implement our framework,Internet Explorer has been adapted to work withAST. This TeleWeb concept demonstratorsupports web session teleportation within PANs.

3 Session adaptationWe are also separately carrying out research intothe AST Content Adaptation component (seeHoh, Gillies and Gardner 2003). Generally, alimiting factor of mobile devices is their smallscreen size and limited processing/storagecapabilities. Application teleportation may involvetwo or more PAN devices with differingcapabilities. Therefore it will be important toadapt sessions to the characteristics of eachparticular device.

Recently, much work has been undertakenelsewhere to adapt screen content for displaypurposes (see Esenther 2002). For example,many applications adapt web pages to complywith bandwidth, screen size and computationalrestrictions. Traditional content adaptation forsmall devices has involved using techniques suchas discarding irrelevant formatting information.However, novel approaches (eg Fox et al. 1998)are aiming to extend the usability of existing webcontent for small devices.

4 Collaborating fromdifferent devicesOnce users can access applications from anydevice, we can then consider scenarios thatinvolve users collaborating by means of theseapplications on shared learning tasks. However, ifthey are accessing the application from differentdevices with different display characteristics,then they may well have completely differentviews of the same shared information. It may notbe possible to view the contents of all these webobjects on screen at the same time, especiallywith smaller device display screens that can onlydisplay a limited set of web objects. The mainobjective of our research is to adapt the relevantinformation based on the perceived ‘utility’ of theweb objects and the capabilities of the devicesbeing used. The determination of ‘utility’ could bean automatic or a manual process. We arecurrently considering a manual method wherebyusers pre-select the utility of the web objects.Figure 2 (opposite) illustrates a scenario withtwo users collaborating from different devices.


5 Design guidelinesfor co-browsingGiven this scenario, how can we adapt learningmaterials so that they are suitable forcollaboration and co-browsing (see Chua et al.2004) from different types of device? We arecurrently carrying out research to investigatewhether the design metrics of web pages thatare perceived as being ‘good’ for single-userbrowsing are the same as the design metrics forpages that are perceived as being ‘good’ forco-browsing. One reason for believing that thedesign metrics might differ is that co-browsingrequires the users not only to search through thecontent for relevant material, but also to describethat material and their interactions with it to theirfellow users with whom they are co-browsing.If the metrics of ‘good’ pages do differ, thenweb-page content adaptation schemesimplemented in co-browsers will need to alterfurther the composition and formatting of thecontents within the page, to take into accountthe task at hand as well as the capabilities of thedevices being used.

Analysis of our preliminary results indicates thatthere are only a few differences between whatconstitutes ‘good’ and ‘bad’ web pages forbrowsing and co-browsing tasks. However, theresults do indicate that the gap between ‘good’and ‘bad’ pages in co-browsing is morepronounced, which suggests that accounting fortask metrics in the adaptation of web pages willbe important.

6 SummaryThis paper has presented an overview of anumber of research activities based around ourAST framework and focused on the issues ofapplication session teleportation, sessionadaptation, heterogeneous device collaborationand design metrics for co-browsing. We haveused the AST framework as the basis for anumber of research experiments. So far, theexperimentation we have done on co-browsingsuggests that adapting web-page content for theco-browsing task is best done by emphasisingmore text and adding large graphics. While itwould be very difficult to insert suitable graphicsas useful visual cues within a web page, theadding of emphasis to text headers andkeywords in web pages is a real possibility forinclusion within automatic content adaptationsystems. In the future, this may have a directimpact on the design and development ofcontent for mobile learning services.

Gardner Chua Shahi 87

Figure 2 Shared viewpoints

ReferencesBraley R, Gifford I and Heile R (2000). Wirelesspersonal area networks: an overview of the IEEEP802.15 working group. ACM SIGMOBILEMobile Computing and Communications Review,4(1), 26–33.

Chua HN, Scott SD, Tham VK, Gardner M andBlanfield P (2004). Design metrics forco-browsing. Paper presented to the 7thInternational Conference on Work withComputing Systems, Kuala Lumpur,29 June–2 July.

Esenther AW (2002). Instant co-browsing:lightweight real-time collaborative web browsing.In Proceedings of the Eleventh InternationalWorld Wide Web Conference, Honolulu,7–11 May.

Fowler CJH and Mayes JT (1999). Learningrelationships from theory to design. Associationfor Learning Technology Journal, 7(3), 6–16.

Fox A, Gribble SD, Chawathe Y and Brewer EA(1998). Adapting to network and client variationusing active proxies: lessons and perspectives.IEEE Personal Communications, 5(4), 10–19.

Hoh S, Gillies S and Gardner M (2003). Devicepersonalisation – where content meets device.BT Technology Journal, 21(1).


AbstractExploiting the capabilities of an increasinglypowerful and ubiquitous network is the keyenabling factor in introducing new collaborationand learning paradigms, where participants donot need to be physically located in the sameplace. Even more important, in the context of theresearch and university world, this model permitsthe sharing of costly equipment betweengeographically distant laboratories, as well asproviding students in remote locations withaccess to concrete scientific instruments.

While most of the network issues, such asquality of service (QoS), mobility and security aregradually being solved, an adequate set ofapplication-level technologies must be put inplace to allow for this model to be deployed.

KeywordsMPEG–4, distance learning, IP QoS

1 IntroductionThis paper presents the activities andexperiments carried out by the authors in thecontext of several European and Italian researchprojects, which initially aimed at providing atechnological platform for accessing culturalheritage information through advanced andinteractive multimedia systems. As a naturalextension of this platform, instruments forproviding an enhanced learning experience havebeen developed, which add support for external,media-synchronised learning objects.

The context of such activities is a real-worldscenario where the teacher, the support material,the application service provider and the studentsneed not necessarily be co-located, either in spaceor in time. This approach opens up a scenario wheredidactic material is not limited to written text orslides, becoming a true multimedia and hypertextualcollection of objects supporting the teacher intransferring knowledge to his or her audience.

From virtual cooperation to distance learning:realising remote multimedia platforms on a QoS-enabled network

Embracing the most recent networkingtechnologies and architectural models enablesstudents to attend their lessons from a remotelocation (be it their home, a remote classroom orconference centre), extending the reach of theplatform to telecommuting students, workersand older people; and overcoming the constraintsof time and place imposed by the traditionalface-to-face training methods.

2 MPEG–4 and the e-learningmedia platformThe International Organization forStandardization/International ElectrotechnicalCommission (ISO/IEC) MPEG–4 standard(ISO/IEC Standards Committee 2001) is acomprehensive set of specifications relating tothe encoded representation of audio/visualcontents. The outstanding improvement providedby MPEG–4 over its predecessors (MPEG–1 andMPEG–2), besides the availability of betteroptimised encoding algorithms and tools, lies in anew approach to contents description: media isnot just a video frame and an associated audiotrack, but rather a complex scene composed ofseveral audio-video objects (AVOs) which areassembled together with precisespatial-temporal relationships. Video objects canbe just plain old rectangular frames, but they mayalso be shaped objects which are ‘composed’ toform the final view. Both natural (ie coming froma camera or a microphone) and synthetic (ie 3-Dmodels and parametric descriptions) AVOs canbe freely intermixed in an MPEG–4 scene;furthermore, they can be made ‘interactive’ byassociating commands and actions with them.Finally, AVOs can be associated with quality ofservice (QoS) descriptors and transportedthrough the network over distinct channels (each,potentially, having a tailored QoS profile).

Giordano Insolvibile Chionsini Polese 89

Stefano [email protected] ofInformation EngineeringUniversity of Pisavia Caruso56122 PisaItaly

Gianluca [email protected] and Multimedia R&DneXtworks SrlVia Turati, 43/4556125 PisaItaly

Valentina [email protected] Informatica eTelecomunicazioniConsorzio Pisa RicercheCorso Italia 11656125 PisaItaly

Pietro [email protected] ProgramsAlcatel Italia SpAVia Trento, 3020059 Vimercate (MI)Italy

This broad set of tools makes MPEG–4 the idealchoice for implementing a technological platformwhich is able to deliver enhanced contents for aninnovative user experience. The components ofthe e-learning media platform endowed withMPEG–4 functionalities are the media productionand delivery chain (authoring tools, media serverand client) and the so-called QoS provision layer.The production chain includes some basic toolsfor the creation of content in the format specifiedby the standard (MP4 file format) and for itssegmentation into tracks suitable for delivery viathe network. The delivery chain comprises themedia server and client, both of which are able tohandle MPEG–4 AVOs and scene descriptions.

The QoS provision layer is an implementation ofMPEG–4 DMIF (Delivery Multimedia IntegrationFramework), a media-unaware andnetwork-aware signalling protocol allowing theset-up of individual channels and their associationwith AVOs. The signalling messages also carryinformation on the QoS requirements foreach AVO.

3 Extensions for distance learningapplicationsThis e-learning platform, initially designed totarget the delivery of enhanced media in acultural heritage context, has proved to besuitable also as the basic building block of adistance learning system. The ability to intermixsynthetic and natural AVOs and to deliver themwith QoS guarantees has provided a goodstarting point for experimenting withdistance-based training courses.

The need to have a virtual whiteboard applicationhas been easily satisfied by making the teacher’sfree-hand drawing application (MicrosoftWindows Journal running on a Tablet PC) workremotely.The particular nature of the distancelearning scenario requires full synchronisationbetween the audio/video streams and thewhiteboard data stream, as well as the ability totransmit the latter using the same network channelsand paradigms (unicast/multicast/broadcast).Hence, a custom application has beendeveloped, whose purpose is to capture periodicscreenshots of the selected application and tofeed them to the video encoder as syntheticobjects. The media server then delivers the‘whiteboard video stream’ alongside theteacher’s voice and video streams. As expected,the image quality obtained with this approach is quite high, since no intermediatedigital–analog–digital conversion takes placebetween the whiteboard and the end-user screen.

More ambitious applications of the technologyreveal the need for more sophisticatedfunctionalities to be built into the platform.Unfortunately, due to limitations both in thestandard and in the current implementation, theintroduction of such functionalities has implied aslight detour from the MPEG–4 specifications.

Pursuing a more generalised approach, thetraditional audio/video stream is augmented withsynchronous external objects which, according tothe context and adhering to literature usage,have been called learning objects. This genericterm designates, in the context of our research,any kind of additional material that can be usedby the teacher to provide the students with anenriched learning experience. More specifically,the implemented mechanism embedsproprietary triggers into the media streams,which bring to the client a specific ‘message’along with a timestamp. The message iscurrently a URI (Uniform Resource Identifier)pointing to the remote learning object or aparameter to be passed to the player.

As simple as this mechanism is, it actually allowssome interesting functionalities to be realised.The simplest example is the transmission ofslides synchronised with audio/video frames:presentation of each slide is activated by triggerswhich are inserted automatically whenever thedistantly located teacher advances his/herpresentation. Since triggers are saved togetherwith the media stream when archiving a livepresentation for later utilisation, this mechanismalso provides for a complete reproduction of thelive lesson.

More generally, a wide range of synchronisedevents on the remote user’s side are enabled.More complex examples include ‘active’ objects(eg simulations and virtual instrument panels) tobe downloaded to the client and synchronisedwith the presentation by using the triggers asclock references and input signals. Support formobile, platform-independent code (likeJava applets or .NET code) is currentlybeing investigated.


4 Network and mobility issues

4.1 Tackling the QoS issuesThe well-known ‘best-effort’ behaviour ofcurrent IP networks is a significant hindrance tothe wide deployment of enriched real-timemultimedia platforms, whose demanding QoSrequirements cannot be met without anadequate provisioning architecture.

The implementation of the experimental networkinfrastructure has been derived from thearchitecture envisioned by the InternetEngineering Task Force (IETF) in several Requestsfor Comment (RFCs) (Blake et al. 1998; Durhamet al. 2000; Bernet et al. 2002; and related RFCs2205, 2474, 2597, 2598, 2749, 3140). This isbased on the integrated services approach in theperiphery of the network and on thedifferentiated services approach in the core. Thisnetwork infrastructure fulfils the needs ofapplications that are demanding in terms of QoSby being able to differentiate between trafficflows, provide the applications with strictbandwidth and delay guarantees and efficientlymanage dynamic resources allocation.

The integration between the application platformand the network is granted by the interfacebetween the QoS provision layer (DMIF) and thenetwork’s border routers.

4.2 Mobility and making e-learningapplications work remotelyLearners want their content to be as mobile asthey are. At the same time, the power ofpersonal digital assistants (PDAs) and othermobile devices is increasing to the extent thatthis desire can now be fulfilled: learning materialsand multimedia contents can efficiently be madeaccessible to people who are on planes, or otherforms of transport, or just in the office. Thepossibility of mobile utilisation of e-learningservices is definitely one of the areas where themost important innovations are expected.

Integration between IP networks – internet,corporate local area networks (LANs) and virtualprivate networks (VPNs) – and next-generationmobile phone infrastructures [eg 3rd GenerationPartnership Project (3GPP)] is the majorchallenge that must be faced from this point ofview. Also, interworking with satellite networksis the ideal way to support provision of servicesin rural and mountain areas, thus reducing the‘digital divide’ and providing innovative ways ofteaching people in remote regions.

Early experiments with the e-learning platformhave been performed by making lessons taughtfrom the University of Pisa available remotely tostudents located in a suitably equipped classroomin the nearby town of Lucca, over a 34 Mbpsradio link. Figure 1 (overleaf) illustrates the virtualclassroom environment. External facilities, whichhave been introduced into the platform to make itmore usable, include a content managementsystem (for lesson planning and students’subscription) and an H.323 Voice-over-IP (VoIP)service platform (to provide real-time interactionbetween the remotely located students and theteacher). The same lessons have been madeavailable to students attending from their homeand accessing the network over 320/128 KbpsADSL links.

The e-learning platform is also being madeaccessible from within the Alcatel MultimediaCommunication Centre (MMC: see Figure 2overleaf) in Vimercate in Italy, as part of atechnology programme whose main objectivesare to extend available e-learning platforms andto realise a distributed testbed interconnectingseveral ‘islands’ (isolated testbeds) includingPisa, Rome, Naples and Vimercate. The testbed,which also includes the Alcatel 3G Reality Centrefor mobile demonstrations and integrationwith 3rd-generation Universal MobileTelecommunications System (UMTS) mobilephones, is currently used for experimentationand development.

5 ConclusionsThe technologies required for the realisation ofenhanced e-learning and distance learningplatforms are definitely mature enough to startthinking about real-world commercialdeployment. The synergy of hardwarecapabilities, network infrastructure andmultimedia applications cannot be exploitedwithout a convergent approach, combiningthese complementary aspects in a uniquearchitectural view.

Dealing with established working methodologieswhile proposing innovative services to the endusers, who here include both the teachers andthe students, is another critical issue. The mostpromising approach is to adhere as closely aspossible to commonly used teaching instrumentsand to develop easy-to-use, easy-to-customisesoftware platforms.


Figure 1 The virtual classroom environment

Figure 2 The Alcatel Italia Multimedia Communication Centre (MMC)


ReferencesAlbaugh P (1999). The role of skepticism inpreparing teachers for the implementation oftechnology. In Proceedings of the Society forInformation Technology and Teacher EducationInternational Conference, 1906–1909.

Blake S et al. (1998). An architecture fordifferentiated services. IETF RFC 2475. Reston,VA: Internet Engineering Task Force. Available viawww.ietf.org, accessed 12 September 2005.

Bernet Y et al. (2000). A framework for integratedservices operation over diffserv networks. IETFRFC 2998. Reston, VA: Internet Engineering TaskForce. Available via www.ietf.org, accessed12 September 2005.

Durham D et al. (2000). The COPS (CommonOpen Policy Service) Protocol. IETF RFC 2748.Reston, VA: Internet Engineering Task Force.Available via www.ietf.org, accessed12 September 2005.

ISO/IEC Standards Committee (2001). MPEG–4 –coding of audio-visual objects. ISO/IEC14496.Geneva: International Organization forStandardization/International ElectrotechnicalCommission.

Peters O and Collis B (2000). A constructivistframework for WWW-based learningenvironments that combine streaming media andWWW functionalities. In Proceedings of theWorld Conference on Educational Multimedia,Hypermedia and Telecommunications, 904–909.

PLS Ramboll Management (2004). Studies inthe context of the E-learning Initiative: virtualmodels of European universities. Århus:PLS Ramboll Management.


AbstractMobile technologies offer the opportunity toembed learning in a natural environment. Thispaper describes the design of the mobileGameprototype. The mobileGame exploresopportunities to support learning through anorientation game in a university setting. Thepaper first introduces the scenario and thendescribes the general architecture of theprototype. The second half of the paper focuseson requirements that have evolved during thedesign, implementation and testing of theprototype: supporting work on the move posesdifficult interface questions – the accuracy ofcurrent outdoor and indoor positioning systemsis still problematic; and the game requires a nearreal-time response time. This requires adistribution of functionality and data between theserver and the client and a careful design of theinterface. The success of the game dependsheavily on the extent to which its design (ie theset-up and its rules) motivates the players. Asurprising number of user roles surface once thegame is implemented in a natural environment.

Keywordsmobile and wireless games, mobile learning,e-learning, computer-supported cooperative play,positioning systems

1 IntroductionToday, we work increasingly with mobiletechnologies. There are mobile phones forcommunication nearly everywhere, laptops towork wherever we want and personal digitalassistants (PDAs) to get access to informationwhenever we want. It is just a question of timeuntil mobile technologies are used for otheraspects of our life. One of these is learning.Sharples, Corlett and Westmancott (2002) saythat the technology for mobile learning is nowavailable. Research in the area of mobile learning,

Requirements for mobile learning games shown on a mobile game prototypeChristoph GöthUrs-Peter HässGerhard [email protected]@[email protected]

Information Management Research GroupDepartment of InformaticsUniversity of ZurichWinterthurerstrasse 190CH–8057 ZürichSwitzerland

however, is still in its early stages, but the resultsshow that mobile technology provides a new wayof learning.

One of these new ways could be the integrationof learning and playing, supported by a mobileenvironment. There are a few systems thatintegrate playing and learning such as theCooties Game or Geney (Savill-Smith and Kent2003), but they focus on role-play or simulation.Prototypes and commercial versions oflocation-based games set in a real-lifeenvironment, like CYSMN (Benford et al. 2003),Pirates! (Björk et al. 2001) or Mogi (Hall 2004),show that people like to play with the newoptions, but these games are focused only onentertainment. Below we describe a scenariowhich shows a new way of learning, combininglocation-based games and mobile learning. Thisscenario was implemented in a prototype fortesting. We describe the architecture of thisprototype before we discuss the requirementswhich we found in our first test with users.

2 The scenarioMark Finder is a new student at the University ofZurich and has just arrived for the orientation day.This year, the traditional orientation rally iselectronically supplemented with handhelddevices. The orientation rally is a fun event whichhelps students to get to know the university andits surroundings. Therefore, the rally will lead allparticipants through an area with several tasks tofulfil at certain spots. Mark is asked to fill in anonline form with his personal profile (nationality,gender, age, personal interests, hobbies,favourite food, etc). All new participatingstudents are automatically grouped by theirprofiles in order to obtain homogeneous groups.Mark meets five other students with similarinterests in theatre, jazz, history and biking. Eachgroup receives a handheld computer.

Göth Häss Schwabe 95

Figure 1 Playing the game

During the orientation rally, each group receivesdifferent tasks referring to significant places,people and events (explained below). Thehandheld device shows the current position ofthe group on the digital map of the university(see Figure 1).

When the group enters a building, the outdoormap switches to an indoor map of that building.The whole rally is structured as a cooperative andcompetitive game. The competitive element isbased on hunting rules: each group tries to catchanother group and equally, is hunted by anothergroup.1 The handheld device shows each groupwhere its hunter and its prey are located.Cooperation rules force group members to meetmembers from other groups as well as teachersand to exchange information with them – again,they are supported with location-basedinformation on their displays. The tasks given tothem provide them with basic information onuniversity life. There are the following typesof task.

p Significant place tasks: the students have to findimportant places such as the library, the cafeteriaor the laboratories. At each location, they have toperform a typical task (find a book, have lunch,etc). The specific tasks are context-dependent(they depend not only on the location, but also onthe time of day, or they build on the activity of aprevious group). The task execution is supportedby the handheld device (eg serving as a front endto the library information system or providingthem with needed information).

p Significant people tasks: the students have tofind important people at the university (thepresident, the study coordinator, the caretaker)and interview them on their jobs – these peopleeither actually participate in the game or areplayed by older students. If those people aretypically mobile, they can be located by amobile device.

p Significant event tasks: the significant events canbe scheduled or be ‘by surprise’. Scheduledevents include introductory lectures and courses.Here tasks relate to the organisation of studies(eg set up a course schedule; how to findimportant information) and some initial content.Unscheduled events include ‘spontaneous’welcome parties by student groups, but also thesign-up of each group member to importantuniversity services (eg computer account,library card).

Each task requires the group to answer one ortwo simple questions displayed on the handhelddevice. For example, one task might be to findthe cafeteria. There they get the question ‘Whatis the price of an apple pie?’ They will not get thenext task until the correct answer is given.

In addition to the game, the students canannotate real objects with virtual ‘post-its’. Othergroups can read these short messages andanswer with their own ‘post-its’. Spectators arealso integrated. They can watch the game via aweb browser. The groups and the spectators canuse the integrated instant messenger tocommunicate with each other.

The mobileGame scenario substitutes theteacher with a PDA that gives tasks to thestudent. This corresponds to constructivisttheory, which requires the teacher to act as acoach. Learners are encouraged by the tasks toact on their own, in the real-world context of theuniversity.They individually construct their ownknowledge of the university. Additionally, thegame is played by groups of students with thesame interests. This links the game with asuitable social setting, as postulated by thesituated cognition theory (Hutchins 1994).


3 The mobileGame prototypeThe scenario describes a sort of synchronouscollaboration between the players. So wedecided to derive the architecture of themobileGame prototype from a commoncomputer-supported cooperative work (CSCW)architecture concept for synchronouscollaboration (see Figure 2; Schümmer andSchluckmann 2001). There are the clients withtheir own private view of the game in progress.Generally, all relevant data is stored on the clientside, so the player could act even if there is noconnection to the server available. The serverworks as the central coordination point for thewhole game. Dynamic changes in the game aretransferred to the server which distributes thedata. Awareness regarding the accuracy of thetransferred information is handled by atimestamp mechanism. Thus, a player alwaysknows if s/he is working on ‘fresh’ data or if themessages from other players are obsolete.

3.1 The technique of the prototypeThe architecture integrates three components:the mobile client on the PDA, the web client for abrowser and the server. The mobile client runs ina Java virtual machine2 on the PDA. Itcommunicates with the server via RemoteMethod Invocation (RMI). For example, the client receives information about the current position of the group, new tasks to solve, events or messages. The client displays the data on the screen of the PDA.

If the group captures other groups of players orsolves tasks, this information is sent back to theserver. Annotations are displayed in the PocketInternet Explorer. The browser exchanges datawith the server over HyperText Transfer Protocol(HTTP). On the PDA, there also runs an instanceof the Ekahau client. This program is independentfrom our own code and communicates with theEkahau positioning engine autonomously. If theplaying ground is not completely covered by aWireless Local Area Network (WLAN), the clientcan also get positioning information over a serialconnection to a Global Positioning System(GPS) receiver.

The web client is a Java applet running in theJava virtual machine of a web browser. It requiresthe Java 1.4 plug-in from Sun. The applet showsthe current status of the game and allowschatting between the playing groups. Thecommunication with the server is realisedover HTTP.

The server components are also written in Java.As described above, it communicates with thedifferent clients over RMI and HTTP. For thecommunication over HTTP, we use the Jettyservlet engine, a simple but efficient servletcontainer. MySQL and the ApacheObject/Relational Bridge (OJB)3 are used for thepersistent storage of game data. The server getsthe positioning information of the playing groupsfrom the Ekahau positioning engine, which itselfgets them from the Ekahau clients running onthe PDAs.


Figure 2 Architecture for distribution with centralised coordination

3.2 Positioning of the clientThere are two ways to localise the position of theclient. The first one is used in outdoor areas, whereno WLAN is available. A GPS receiver is pluggedinto the PDA which computes its position withthe GPS signals only. The problem of GPSpositioning is its inaccuracy. The accuracy rangesfrom 3m to 100m, depending on the weather andthe surroundings. There are many unwantedside-effects like position indicators jumping overthe whole digital map or groups not beingavailable at the point shown on the digital map.

The second way, which is much more accurate,is to use the WLAN to compute the position. Weused the Ekahau positioning engine for thispurpose. The Ekahau client on the PDA sends theinformation on the available WLAN access pointsto the Ekahau positioning server. The servercomputes the position, which is read from themobileGame server. The accuracy achieved isbetween 1m and 3m, if four access points areavailable. The disadvantage is that the engineneeds to be calibrated before you can use it. Thiscalibration took up to 1 hour per floor of abuilding, so you cannot use it ‘out of the box’.

4 Requirements andlessons learnedThe following requirements are based on a usertest which took place at the University ofKoblenz. The test was integrated into apresentation day for the Faculty of ComputerSciences. The mobileGame was shown to thevisitors and they could play a little game, in whichthey had to find a hidden PDA.

4.1 Work on the moveThe players got a PDA showing a map of twofloors, their own position and the position of thePDA that they had to find. Although the mapfilled 50% of the PDA screen and the floor wasonly a straight corridor, the players mentionedthat it was not possible to move and navigatewith the PDA at the same time. If they wanted tolook for their current position, they had to stopand look at the PDA. Even though they did nothave to interact with the PDA, they had to lookdown and compare the map on the screen withtheir surroundings. It is really hard for people towork on the move.

In the mobileGame, the players have to positionthemselves constantly. Of course, they do notwant to stop every 10 seconds and look at thePDA. Therefore, we have done someexperiments with head-mounted display thatpresents information close to the line of sight.

This option, however, has two disadvantages: theplayers have to wear big glasses and noteverybody likes to look like a cyborg or a robot.The other disadvantage is the distorted field ofvision. There are two kinds of head-mounteddisplay. The first is a little display, which isattached to normal glasses. The second is a pairof big glasses with integrated video input andoutput. Two video cameras in the glass recordparts of the field of vision and display it on twoscreens in front of the player’s eyes. This can beaugmented with additional information.Depending on the model, the player either has torefocus the eyes on the little display on theglasses; or wear big glasses which only present arestricted field of vision.

We believe that an audio interface, which givesthe user the necessary information throughheadphones, may be the answer to that problem.Björk et al. (2001) have experimented with audiofor status information in their game Pirates!However, they found that often this informationcould not be heard because the noise level withinthe gaming area was too high. So there must bea trade-off between the transmission ofinformation through transient audio andpersistent visualisation on the PDA screen.

4.2 Accuracy of positioningOne of the biggest mobile games was Can yousee me now? designed by Benford et al. (2003).Mobile actors hunt virtual players in Sheffield. Thepositioning system for the actors is based on aGPS. Their main problem was ensuring accuracyof up to 106m, but with some tricks, the actorshave managed to overcome this problem. Ourprototype uses GPS and WLAN for positioning,as described above, with the result of anaccuracy of 3–5m. The players, who need thepositioning information for their orientation, havetold us that the accuracy was quite good as longthey only needed to know their approximateposition. As soon as they had to know the exactposition of an object – in our test, the position ofthe hidden PDA – the accuracy was notsufficient. They had to search in up to threerooms for the PDA.

For each domain of positioning, you need aseparate system. Global and local systems canbe integrated to get a much better result for theplayers. We are planning to add a third systembeside the GPS and the WLAN system, which isbased on Radio Frequency Identification (RFID).Points of special interest can be tagged withRFIDs, so a player who reaches this pointreceives a signal and knows that his/her positionis accurate within up to 1m.


4.3 Offline areas and shortreaction timeOne of the main reactions of the players in thetest was that the update time of their positionwas much too slow. We update their positionevery 3 seconds, but in this time, the playersmove 5m or more. This is one of the mostimportant requirements of mobile games.Usually you do not have an area-wide WLAN, sothe mobile clients are not always connected tothe server. But the players want a near real-timereaction of the client. Game objects, which arechanging all the time, like the position of thegamers, have to be updated by client as fast aspossible. This means on the one hand, that themobile game must have a very good cachingalgorithm; and on the other hand, that a veryefficient data transmission strategy is needed.

In our prototype, all static information, such asthe game maps, is stored on the PDA. Only thedynamic information, such as a new position or anew task, is transmitted. The PDA gets all theinformation from the server that it needs foroffline work, like the answer on the current task,so that the players can interact even if they are inan area without a WLAN. Additionally, they getthe status information about their network statusand the status of other players with whom theywant to interact.

4.4 Easy-to-use interfaceFigure 3 shows the new and the old interface.The first interface uses drop-down menus as ondesktop programs. The observation of the playersshowed that navigation with the drop-downmenu and the pen of the PDA is not reallyintuitive. The use of the PDA is much more likethe use of an automat. So we redesigned theGUI and substituted a button bar for the oldmenu. Now all functions can be reached by oneor two clicks. We also use symbols instead oftext because they are smaller than text and theusers understand symbols much better.

One big problem is still the interface foranswering the questions of the task. On the onehand, you can use open questions, which givesthe game designer a broad range. Then the usershave to write the answer down with the pen anda little virtual keyboard on the PDA, which is hardto use. On the other hand, you can usemultiple-choice questions, which restrict thegame design, but users can answer questions bysimply clicking. At this point in time, we supportopen questions, multiple-choice questions andquestions with sliders, where you can ask fornumbers. We think an audio interface could helphere too.

4.5 Roles in the gameIn the test in Koblenz, we used only one client –the mobile client. The players as well as thegame master use the same functions with thesame client. Some functions werespontaneously ‘re-defined’ because there wereno adequate functions given by the client. Afterthe test, the results were discussed in a meetingwith experts. As a result of this discussion, it wasagreed that a mobile game needs to supportmany roles in the game. There is the player, thegame master, the spectators, the visitors, thetutors and the actors for special events. Alsothere is the game designer and administratorsupporting the game. All these people need aspecial client with special functions. At this pointin time, we only have a client for the players andfor the spectators.

Figure 3The new and the old graphical user interface (GUI)


5 Ongoing researchThe next step of our research was a big user testwith new students at the University of Koblenz atthe end of April 2004. As described in thescenario (section 2), new students should learnfrom mobileGame how to live on campus andwhere to find the most important places. In thepresentation at this conference, a first result ofthis test can be presented. After the analysis ofthe test, the prototype should be revised. Thereare plans to test the revised version at theUniversity of Zurich in the autumn of 2004.

ReferencesBenford S, Anastasi R, Flintham M, Drozd A,Crabtree A, Greenhalgh C, Tandavanitj N,Adams M and Row-Farr J (2003). Coping withuncertainty in a location-based game.IEEE Pervasive Computing Journal,July–September issue, 34–41.

Björk S, Fals J, Hansson R and Ljungstrand P(2001). Pirates! Using the physical world as agame board. Paper presented at the 7thIFIPTC13 Conference on Human–ComputerInteraction, 9–13 July,Tokyo, Japan.

Hall J (2004). Mogi: second generationlocation-based gaming. At www.thefeature.com/article?articleid=100501&threshold=–1,accessed 2004.

Hutchins E (1994). Cognition in the wild.Cambridge, MA: MIT Press.

Savill-Smith C and Kent P (2003). The use ofpalmtop computers for learning. A review of theliterature. London: Learning and SkillsDevelopment Agency.

Schümmer J and Schluckmann C (2001).Synchrone Softwarearchitekturenp. InG Schwabe, N Streiz and R Unland (eds)CSCW-Kompendium. Heidelberg:Springer Verlag, 297–309.

Sharples M, Corlett D and Westmancott O(2002). The design and implementation of amobile learning resource. Personal andUbiquitous Computing, 6, 220–234.

Endnotes1 The didactic reason for hunting rules is to keep

the groups moving. Of course there need to behunting-free areas and times; for example,during courses.

2 This JVM has to support the Personal Java 1.2specification (http://java.sun.com/products/personaljava/).

3 The OJB is an object/relational mapping tool that allows transparent persistence for Java objects against relational databases.See http://db.apache.org/ojb/


AbstractThis paper explores Wenger’s ‘community ofpractice’ concept and relates his theory to acommunity of practitioners implementingportable and wireless technologies in schoolsand colleges in the UK.

Keywordsportable, wireless, community of practice, Toshiba

1 Background – information andcommunications technology(ICT) in schoolsDepartment for Education and Skills (DfES)funding for e-learning in schools in the UK hasincreased substantially since it began in April1998. In 2002/03, it totalled £510m comparedwith a total of £657m over the 4 previous years.In addition, £230m was made available for stafftraining through the National Opportunities Fund.

A recent Office for Standards in Education(Ofsted) report (2004) entitled ICT in schools –the impact of government initiatives five years onis a timely reminder of the constant pressure oneducators to demonstrate that the investment intechnology is having the required impact on thelearning process and outcomes. The reporthighlights the main positive aspects of the workof The National Grid for Learning including:laptops for teachers, strategic leadership in ICT,Curriculum Online, enhancing subject teachingusing ICT, the continued funding of the RegionalBroadband Consortium, The Test Bed Project.

The main findings suggest that the continuedimpact of government initiatives for ICT inschools has been significant. The outcomes ofthe initiatives are more evident in improvementsin pupils’ achievements in ICT capability than intheir application of this learning in other subjects.The incidence of effective application of ICT inlessons across subjects is, however, increasingslowly but surely.

A community of practice? The Toshiba Ambassadors programmeBob [email protected] Adviser, Toshiba Information Systems(UK) Ltde-Learning Consultant, DfES Standards UnitSupport for Education and Training16 MeadowgateUrmstonManchester M41 9LBUnited Kingdom

However, the gap between the best and worst inICT provision is unacceptably wide andincreasing. As yet, the government’s aim –for ICT to become embedded in the work ofschools – is a reality in only a small minority ofschools. More typical is a picture in which pupils’ICT experiences across the curriculum aresporadic and dependent on individual teachers; inmany schools, opportunities to exploit technologyare not taken on a daily basis.

Some schools have been highlighted in ICTSchool Portraits, a European initiative (see ICT2005). Another European initiative is the ToshibaAmbassadors programme and case studies(Toshiba 2005).

2 Toshiba Ambassadors (UK)Formed in 2001, the Toshiba Ambassadors (UK)are part of a European network of leaders andmanagers in schools and colleges. The focus ofthe network is the implementation of wirelessnetworks and the impact on teaching, learningand management.

Toshiba Ambassadors case studies include anumber of schools, colleges, Local EducationAuthorities (LEAs), Education Action Zones(EAZs) and Excellence in Cities institutions (EICs).The pupils of two schools, Leigh City TechnologyCollege (CTC) and Arnewood School ran aworkshop at the MLEARN Conference in 2003.

Each institution has its own unique experience ofimplementing wireless technology in teachingand learning.

Toshiba provides the support for the network andthe group receives regular visits from Toshibatechnical and business development staff.

Harrison 101

The most beneficial aspect for the educators isthe seminar programme, comprising workshopsheld three or four times a year and hosted byToshiba in partnership with one of theAmbassador organisations. The format forthe workshops varies, but usually has thefollowing components:

p an overview of recent developments in ICT ineducation from the DfES

p a research update from the British EducationalCommunications and Technology Agency (Becta)/the DfES

p presentations from each of the Ambassadors onrecent developments and the evidence for impact

p a technical update from Toshiba

p a presentation from the host organisation oncurrent issues

p feedback from Ambassadors on technical andpedagogical issues in wireless/portability

p new product testing/trialling.

The initiative also has an international dimension:activities have included a study visit to theBenelux countries to exchange ideas with otherEuropean Ambassadors and a visit to the Toshibamanufacturing plant in Germany. SeveralAmbassadors have cooperated on joint projectapplications and a successful NetworkedLearning Communities project includes threeAmbassadors and a school in South Australiawhere a Toshiba Ambassador went to teach.

The Ambassadors attend the British EducationalTechnology Conference and exhibition (BETT) andprovide advice and support not only to eachother, but also to other schools and collegeswhich are beginning their wireless journey.

The Ambassadors also play an active role in avariety of other communities including theNational Association of Advisers in ComputerEducation (NAACE), the Association of LearningTechnology (ALT), the Becta ICT ResearchNetwork, the British Educational Leadership,Administration and Management Society(BELMAS), the Specialist Schools Trust and thee-Learning Foundation.

3 Wireless networkingand portabilityWireless networking in schools and the use ofportable devices featured in Ofsted’s recentreport (2004): ‘The purchasing of wirelesslaptops can greatly increase flexibility … the useof laptops on trolleys that can be wirelesslylinked to the school network in different parts ofthe school has had a profound impact where thishas been part of a co-ordinated and sustaineddrive to improve the uptake of ICT acrosssubjects.’ The ‘e class 2go’ was featured at theMLEARN Conference of 2002.

Indeed Perry (2002) suggests that a wirelessnetwork can help teachers to:

p work more efficiently

p better support their pupils’ learning throughtheir own use of ICT

p use ICT to extract greater value from their teaching

p work wherever and whenever suits them best.

He claims that:

The flexibility of wireless networks and portablecomputers means that lessons using ICT indepth can take place in ordinary classrooms,allowing it to serve children’s learning needs andnot dominate them, and to improve teachers’confidence in innovating. Schools have found thatthe flexibility of wireless networks supports bothteachers’ teaching and administrativeresponsibilities highly effectively.

He uses three examples – from the City of YorkEducation Authority, Dorrington Primary Schooland Sawtry Community Technology College.

A recent Becta report (2004) which analysed theavailable research into the use of portable ICTdevices in teaching and learning summarises thekey benefits of using portable devices as follows.

3.1 General benefitsp Portable ICT devices do not dominate in the

same way that desktop computers can, and maybe more readily integrated into classroom useand across the curriculum with the minimum ofdisruption to existing practices.

p The use of notebook computers, together withwireless networking technology, allows ICT workto be done in the classroom, saving both spaceand the time needed to move to speciallyequipped ICT suites.

p Portability enables students to take work hometo continue working, and this can foster greaterfeelings of ownership over work.


3.2 Benefits for studentsp There are gains in understanding and

analytical skills, including improvements inreading comprehension.

p Writing skills are developed: this includes spelling,grammar, punctuation, editing and redrafting aswell as fluency, originality and elaboration.

p There is increased motivation, organisationalskills and responsibility among pupils.

p Portable devices encourage independent andactive learning, and self-responsibility for learning.

3.3 Benefits for teachersp There are gains in ICT literacy skills, confidence

and enthusiasm.

p It is easier to plan and prepare lessons and todesign materials.

p They have greater ability and confidence tosupport students’ learning with ICT.

p They have access to up-to-date pupil and schooldata, any time and anywhere.

p It increases the efficiency and accuracy ofday-to-day registration of pupils.

p It enhances the professional image that theyproject to colleagues.

3.4 Benefits for parentsp There is increased involvement in education

for parents and, in some cases, improvedself-esteem.

p There is increased knowledge of their children’slearning and capabilities, owing to an increasein learning activity which is situated in the home.

4 Wenger and communitiesof practiceEtienne Wenger (1998) poses the question:‘What if the key to complex knowledgechallenges faced by most organisations todaylies in the age old, utterly familiar, and largelyinformal social structures known today ascommunities of practice.

Wenger (2001) further developed the conceptand related it to the role of technology insupporting communities in practice, and whilethe main thrust of this work was the explorationof appropriate technological platforms, some ofthe key principles resonate with the activities ofthe Toshiba Ambassadors.

According to Wenger, not every community is acommunity of practice. The fans who cometogether in the City Of Manchester Stadium tosupport the Manchester City football team canbe described as a community, but not acommunity of practice. He asserts that threecharacteristics are essential:

The domain: since a community of practice isfocused on a domain of shared interest, it is notmerely a club of friends or a network ofconnections between people. Membershipimplies a minimum level of knowledge of thatdomain – a shared competence thatdistinguishes members from other people.

The community: in pursuing their interest in theirdomain, members engage in joint activities anddiscussions, help each other and shareinformation. That is how they form a communityaround their domain and build relationships.Having the same job or the same title does notmake a community of practice unless themembers interact and learn together.

The practice: a community of practice is notmerely a community of interest or people wholike certain kinds of film, for example. Membersof a community of practice develop a sharedrepertoire of resources, experiences, stories,tools, ways of addressing recurring problems –in short, a shared practice.

Wenger goes on to say that communities of practiceare not a novelty. They are not a new solution toexisting problems: in fact, they are just as likely tohave been involved in the development of theseproblems. In particular, they are not a design fad,a new kind of organisational unit or a pedagogicaldevice to be implemented. They are about content,learning and a living experience of negotiatingmeaning, they are not about form. They cannotbe legislated into existence or defined by decree.They can be recognised, supported, encouragedand nurtured, but they are not reified designableunits. Practice itself is not amenable to design.

Harrison 103

5 Conclusion and future workThere are currently 12 members of the ToshibaAmbassadors (UK). There is strong evidence thatthey have the potential to mature into acommunity of practice, as Wenger (1998, 2001)defines it. They have formed a symbioticrelationship that has resulted in organisationaland individual development. Most importantly,they have had a significant impact on teachingand learning within their own schools andcolleges and shared best practice at local,national and international levels.

Further work will focus on the maturation of thecommunity and the addition of new members aswell as an evaluation of the impact of theAmbassador programme.

ReferencesBecta (2004). What the research says aboutportable ICT devices in teaching and learning.Coventry: British Educational Communicationsand Technology Agency.

ICT (2005). ICT school portraits.At http://schoolportraits.eur.org, accessed10 September 2005.

Ofsted (2004). ICT in schools – the impact ofgovernment initiatives five years on. London:Office for Standards in Education.

Perry D (2002). Wireless networking in schools.Coventry: Becta/DfES/Technology College Trust.

Seale J (2004). The development of accessibilitypractices in e learning: an exploration ofcommunities of practice. ALT Journal ofResearch in Learning Technology, 12(1).

Toshiba (2005). Toshiba Ambassadorsprogramme. At www.freedomlearning.co.uk,accessed 10 September 2005.

Wenger E (1998). Communities of practice:learning, meaning and identity. New York:Cambridge University Press.

Wenger E (2001). Supporting communities ofpractice – a survey of community orientedtechnologies. A shareware report.At www.ewenger.com/tech, accessed14 September 2005.


AbstractConsuming learning material via mobile devicesis deemed to be much more tiring andsignificantly less efficient than learning fromprinted material. In the likely scenario that usersof online learning environments will escape themby printing out the content, they will typically losedidactically relevant features like interactivity,hyperlinks, colour, video and audio. ThePaperLinks approach tries to overcome thisdilemma by supporting the learner in accessingsuch computer-based features while they arereading the corresponding printed material.PaperLinks are systematic augmentations ofprinted versions of hypermedia content that workin such a way that accessing online features isstill only one ‘click’ away, provided that a mobiledevice is kept handy.

Keywordsdigitally augmented paper, hypermedia,Extensible Stylesheet Language (XSLT)

1 IntroductionWhile there seems to be a unanimous trendtowards e-learning environments, the fact thatreading text from computer screens isconsidered much more fatiguing and significantlyless efficient than reading paper text cannot beignored, especially in the domain of mobilecomputing with small screen sizes, andparticularly in the context of learning, where textis being read over and over (Mills and Weldon1987; Learn and Mirski 2003). Thus, it is verylikely that users will be printing out the contentprovided by their learning environment, therebytypically losing didactically relevant features likeinteractivity, hyperlinks, colour, video and audio.

PaperLinks – linking printouts to mobile devicesMartin HitzStefan [email protected]@isys.uni-klu.ac.atDepartment of Informatics SystemsUniversity of KlagenfurtUniversitätsstrasse 65–67A–9020 KlagenfurtAustria

To save the ‘didactic power’ of such features, it isessential that the reader of paper text is offeredimmediate access to those computer-basedteaching elements when necessary. Theefficiency of the access method is absolutelycrucial – entering a complete URL via a PDA iscertainly not a realistic option.

In section 2, we give an overview of ourapproach to solving the problem, based on theconcept of digitally augmented paper(Guimbretière 2003). In sections 3 and 4respectively, the architecture of animplementation and specific input modessupported are discussed. An overview of futurework concludes the paper (section 5).

2 The PaperLink approachThe concept of PaperLinks is a straightforwardattempt to combine the advantages of two verydifferent but complementary media – traditionalpaper and hypermedia. The challenge is to bridgethe gap between the two in a technically soundand usable way. In the first step of thetransformation of a hypermedia document to apaper printout, all information which wouldnormally get lost is identified and encoded in theprintout as so-called PaperLinks. Thisaugmentation step is totally transparent for theuser and will be explained in section 3.

PaperLinks are realised as short numericalidentifiers (IDs) associated to objects exclusivelyavailable in hypermedia. The crucial part of theconcept is to provide a simple and usable way toinput such IDs into a mobile device in the courseof reading the printed material.

Hitz Plattner 105

To achieve this, several approaches exist,determining the final representation of thePaperLinks in the printout.

Key: + = pros; – = cons

p Direct input of a short number (eg 2–4 digits)(see also section 4.1)

+ No special hardware needed

+ Easy to understand and implement; reliable

– Requires manual user input over keyboardor similar

– Only valid in a restricted scope

p Input via barcode-scanner (pencil)

+ Fast and convenient

– Requires additional (relatively exotic) hardware

– Less reliable than direct input

p Input via voice recognition

+ Fast and convenient (numbers representa small, closed domain)

+ No exotic hardware needed

– Less reliable than direct input

– Implementation can be complex onmobile devices

3 ArchitectureThis description of the general architecture isbased on a prototypical implementation in anongoing national e-learning project [part of theinternational MOBIlearn initiative (2005)], wherethe content is organised in small building blockscalled ‘learning units’ (LUs). LUs are available forthree different platforms (PC, PDA andSmartphone). Such an LU is the starting point ofthe augmentation process shown in Figure 1.

Figure 1PaperLink processing

FOP Formatting Object Processor

LU Learning Unit

ML MOBILearn

URI Universal Resource Identifier

Xerces Apache XML parser

XSL Extensible Stylesheet Language

XSL–FO XSL-Formatting Objects

XSLT XSL Transformations

An LU is an XML document, which is the basicinput for the ‘Augmentor’ process, in which allobjects of interest (interactions, videos, images,external hyperlinks, links to glossary entries, etc)are identified and unique identifiers (IDs) aregenerated, constituting their correspondingPaperLinks. These IDs, in combination with anobject-discriminating Uniform Resource Identifier(URI), are stored in a database (contentmetadata). The IDs are embedded into the LUdocument as XML processing instructions (PI) –thus, the document remains valid with respect toa given Document Type Definition (DTD) or XMLSchema. Finally, the augmented document istransformed into a printable PDF document.

The resulting printout – which now contains aspecific representation of the PaperLinks(see Figure 2 opposite) – is ‘processed’ by theuser who has a device with an active‘ID–Listener’ nearby. The ID–Listener is aplatform-specific process that waits for the userto input PaperLinks, resolving the associatedobjects and triggering the appropriate renderingprocesses (eg video player, web browser).


Figure 2 PaperLink representation in printout

4 ID–Listeners implementedTwo ID–Listeners have been implemented in theproject so far and are presented in the followingsub-sections.

4.1 Touchscreen dialThis implementation of a direct input variant of anID–Listener is based on the Windows CE-drivenPocket PC platform. Whereas the standard inputmethods (transcriber, virtual keyboard, etc) arerather cumbersome in use and do require theuse of the stylus, PDAs also feature quite a large(240x320 pixel) touch-sensitive display whichmay also be operated just with fingers. In ourimplementation, the screen shows a numeric keypad with two additional buttons to cancel and toset bookmarks (Figure 3).

Figure 3 Touchscreen dial

For this ID–Listener, all PaperLinks in the printoutare represented by numbers of fixed length (fourdigits in our application) so the correspondingaction can be triggered without the need for anexplicit input sequence termination. Without anytraining, such an ID can be entered within asecond, so we consider it to be fast and reliable,but of course, the value domain is very limited(0000–9999). While it would be possible toexpand this limitation by adding more digits, thiswould affect every single input operation andsignificantly decrease the efficiency of thesystem. Rather, the scope of the IDs visible inthe printout is restricted to a single contentmodule which, in our case, has the approximatesize of a semester’s lecture course (15 x 45minutes). In this variant, the user has to selectthe module (thereby establishing the context forID resolution) before he or she can use thePaperLinks. This adds some set-up cost, but hasto be done only once in a learning session anddoes not hamper every single input operation.

At the moment, PDAs are not really widespread,but there is a strong indication that these deviceswill quickly be outnumbered by the upcomingSmartphones which it is believed will reachnearly total penetration within the next 3–5years. As Smartphones offer similar capabilitiesto PDAs and already feature numeric key pads,they would be ideal for this usage scenario.

4.2 Bluetooth barcode scannerOur second implementation of an ID–Listener isalso for the Pocket PC platform, but uses aBluetooth-enabled barcode scanner to ‘follow’the PaperLinks. The scanner is very lightweight,small in size and can be used like a pen to easilyscan the barcodes representing IDs in theaugmented printout. The wireless scanner andan example of a PaperLink can be seen inFigure 4 (overleaf).

Hitz Plattner 107

Figure 4 Barcode scanner plus PaperLink to an interactive experiment

With this ID–Listener, code length is not anissue, so there is no real problem in associatingevery desired object with a project global ID, andso the additional step of ‘establishing context’can be omitted. On the other hand, this inputmethod is not necessarily faster than thetouchscreen dial and the barcode scanner is quiterare and expensive. These are serious problemsfor e-learning platforms, but the scanner is a niceoption for scenarios in which the requirementsfor initial cost and availability are less oppressive.

5 Future workBecause of the early state of the overallMOBIlearn project, we only have very limiteddata regarding the usefulness and acceptance ofthe current PaperLink implementation. A smallreal-world test with about 30 users will beperformed in mid-July 2004. The results will beavailable from the authors.

The current prototypical implementation of theconcept is tailored to the MOBIlearn e-learningenvironment. If the usability tests yieldsatisfactory results, there are plans to do ageneric implementation which will support anyXML-structured content and, as far as possible,HTML-encoded documents too.

AcknowledgementsThe work presented in this paper has beensupported by the Neue Medien in der Lehre anUniversitäten und Fachhochschulen initiative ofthe Austrian Federal Ministry for Education,Science and Culture.

ReferencesGuimbretière F (2003). Paper augmented digitaldocuments. In Proceedings of the ACMSymposium on User Interface Software andTechnology, Vancouver, Canada. New York:ACM Press, 51–60.

Learn MI and Mirski PJ (2003). Mobile learning –cui bono? In M Auer (ed) Proceedings of theInternational Workshop on InteractiveComputer-aided Learning, September 2003.Kassel: Kassel University Press.

Mills CB and Weldon LJ (1987). Reading textfrom computer screens. ACM ComputingSurveys, 19(4), 329–358.

MOBIlearn (2005). Website.At www.mobilearn.at or www.mobilearn.org


AbstractThis paper introduces questions that areimportant when designing adaptive mobilelearning systems. Adaptivity is defined here asone form of adaptation; or as a quality of asystem to automatically and autonomicallyregulate and organise its functioning, theappearance of its UI and the order of informationit offers. The most important thing is to weigh theusefulness or limitations of adaptivity in learningsystems. Other essential questions includeissues of choice between different adaptivetechniques and choosing the foundation forthem – in this case, user modelling; and thechallenges of gathering information on learners.

Keywordsmobile learning, adaptivity, user modelling

1 Why or when is adaptivity neededin mobile learning systems?Learning through mobile devices with smallscreens requires more from the contentmanagement and from the user interface (UI)than is required in more traditional environments,since there is a large variety of terminals (morethan in the case of PCs, for example) and thefunctionality of networks (coverage andbandwidth) varies as well. Adaptivity can bedefined as one form of adaptation or as a qualityof a system to automatically and autonomicallyregulate or organise its functioning, theappearance of its UI and the order of informationthat it offers. Defined this way, it is close to theartificial intelligence (AI) research that has beendone since the 1950s (see eg Self 1988), whereproblems relating to knowledge-based systemsand inferencing are important themes [see alsoresearch on uncertainty in AI (UAI 2005)].

Adaptive mobile learning systems – the essential issues from the design perspective

Combining adaptivity and mobile learning meansfrom the point of view of learning that:

p the screen space may be very limited and thefunctions may be difficult to control

p the learning online may not be possible at alltimes or everywhere

p the learning may be restricted in terms ofavailability (ie because of the costs of datatransfer or the skills needed to use the devices).

Even if we think about mobility as does Cooper(2002) – ie as learning independent of a certainplace or time; learning in any place, sitting orstanding, but not necessarily as learning on thego – the devotion of time and conscious attentioncan likewise be very limited, and the context canaffect the learning in a number of ways. Theseare challenges posed by mobility to the design ofmobile learning systems that can be met – atleast in some ways – through applying adaptivityin the learning systems.

Furthermore, even though with mobile deviceslearners are able to share information andorganise tasks and other educational activities innew ways, all this is happening in existing socialenvironments, and with existing patterns ofinteraction, and in existing cultures. From thedesigner’s point of view, it is important toconsider mobile computers not only as ‘dialoguepartners’, but also to consider the changing rolesof learners and teachers and types of activitywhen we are designing mobile learning systems.The technology itself should adapt to the learningneeds (Milrad 2003). In this sense, adaptivity canbe seen as a general feature of technology, andnot only a property of a mobile (learning) system.

However, adaptivity is not necessarily onlyan asset: if improperly applied, it can diminishthe controllability and predictability of theapplication and its functions (Jameson 2001).

Jäppinen Ahonen Vainio Tanhua-Piiroinen 109

Anu JäppinenMikko Ahonen Teija VainioErika [email protected]@[email protected]@uta.fi

Hypermedia LaboratoryUniversity of TampereFI–33 014 Tampereen yliopistoFinlandwww.mobilearn.org

In the worst case scenario, adaptive changes canmake learning harder if the system works in anunstable way and users then need to becomeaware even of the actions that they areaccustomed to perform automatically (seeKramer, Norotha and Vergo 2000). In this respect,the biggest problem usually is that adaptivityis just an additional ‘feature’ that is built ontothe system later on, or a need that emergesduring the construction process. Instead,adaptivity-related research questions and relatedneeds should be integrated into the early phases,since ‘research methods should not be selecteduntil a researcher is clear about his/her researchgoals as well as the nature of the researchquestions to be addressed within a particularstudy’ (Reeves 2000). In our opinion, this appliesto adaptive learning system design as well.

Different theories of learning have effects on thedesign: they produce different kinds of solutionin the adaptive mobile learning system. Thesealternatives should be evaluated at the beginningof the design process. Reeves (1997) hasintroduced a model of the 14 pedagogicaldimensions for evaluating different formsof computer-based education (CBE). Thedimensions can be used also for comparingdifferent implementations of CBE. They include:pedagogical philosophy, teacher’s role, learnercontrol and user activity. A mobile learningsystem is one form of CBE and therefore theimportant decision is whether the learning isgoing to be learner-centred or controlled by ateacher – or possibly by the system. Anotherdimension worth considering is learner activity(or ‘user activity’ in Reeves’s model).

2 User modelling as a basisfor adaptivityMost adaptive techniques are based on usermodelling; Benyon and Murray (1993) even claimit has to be an integral part of any adaptivesystem. In the case of mobile learning, learnermodelling can be quite challenging because it caninclude so many different, very complicatedfactors. Factors related to the learners –preferences, interests, goals, cognitive orlearning styles or other traits – presentchallenges for modelling, not only because thereare very few (computer-applicable) methods ofidentifying them, but also because they can varyand change over time (Schwab and Kobsa 2002).Adaptivity can be linked to these factors inmany ways; for example, by saving thelearning histories of learners and using themto make assumptions on the learners’ progress,

the system could help a learner to find moreeasily the functions they customarily use, orlearning material they still have not found butothers have (see also Schwab and Kobsa 2000 ongroup-based modelling and Bayesian networks).

Context may seem at first sight an easy part ofmodelling, but it is not: contexts affect people’sactivities mainly through the process ofinterpreting the contexts (eg Berger andLuckmann 1991) and are therefore closely tied inwith factors related to learners. However,learning can be supported with adaptivity withouttoo much effort – for example, through orderinginformation about a) the locations of learningopportunities; or b) experts according to theirphysical proximity. On the other hand, factors likethe experience (or lack of it) of learners in using thesystem or in terms of learning content providemore interesting and more meaningful startingpoints for modelling the contexts. For instance, inadaptive mobile learning environments, learnerscan be offered information about other people byorganising it according to their availability forcooperation (students), or advice (experts,instructors), or according to their closeness in thephysical space.

Device-related factors refer here to the differencesbetween different kinds of mobile device and thevarying size of their displays, input and outputmethods and their limitations in presentinginformation. In mobile learning, it would beuseful for learners to know, for example, if somelearning material cannot be presented due to lackof graphical display; or if the loading of picturestakes a considerable amount of time due to lowbandwidth. These factors are so numerous thatthe first difficulty is choosing which of them arethe most important. Second, it is also difficult todecide how exactly they should be made use ofwhen designing adaptivity.

3 Methods of modelling the userApart from the methods for making assumptionsabout the devices used, the users and theircontext, user modelling requires two other kindsof method:

p one for acquiring information, or makingobservations and assumptions based on them,about the users (ie upward inference); for example,input automatically or by learners, or by askingthe learner to confirm automatic observations.

p one for making more assumptions about usersbased on the assumptions made by the firstmethod (ie downward inference) (Jameson 2001).Moreover, a method for validating the user modelwould be valuable (Dillenbourg and Self 1992).


Different methods have different benefits andrequirements for qualities of data. The particularmethod that is applied can depend, for example,on how well the users’ behaviour in the system isknown (Jameson 2001). In the case of mobilelearning, one could think that there is not enoughknowledge to model the learners’ behaviour inmobile learning environments, as they are sonew. However, there is a lot of knowledge aboutpeople’s behaviour in web-based learningenvironments that can be at least partly appliedto mobile environments (see Benyon and Murray1993). However, taking a decision between (a)content-based methods (ie categorising thecontent) an(d b) collaborative methods (ie basedon other people’s choices of the content) thatdepend on the number of learners can be morecrucial in mobile learning systems, since thecollaborative methods require lots of learners orlots of action in the system. This may be aproblem in mobile learning, at least whenstarting up a new system or introducingnew material.

4 ConclusionsThe challenges posed by mobility to the designof mobile learning systems can be met throughapplying adaptivity in the learning systems, but itis not easy, since adaptivity in mobile learningcan refer to a host of different features, methodsof organising, automatic functions, etc, thepurpose of which is to encode and translateinformation between ‘two interacting agents(eg a human and a computer) who possess anincomplete understanding of the other’sknowledge’ (Chignell and Hancock 1988).The essential issues concern mainly:

p the actual need for adaptivity

p the choice of theoretical background for learning

p the kind of data needed as the base foruser modelling

p advance knowledge of the behaviour of learnersin the system

p the suitability of the learning materialfor classification

p the assessed quantity of activity in the system.

AcknowledgementsWe would like to thank our partners in theMOBIlearn project for their valuable supportduring the project. We would also like to thankJani Nummela and Tere Vaden at the HypermediaLaboratory at the University of Tampere fortheir comments.

ReferencesBenyon DR and Murray DM (1993). Applyinguser modelling to human–computerinteraction design. AI Review, 6, 43–69. Atwww.dcs.napier.ac.uk/~dbenyon/AIReview.pdf

Berger PL and Luckmann T (1991). The socialconstruction of reality: a treatise of the sociologyof knowledge. Garden City, NY: Anchor Books.

Chignell MH and Hancock PA (1988). Intelligentinterfaces. In M Helander (ed.) Handbook ofHuman–Computer Interaction. Amsterdam:Elsevier Science Publishers BV.

Cooper G (2002). The mutable mobile: socialtheory in the wireless world. In B Brown,N Green and R Harper (eds) Wireless World.London: Springer–Verlag.

Dillenbourg P and Self J (1992). A framework forlearning modelling. Interactive LearningEnvironments, 2(2), 111–137.

Jameson A (2001). Designing systems that adaptto their users. Notes from tutorial held at theConference on Human Factors in ComputingSystems, Seattle, 31 March–5 April.

Kramer J, Norotha S and Vergo J (2000).A user-centered design approach topersonalization. Communications of ACM,43(8), 45–51.

Milrad M (2003). Mobile learning: challenges,perspectives, and reality. In K Nyíri (ed.) Mobilelearning. Essays on philosophy, psychology andeducation. Vienna: Passagen Verlag, 151–164.

Papert S (1980). Mindstorms: children,computers and powerful ideas. Brighton:Harvester Press.

Reeves T (1997). Evaluating what really mattersin computer-based education.At www.educationau.edu.au/archives/CP/reeves.htm

Reeves T (2000). Enhancing the worth ofinstructional technology research through ‘designexperiments’ and other development researchstrategies. In International Perspectives onInstructional Technology Research for the21st Century Symposium. New Orleans:American Educational Research Association.

Self JA (ed.) (1998). Artificial intelligence andhuman learning: intelligent computer-aidedinstruction. London: Chapman and Hall.

Schwab I and Kobsa A (2002). Adaptivity throughunobstrusive learning. Kunstliche Intelligenz,2002/03, 5–9.

UAI (2005). Uncertainty in artificial intelligencewebsite. At www.sis.pitt.edu/~dsl/UAI/

Jäppinen Ahonen Vainio Tanhua-Piiroinen 111

Building a wireless learning management system

Jeffrey Webb 113

Ann [email protected] of Huddersfield and CPCETLTIUQueensgateHuddersfield HD1 3DHUnited Kingdom

Keith [email protected] of Huddersfield and CPCETSEDUQueensgateHuddersfield HD1 3DHUnited Kingdom

AbstractFor over a year, the Consortium forPost-Compulsory Education and Trainingmanaged learning environment (MLE) project has been implementing a learning managementsystem (LMS) across the north of England.Learners and tutors in colleges often lack theequipment needed for successful e-learning. Ourproject has been involved with providing resourcesfor the sharing of materials, and the means touse them via a wireless LMS.The LMS includesmany useful tools and resources, and the wirelessLMS/MLE is intended to help overcome obstaclesto using IT and learning technology.

KeywordsLMS, wireless

BackgroundThe Consortium for Post-Compulsory Educationand Training delivers the same University ofHuddersfield teacher training qualificationsthrough partners in 36 different locations acrossthe north of England. The programmes requirethat all learners have a positive and equivalentexperience of e-learning but, despite the best efforts of programme managers, the actual patterns of access and use have shownsignificant (and unacceptable) diversity in studentexperience with web-based learning activities.

Definition of the problemReasons for the variation have included theavailability of hardware and the disposition oftutors (two things that have not been unrelated).Another difficulty has been the need to book ITrooms in colleges to use e-learning resources.This can often involve a fair bit of advanceplanning on the tutor’s behalf, and the lesson has to be tailored to fill the IT time booked.

On this basis, the Consortium is now providingwireless facilities for engagement with its managedlearning environment among all its partners.

The wireless networks are in the form of trolleyscontaining laptops that can be used in any classroomthat is within 50 metres of an internet connection,and this facility is designed to undermine barriersto use of web-based opportunities, (including a virtual learning environment and a digitalrepository). The critical issue now is the extent to which flexibility is related to utility.

Trials and training with the wireless LMSThe reluctance of tutors to use the LMS hasbeen passed on to their learners. Even thoughtraining has been provided for the resourcesavailable, often the information has not beenpassed on to learners. This creates difficultiesdisseminating important information andimplementing learning technology resources.

Recently, our training and awareness raising hasinvolved a wireless laptop trolley. Our studentshave expressed excitement and an eagerness to use the laptops. The wireless nature of thelaptops adds a novelty value that captures theirinterest. Learners gain hands-on experience of the LMS and learning resources using thewireless equipment. The focus is on discoverylearning rather than instruction.

First findingsSo far we have found from our feedback withtutors that they are positive about the wirelesslaptops. For them it will mean the chance to use the LMS and learning technology resourcesin any classroom. Often IT rooms are notconsidered suitable for teaching purposes.

Our students are very enthusiastic and impressedthat their colleges are receiving this sort ofequipment. It is important for our project that weencourage new teachers to use learning technologyin their teaching. For these student teachers, it is important that they can use this learningtechnology in situations that suit their teachingpurposes. We have found from our trial eventsthat these needs can by met by a wireless LMS.

AbstractThe MOBIlearn project aims to develop areusable architecture for delivering mobilelearning experiences. A key component of thisarchitecture is a context-awareness subsystemthat is intended to tailor the content and optionsmade available to a learner, depending on theircurrent situation, preferences and learninghistory. The context-awareness subsystem hasbeen developed alongside a hierarchical model ofcontext, and has been subjected to formativeevaluation. With reference to our context model,preliminary user trials and input from museumstaff, we describe the planned deployment ofthis system in an art museum learning scenario.

Keywordscontext awareness, mobile learning

1 Context awarenessfor m-learningContext-aware computing (for a recent review,see Chen and Kotz 2000) has a lot to offer mobilelearning. By taking account of the learner’ssurroundings, we can create engaging learningexperiences, providing content and options thatare tailored to the current context.

One of the aims of the MOBIlearn project is tocreate an architecture for mobile learning thatincludes context-aware recommendations ofcontent, options and services. Work so far hasproduced a hierarchical model of context(Lonsdale et al. 2004) and we have performedsome preliminary formative studies of thesystem (Beale and Lonsdale 2004).

Context awareness for MOBIlearn: creating anengaging learning experience in an art museum

Within the MOBIlearn project, we havedeveloped several learning scenarios designed topermit the deployment and testing of thesystem, including the context-awarenesssubsystem. Here we present our plans fordeploying and evaluating our context-awarenessarchitecture for students in an art museum. Thisresearch is work in progress, currently beingdeveloped and planned for deployment in a localmuseum in September 2004.

1.1 Museum scenarioIn our art museum scenario, two students visitthe museum with a number of goals. Our aimis to support them in carrying out thefollowing activities:

p planning the visit according to personal interestand requirements

p reaching the area of interest as quickly as possible

p knowing exactly where co-learners, guides, etc,are located in the museum

p viewing a catalogue of works by a particular artist

p receiving detailed information about the work

p receiving information about the artist

p receiving information on other artists belongingto the same period of history

p using an audio guide on the mobile device

p downloading and saving information relating tothe work of art

p adding comments/notes next to each work

p communicating with other learners andsharing opinions.

Lonsdale Baber Sharples Byrne Arvanitis Brundell Beale 115

Peter LonsdaleChris BaberMike SharplesWill ByrneTheodoros N ArvanitisPat [email protected] of EngineeringUniversity of BirminghamEdgbastonBirmingham B15 2TTUnited Kingdomwww.mobilearn.org

Russell [email protected] of Computer ScienceUniversity of BirminghamEdgbastonBirmingham B15 2TTUnited Kingdomwww.mobilearn.org

2 ImplementationThe context-awareness system is implementedas a web service in Java, running under a Tomcatserver. We have defined appropriate ExtensibleMarkup Language (XML) Schemas for sendingand receiving contextual information, and forproviding recommendations to the contextdelivery subsystems. This work has been drivenand informed by our original context model, webservices architecture standards and pragmaticconcerns for the MOBIlearn architecture as a whole.

2.1 Context modelThere is growing recognition of the need forflexible, scalable, reusable models of context thatcan be deployed for a range of applications. InMOBIlearn, we have been working to produce areusable model of context for mobile learningapplications (see Beale and Lonsdale 2004;Lonsdale et al. 2004). There have also been callsfor a reconceptualisation of context to representbetter the sociocultural aspects of activity,moving away from the predominantlytechnology-centred approaches to context(Lueg 2002a, 2002b; Dourish 2004).

For MOBIlearn, the purpose of contextawareness is to enable learning on mobiledevices, and so our approach to describingcontext and applying this description toproducing a usable software architecture is basedon this focus. Figure 1 shows the basic hierarchyfor our description of context.

Figure 1 Context model hierarchy

Instead of a rigid definition, our intention is toprovide a hierarchical description of context as adynamic process with historical dependencies.By this, we mean that context is a set ofchanging relationships that may be shaped by thehistory of those relationships. For example,learners visiting a museum for the second timecould have their content recommendationsinfluenced by their activities on a previous visit.More details of our context model can be foundin Beale and Lonsdale (2004).

Contextual information is made available to othercomponents of the MOBIlearn system by meansof XML documents in an agreed format. At anygiven time, the current context state isrepresented as a nested set of context features,all described in XML form. An XML Schema forthis XML object is an agreed format that allowsall components of the MOBIlearn architecture toaccess this information as and when it isrequired. Storage of a set of timestamped XMLcontext objects provides the historical contexttrace that can be inspected and used bysubsequent sessions.

2.2 ApproachMuch of the current work on context awarenessis technologically driven – that is, systems aredeveloped to take account of the capabilities ofthe available technology, but there is currently adebate about the exact nature of ‘context’ forcontext-aware computing (see eg Dourish 2004).Our aim, in MOBIlearn, is to produce alearner-centred approach to context awarenessby including contextual elements that are usableand useful for a given learning scenario. Theseelements also need to be available from either theenvironment, a learner model/profile, or directlyfrom the learners themselves. We aim to involvethe users in the process of determining and usingcontextual information by consulting them forinformation and by making their current derivedcontext both open for inspection and modifiable.

2.3 Elements of contextWhat is becoming clear is that there aredifficulties in implementing context awareness.First, how do we get hold of contextualinformation; and second, what do we do with itonce we have it?

Within the MOBIlearn project, we are centreingour designs on specific learning scenarios. Atpresent, we are working on deploying thesystem in an art museum setting.


For this scenario, the context-awareness systemis being set up to take account of severalcontextual elements from both the environmentand the learner. These contextual elements willbe used to derive a usable context substate(see Lonsdale et al. 2004). Elements to be usedin the context-awareness subsystem include:

p current learning topic: this will be eitherexplicitly indicated by the learner, or obtainedfrom his/her profile

p time spent on each artefact: our location-trackingsystem not only tracks learners’ locations, it alsotimes how long they spend in any one position.Standing in front of an artefact for a longer periodis used to infer a higher level of interest andcontent recommendations are updated accordingly

p artefacts or content annotated (through theMOBIlearn system on a handheld device): theMOBIlearn system supports the annotation ofcontent and also of artefacts themselves (a formof virtual graffiti) – this information will be trackedby the context-awareness system to providerelevant recommendations (eg recommendingitems that have been annotated by friends)

p content items shared with others (through theMOBIlearn system on a handheld device): theMOBIlearn system supports the sharing of contentwith other learners – this activity will be trackedand used as a source of contextual information.

These elements of context will interact to give usa useful way of determining what is appropriateto show the user. For example, the time spentlooking at each painting will be used to derive ameasure of ‘interest’. A low level of interest willmean that only the title and artist of the currentartefact will be shown. However, if the subject ofthe current artefact matches the user’s currentlearning goal, then the system will assume ahigher level of interest and show more detailedinformation. The user will also be able to overridethe system. For example, if their goals changeduring the session and this is not reflected bytheir current profile, then they will be able tospecify their interests, or revert to a default contentdelivery that is not customised in any way.

2.4 Issues raised by stakeholders –what do we know about people’sbehaviour in museums?

Museum curators

We have engaged with the stakeholders of themuseum experience by inviting input from thecurators of the museum in which we plan to deploythe system. Several characteristics of visitorbehaviour lend themselves to context-aware support.

p No follow-up, no introduction: visitors to amuseum typically have little or no lead up to theirvisit, and are subsequently unprepared for thematerials they encounter. Similarly, there is noopportunity for them to follow up their visit. Thevisit itself must be designed to accommodatethe lack of lead-in and lead-out activities.

p Lack of focused interest: museums contain manyitems, relating to a wide range of topics. Visitorsenter exhibition spaces and take in the collection,but often fail to give any specific attention toparticular items. Being able to point them in the rightdirection with context awareness will, it is hoped,be an effective way of focusing their attention.

p How can we present the right level ofinformation to each person? Museum visitorscome from many different backgrounds and havedifferent goals, interests and experience. It isimpossible to pitch the information provided atthe right level to satisfy all visitors. The key to thisproblem is personalising the information weprovide, using whatever contextual informationwe can gather about the visitor.

Issues from user trials

To understand better the impact of context-awarecontent delivery in an art museum setting, weran some user trials where learners were giventhe chance to receive recommendations fromthe context-awareness system.

The system provided recommendations ofcontent, questions and communication withother learners, depending on the participant’scurrent location and question (selected from apredefined list). For example, a learner standingin front of Botticelli’s painting La Primavera wouldsee content relevant to that painting near the topof their content list, with the top item being mostrelevant to La Primavera and their current question.If another participant who had already answeredthe current question was also standing in front ofLa Primavera, the system would suggest talkingto them in order to discuss the answer.

During the session, the experimenter employeda ‘Wizard of Oz’ evaluation method, monitoringeach participant’s location and updating theirclient software, using a remote managementapplication. (The Wizard of Oz evaluation methodinvolves the experimenter(s) acting to simulatesome or all functionalities of the system, with orwithout the participants’ knowledge, allowingevaluation before a fully functional systemis available.) The experimenter was also ableto monitor which question each participantwas currently working on and whichquestions had already been answered.

Lonsdale Baber Sharples Byrne Arvanitis Brundell Beale 117

At the end of the session, the concept ofcontext-aware content delivery was discussedwith the participants, and they were asked forfeedback about their use of the system in aninformal, free-form interview.

Feedback was gathered from users about theusefulness and usability of the system. Thisfeedback was used to derive a formativeevaluation of our current implementation. Thefollowing issues were identified.

p It worked: most users were able to findrelevant information quickly, and successfullyanswer the questions.

p Interface and representations: many users wereconfused about what we were trying torepresent with the interface, and were not surewhy their recommendations were changing orhow they could best use the recommendationslist to answer the questions.

p Understanding: some people were not quite surewhy the system did what it did, and weresurprised by the constantly changing list ofoptions. Demonstration and explanation did notseem to help with this – when there was amisunderstanding, it was due to a lack ofintuitiveness about the display of thecontext-dependent recommendations.

p Distraction versus engagement: offering multiplechoices either led to sidetracking or encouragedpeople to take their exploration of the contentfurther. Both of these outcomes suggest thatusers were engaging with the experience, butthis could become a concern if we are trying todesign a specific programme of learning. Optionsthat distract users from their current task focusneed to be avoided, and so it is possible thatsome limits need to be set on exactly how muchcontextually based recommendation is done.

p Mixed content: there is a need to distinguishquestions, content and physical resources.Offering recommendations on all of these in asingle integrated display seemed to beconfusing, especially in combination with the lackof an intuitive, easily grasped model of what wasactually going on and why.

p Temporal context: context is often used in asnapshot sense – ‘What is happening now?’,‘Where am I at this moment?’, and so on.However, there are many much longer-termaspects to context (eg task, learning progress,life goals) and it is not clear how best to representand use this information in the context system.The fundamental issue is that we need to be ableto model and then provide support for usersacross multiple activities, episodes and projects,with the history of previous support playing anintegral role in determining future actions.

3 Next stepsSeveral specific challenges for implementing thiskind of context-aware experience are apparent.

p How can the context-aware system besupportive in a non-intrusive way?

p How can we make suggestions about content oractivities without causing distraction?

p How can we effectively combine varioussources of contextual information to providerecommendations appropriate to the goals of:a) the learners; b) the experience designers;c) the system designers; andd) the teachers/docents?

The context-awareness system, along with theother major components of the MOBIlearnsystem, will be deployed in a local museum inSeptember 2004. There are also ongoing trials ofvarious context-sensing mechanisms at theUniversity of Birmingham.

ReferencesBeale R and Lonsdale P (2004). Mobile contextaware systems: the intelligence to support tasksand effectively utilise resources. In Proceedingsof the 6th International Conference onHuman–Computer Interaction with MobileDevices and Services, Glasgow,13–16 September, 240–251.

Chen G and Kotz D (2000). A survey ofcontext-aware mobile computing research.Technical Report TR2000–381. Hanover, NH:Dartmouth College.

Dourish P (2004). What we talk about when wetalk about context. Personal and UbiquitousComputing, 8(1), 19–30.

Lonsdale P, Baber C, Sharples M and Arvanitis T(2004). A context awareness architecture forfacilitating mobile learning. In J Attewell andC Savill-Smith (eds) Learning with mobiledevices: research and development – a bookof papers. London: Learning and SkillsDevelopment Agency.

Lueg C (2002a). Looking under the rug oncontext-aware artefacts and socially adepttechnologies. Paper presented at the Workshopon the Philosophy and Design of Socially AdeptTechnologies at the ACM SIGCHI Conference onHuman Factors in Computing Systems,20–25 April, Minneapolis.

Lueg C (2002b). Operationalizing context incontext-aware artefacts: benefits and pitfalls.Informing Science: Human Technology Interface,5(2), 43–47.


AbstractThe Mobile Technologies for Ad-hoc Learning(MoTFAL) project is a joint initiative ofpedagogical and cognitive science, involvingtechnological experts, educators andpsychologists, to research the possibilities ofusing mobile platforms with internet access foreducational purposes at school level. The projectdesigns, develops, tests and evaluates ahandheld learning environment based onemerging technology that facilitates on-sitelearning. The proposed approach cuts across thetraditional boundary between the classroom,home and other alternative educational settings(museums, libraries, archaeological and historicalsites, etc) as distinct learning environments. Thegoal is to shift away from classroom learning to‘daylong’ learning and to use the mobiletechnology to facilitate that shift. This paperpresents the basic principles of the MoTFALproject, the theoretical background of itsdevelopment as well as the possibilities for itsexploitation for educational purposes.

Keywordsm-learning, mobiles, learning anywhere anytime,handheld devices

1 IntroductionThe MoTFAL project is a joint initiative ofpedagogical and technological experts, educatorsand psychologists to research the possibilities ofusing mobile platforms with internet access foreducational purposes at school level. Thepartnership aims to develop, test and evaluatelearning schemes that are implemented within ahandheld learning environment, based onemerging technology that facilitates on-sitelearning, maximising the impact of informationthat is provided when the motivation of thestudent is highest.

The MOTFAL project: mobile technologies for ad hoc learningEleni MalliouStavros SavvasSofoklis [email protected] and Development DepartmentEllinogermaniki Agogi25D Plakentias AvenueGR–15234ChalandriGreece

Antonis MiliarakisManolis [email protected] and Development DepartmentFORTHnet SAScientific and Technology Park of CreteVassilika Vouton PO Box 2219GR–71003HeraklionCreteGreece

The MoTFAL learning environment includes fullaccess to digital resources, cognitive tools,knowledge visualisations and software mentorsto help with: learning to use devices such asdigital cameras; organising and recalling imagesand the sounds of people and situations;knowledge sharing between students in differentenvironments; using contextual personal toolsthat change their behaviour based on where theyare and the activity in progress.

1.1 On-site learning and learningthrough mobile technologyClassrooms, textbooks, lectures and trainingsessions have at least one thing in common.As is characteristic of traditional learningopportunities, they take the learners out of thecontext of their everyday tasks and otheractivities and put them into specialised learningcontexts. But there is another idea, one thatpromises to complement traditional dedicatedlearning situations with ‘contextual learning’,where learning is a dimension of those everydaytasks, activities and situations. This alternativeapproach is becoming all the more attractivein the light of current trends in work andlearning that emphasise continuous andjust-in-time learning.

Learning happens in various ways. Students learnin classrooms, but they also learn by exploringstreams and parks, trying and failing to performtasks, talking to friends, etc. Adults learn in manyof the same ways, by experience, by involvement,by talking with peers and experts, or by delvinginto a practical problem. Any experience can be alearning opportunity, but often the resources tomake it so are lacking. We are used to thinkingof knowledge as something ‘stored’, ‘held’,or contained in a ‘body of knowledge’.

Malliou Savvas Sotiriou Miliarakis Stratakis 119

We are following a different, complementaryinsight here, that knowledge is something activein situations and contextual in its very nature.Knowledge is something that happens ratherthan something that is stored and appliedwhen appropriate.

2 Educational point of view

2.1 Educational theoriesThe educational theories that underlie thedevelopment of the educational material as wellas the educational framework of the MoTFALproject are as follows.

p Collaborative learning: in collaborative learning,students generally work together in groups oftwo or more. The mobile application of theMoTFAL system enhances students’collaboration, as it gives them the possibility tocommunicate and cooperate just by using theproject’s platform.

p Contextual learning: contextual learning islearning that occurs in close relationship withactual experience. Its main principle is tomotivate students to make connections betweenknowledge and its application to their lives. In theframework of the MoTFAL project, students havethe opportunity to study their curricula inreal environments.

p Autonomous learning: autonomous andself-directed learning sees learners asresponsible owners and managers of their ownlearning process. The MoTFAL project supportsthe autonomy of the learners by providing themwith a variety of resources and materials, with anopen environment for learning and collaboration,and also with opportunities to work alone orwith others.

p Experiential learning: experiential learning takesplace when a person is actively involved in anactivity.The MoTFAL project is based on thepedagogical method of experiential learning as itgives students the opportunity to get personallyinvolved in the learning activity.

2.2 Educational scenarios for learningthrough mobile devices: teachinghistory – an exampleResearch on the methods of teaching history inthe classroom has shown that the main difficultiesthat students face relate to the understanding ofhistorical terms, the placement of historicalevents in space and time, and the propositionof multidimensional causative relationships.

They also face difficulties in understanding thathistorical studies are not political, military ordiplomatic events, but the everyday life of peopleof a particular epoch. Contextual learning inhistory education and the scenario-designmethod offer a different approach andperspective to history teaching and learning.

The educational approach which is adopted in theframework of the MoTFAL project is to usescenario-based design methods as a means ofdefining suitable educational applications of themobile technology. Scenario building is one ofthe main design techniques to explore newforms of interaction, in which the physicalenvironment is made to react to humanbehaviour, using handheld devices as a mediator.In the framework of the MoTFAL project, a seriesof scenarios as well as the relevant educationalmaterial are being designed and developed foruse during the implementation phase of theproject. An example of a learning scenario ispresented below.

The teacher of a high-school class takes thestudents on a field trip to the Parthenon inAthens. As they are visiting the monument (seeFigure 1 opposite), the students are asked toconnect to the specific area of the platformwhere the teacher has already uploaded theselected material concerning the monument’shistory. Students are able to see pictures of themonument during the time, enriched byanimations. They can also have access to a videoshowing what the monument looked like andhow it was related to the everyday life of thepeople at that time. Furthermore, students areable to capture moments of their visit with thecamera of the device and upload them to theserver for future reference, and also add theircomments and continue their research byaccessing relevant websites during their visit.

3 Technological point of viewA true m-learning environment resembles inprinciple a sophisticated content and datamanagement system, with development,delivery and control of the content and thelearning process. Based on this definition,m-learning should be described as part of anintegrated global learning strategy,encompassing a variety of instructional methods,management of learning content and servicesthat supply learners with electronic informationand educational content regardless of wherethey are or the time of day.


3.1 Technical descriptionThe main technological aim of the MoTFALproject is the development of a platform thatmakes internet services available in any placeand at any time. The MoTFAL system iscomposed of a handheld device, a GeneralPacket Radio Service (GPRS)-capable mobilephone and an open web platform. The openstructure of the platform gives teachers thepossibility of developing and uploading theteaching material that is useful for their lesson.

The project develops methods and software forstoring, retrieving and dynamically synthesisingeducational modules to meet each learner’sgoals. To achieve this, the content is broken intosmall, independent multimedia educationalmodules. These are stored and retrievedusing a database management system.Metadata is used to describe the modules.

Figure 1 A high-school visit to the Parthenon in Athens

Aspects that need to be described usingmetadata include the format of each module andother technological aspects, its technologyrequirements, its duration, its role in the learningcycle, etc. The platform services are delivered viaan advanced user interface (UI), where the userhas to log in. The first major component of the UIis the Personal Learner’s Apprentice: this is thecore software agent of the system and the mainpart of the UI, being responsible for interactingwith the user. It will:

p manage the user–system dialogue

p proactively suggest content to the student,based on his/her profile

p attend to users’ queries about content andsuggest modules that meet their declared needs,based on the available educational modules

p break down the user queries and goals intosub-goals to be met by the system and thencooperate with the Response Planner in orderto compile a list of suggestions

Malliou Savvas Sotiriou Miliarakis Stratakis 121

p monitor the correct delivery of courses andrecord the user’s learning behaviour in order toupdate his/her profile

p optimise the delivery of content with respectto momentary network availability anddevice capabilities.

A Multimedia Messaging Platform (MMP) isbeing developed that will provide two-waycommunication. The MMP is the second majorcomponent of the UI. A web-based applicationprovides the interface for the delivery of themultimedia messages. A web server is used tocollect user responses either through the Web ordirectly from the mobile network.

3.2 Usability issues and the project’stechnological requirementsLearning through mobile technology has beenslow to grow because most wireless devices todate have had small screens, low resolution,slow processing and limited storage capabilities.The solution lies in taking a different approach tostreamlining the information and targeting it atthe user.The first step towards this solution is tointegrate a user-centred investigation into them-learning system’s development cycle. Themethodological approach of the MoTFAL projectplays a fundamental role in the development ofsuch a system: user-centred design andscenario-based design are ways of assuring thatthe final system is appropriate to the user and tothe context of use.

The system of the MoTFAL project fulfils thefollowing requirements.

p Interactivity: it should provide a means ofcommunication between learners and teachers.

p Interdisciplinarity: content should be presented inan interdisciplinary way, incorporating informationfrom different disciplines, thus promoting theidea of informal learning.

p Unobtrusiveness: so that the student can capturesituations and retrieve knowledge withouttechnology obtruding on the situation.

p Availability: its functions should be availableanywhere and it should provide seamlesscommunication inside and outside buildings.

p Adaptability: it should adapt to the learners’evolving skills and knowledge.

p Usefulness: it should be suited to everydayneeds for communication, referenceand learning.

p Suitability: content should correspond to thespecific learning needs of users.

4 ConclusionsThe MoTFAL partnership considers that thechallenge for the future generation of educationalsystems at the dawn of the third millennium is todevelop didactic environments for mobile phonesand mobile devices as the availability of mobiledevices spreads to more than 1bn usersworldwide. The mobile telephone is becoming atrusted personal device with internet access,smart-card usage and a range of possibilities forkeeping the learner in touch with the institution’sstudent support services and in contact withlearning materials and fellow students, while athome, at work or travelling.

During the implementation of the project, anextensive usability evaluation will be performedthat will offer guidelines for the human–computerinteraction and psychological content required forthe development of the final version of theMoTFAL system. The MoTFAL consortium aimsto investigate the impact that handheldtechnology has on end users’ experience ofwireless m-learning applications.

ReferencesAnderson JR, Reder LM and Simon HA (1996).Situated learning and education. Educational Researcher, 25(4), 5–11.

Aviram A (1993). Autonomy and the flexible school. International Review of Education,39(5), 419–433.

Fleischman J (2001). Going mobile: newtechnologies in education, Converge Magazine,May issue. Available via www.convergemag.com

Norman DA (1999). The invisible computer.Cambridge, MA: MIT Press.

Ringland G (2001). Scenario planning: managing for the future. Chichester: John Wileyand Sons.


AbstractEnlarging the scope of interaction betweenuniversities, research institutions, networkequipment manufacturers and network operatorsto actors distributed in a worldwide environmentis crucial to the effective advance of the researchcommunity. This synergy puts new requirementson today’s access and backbone networks toprovide the different data traffic that arises fromactivities like e-learning and e-training with theadequate level of service.

The MOICANE project (Multiple OrganisationInterconnection for Collaborative AdvancedNetwork Experiments) aims to test theeffectiveness of the standardised solution forInternet Protocol Quality of Service (IP QoS) –namely IntServ and DiffServ – in a completeintegration of access and backbone parts. Theproject’s main goal was to create a distributedtestbed, interconnecting several remote network‘islands’ characterised by different accesstechnologies and supporting a range of services,such as e-learning, virtual classroom andvirtual laboratory.

To provide these QoS services, the MOICANEproject addressed and achieved three differentobjectives: networked cooperation for remoteexperiments; deployment of diverse accesstechnologies and analysis of their interaction withthe core network; set-up of autonomousQoS-capable islands and their interconnection.

Keywordse-learning, QoS, distributed network

Providing quality of service (QoS) guarantees for e-learning applications over IP networks: the MOICANE project

1 Introduction andproject objectivesEnlarging the scope of interaction betweenuniversities, research institutions, networkequipment manufacturers and network operatorsto actors distributed in a worldwide environmentis crucial to the effective advance of the researchcommunity. This synergy, based on a ‘virtuallaboratory’ service concept and resulting in a‘networked collaborations’ scenario, puts newrequirements on today’s networks. First, there isa strong need for a variety of flexible andpowerful access technologies (fixed, mobile,wireless) in order to extend the serviceavailability everywhere and at any time to a widerpool of actors. Second, the transport capabilitiesof the network backbone need to be enhanced inorder to provide the different traffic types with anadequate level of service.

The MOICANE project (Multiple OrganisationInterconnection for Collaborative AdvancedNetwork Experiments) (2005) is a 24-monthInformation Society Technologies (IST) initiativefunded under FP5 of the EU researchprogramme. In terms of technical detail, itimplemented a QoS-aware architecture based ona Differentiated Services (DiffServ) (IETF 1998)core network and an Integrated Services(IntServ) (IETF 1997) QoS architecture on theaccess part of the network. The definedarchitecture allowed the use of generic signallingprotocols, such as RSVP (ReSerVation Protocol),H.323, SIP (Session Initiation Protocol) on theaccess part of the network. Based on thisnetwork concept, network elements (borderrouters, core routers and bandwidth brokers)were independently developed by otherMOICANE partners (universities and researchinstitutions). In order to do this, the project basedits efforts on the work of the relevantstandardisation bodies and was influenced by thelessons learnt from commercial state-of-the-artnetwork equipment.

Mazza Roddolo Panza Balatti 123

Riccardo MazzaEnrico [email protected]@mail.wind.itWIND Telecomunicazioni SpAViadotto XXV Aprile 910015 Ivrea (Turin)Italy

Gianmarco PanzaEnrico [email protected]@cefriel.itCEFRIEL Network Systems UnitVia Fucini 220133 MilanItaly

The project developed two innovative prototypesof bandwidth broker. This enabled the MOICANEnetwork to provide both static and dynamicresource allocations between different domains.Additionally, the project defined an extension forthe COPS (Common Open Policy Service)protocol, called X–COPS, unifying several accessnetwork signalling protocols. The InternetEngineering Task Force (IETF) and other bodieshave not yet standardised an inter-domainsignalling protocol for communication betweenbandwidth brokers. The project proposed andimplemented an inter-domain signalling protocolbased on the X–COPS protocol. In this way, theproject unified the intra- and inter-domainsignalling between IntServ- and DiffServ-basedislands, providing a great contribution to theinternet community and at the same time, pavingthe way for the commercial equipment to follow.

Finally, the MOICANE pilot project showed thatQoS could be provided to end userstransparently, in a tailored and effective way,using the information carried by well-known andwidely used signalling protocols. Moving thenetwork intelligence towards the edge and usingstandard protocols for the network configurationallows a scenario where the customers can stilluse legacy networked applications and accessthe advanced services of a MOICANE-likeQoS-aware network.

2 Project testbed ande-learning serviceIn the MOICANE project testbed, composed ofdifferent ‘islands’ – each one independent fromthe others in terms of administration,geographical location (Athens, Bucharest, Lisbon,Ivrea, Milan and Pisa) and network architecturalchoices – the network QoS infrastructure neededto be tested first in a local environment, in orderto start evaluating locally the QoS model chosen.Each island was based on one or more DiffServdomains and included an integration of theDiffServ section with an IntServ domain. Afterthe islands had been set up and were correctlyoperating as autonomous entities, they wereinterconnected through high-speed connectionsor through the internet. Then their interactionwas tested in order to ensure that the QoSguarantees supported by each single island werepreserved when traversing one or more domains(in the case of pure DiffServ environments); andwhen passing from an IntServ to a DiffServisland (in the case of mixed environments).

DiffServ domains constituted the backbone ofthe testbed. Since there were cases whereDiffServ and IntServ domains would be presenton the same island, it was necessary to developand integrate devices able to incorporateefficiently the two QoS architectures.

Several effective steps led to the success of thisactivity. First, there was the deployment andpositive test, on each island, of all the modulesfor QoS provision. This was followed by anefficient dimensioning for the network elements,in terms of buffer, scheduling algorithms,bandwidth allocation and admission controlschemes. Hence, the verification of theend-to-end QoS guarantees was maintainedwhen crossing the different islands. Finally, therewas the consolidated knowledge and experiencein the set-up, configuration, traffic and networkengineering and protocol interoperability ofIETF’s IntServ and DiffServ models, thusexploiting on-site experiences to produce usefulinformation and management criteria suitable forsubmission to the relevant standards bodies(specifically IETF).

Furthermore, a novel testing methodology,namely Orione (developed by a collaborationbetween WIND and Tekelec), was adopted to runthe evaluation tests from a functional and aperformance point of view. These tests were runfor the access and core parts of the network, andfor each island separately and end-to-end, inorder to assess properly both the entire networkinfrastructure and the QoS perceived by theperson using the e-learning and ‘virtualclassroom’ services. A test scenario for thee-learning service, involving two differentMOICANE islands, is depicted in Figure 1(opposite). The students are located on island1,while on island2 there is the teacher, using his orher personalised application with enhancedcapabilities by comparison with the students’version. The two islands are connected by ahigh-capacity international link.


3 ConclusionThe MOICANE project has achieved threedifferent objectives: networked cooperation forremote experiments; deployment of diverseaccess technologies and analysis of theirinteraction with the core network; set-up ofautonomous QoS-capable islands and theirinterconnection.

The results collected in the testbed have fullyassessed the proposed architecture for e-learningand e-training services. The QoS provided fulfilsthe requirements of the end user. The developedsolution paves the way for a worldwidedeployment of virtual collaborative services.

ReferencesIETF (1997). The use of RSVP with IETFintegrated services. RFC 2210. Reston, VA:Internet Engineering Task Force.

IETF (1998). An architecture for differentiatedservices. RFC 2475. Reston, VA: InternetEngineering Task Force.

MOICANE (2005). Website.At www.moicane.com/

Mazza Roddolo Panza Balatti 125

Figure 1 E-learning service scenario with QoS guarantees

AbstractThis presentation will explain the concept oflearning objects and metadata and introduce theeduSource suite of tools, applicable for mobilelearning. Learning objects are enabled and madeinteroperable using mobile and other devicesthrough the use of international standards andspecifications. eduSource Canada is apan-Canadian project which aims to develop theopen source infrastructure for a network oflearning object repositories. The suite includesmetadata guidelines, applications, and hardwarespecifications. Metadata is essential forpromoting semantic interoperability, so theproject team has applied the IEEE LOM, creatingCanCore, which is a metadata implementationprofile which facilitates employment of the LOM(Berners-Lee, Hendler and Lassila, 2001;Geroimenko and Chen, 2002).

KeywordsLearning object, metadata, application profile,interoperability

1 BackgroundeduSource Canada partners are building anational test bed of linked and interoperablelearning object repositories. This project is basedon national and international standards. It isbilingual (French/English) and accessible toeveryone on the Internet, including those withdisabilities. eduSource is providing leadership inthe ongoing development of the associatedtools, systems, protocols and practices thatwill support such an infrastructure. Collectively,the contributions of the partners amountto $5,280,000 of the total project value of$9,830,000. CANARIE is contributing up to$4,500,000. The six primary partners areAthabasca University – Canada’s OpenUniversity, the Netera Alliance, TeleEducation NB,Simon Fraser University – Surrey,Téléuniversitédu Québec, and the University of Waterloo.

eduSource Canada: learning object repositories and m-learningRory [email protected] University1 University DriveAthabascaAlbertaCanada T9S 3A3

2 Learning objects and metadata

2.1 A description of learning objectsLearning objects have been defined differentlyby various researchers. Although there is a widedivergence of opinion, at its core, a learningobject is a digital resource that is used inlearning. More specifically, the resource wouldhave a lesson encapsulated in it and appropriatemetadata. There are different levels of granularityused, ranging from a simple component to alesson, unit, module, course, or evenprogramme. These learning objects enablelearning when they are interoperable amongdifferent systems (IMS Learning designspecification, 2003) (Norm Friesen, 2001)(R. McGreal, in press; Wiley, 2001).

2.2 What is metadata?Metadata is often described as being ‘data aboutdata’. This description is not particularly usefulwithout examples. A library card is a commonlyknown type of metadata. The author, title andISBN code are all fields in this standard metadataformat. The principal metadata standard for LOsis the IEEE LTSC LOM. Many organizations suchas ADL, ARIADNE, and IMS (ADL, 2003;ARIADNE, 2003; IMS, 2003) have been involvedeither directly or indirectly in the development ofthe specifications on which this standard isbased (Duval, Hodgins, Sutton and Weibel, 2002;IEEE Learning Technology Standards Committee,2002; Rory McGreal and Roberts, 2001).A variety of application profiles for theimplementation of these standards, such asSCORM, CanCore and SingCore have beendeveloped (ADL, 2003; Athabasca University,2003; SingCORE, 2003).

McGreal 127

2.3 CanCore metadataimplementation profileThe IEEE LOM standard is both complex andgeneral in character. It contains a broad range ofelements, and leaves open many possibilities forinterpretation. CanCore seeks to simplify,interpret and refine this standard to aidimplementers. CanCore provides a great deal offine-grained information about each element inthe LOM (information that takes the form ofrecommendations, examples, and references toother interpretations). The CanCore elementsubset, as well as its best practice guidelines,has been created through close consultation witha community of educational participants (Frenchand English). CanCore has been specificallycreated to facilitate profiling and decision makingby those involved in implementation(Athabasca University, 2003; N Friesen,Roberts and Fisher, 2002).

2.4 ConclusionThe eduSource project supports the creation ofinternationally interoperable repositories forlearning objects. These repositories willempower m-learning. As course developers andteachers gain experience in the use of mobiledevices and as the number of online coursesgrows, the importance and necessity of learningobjects and the metadata standards that supportthem will become more apparent. Efficientlearning using the ever-expanding multimediaresources of the Internet on the rapidlyimproving mobile devices will become a reality.From anything and everything to specific digitallearning resources, the future of m-learning isinextricably linked with the development ofquality learning objects.

ReferencesADL (2003, 15 February). Sharable ContentObject Reference Model (SCORM). AccessedJuly 14, 2002 from www.adlnet.org/ADLDOCS/Documents/SCORM_1.2_ConformanceReq.pdf

ARIADNE (2003). Home page. Accessed July 14,2002 from http://ariadne.unil.ch

Athabasca University. (2003, 23 April). CanCoreMetadata Guidelines Home Page. AccessedJuly 14, 2002 from www.cancore.ca

Berners-Lee T, Hendler J, and Lassila O (2001).The semantic Web [Electronic Version]. ScientificAmerican, May. Accessed July 14, 2002 fromwww.sciam.com/article.cfm?articleID=00048144-10D2-1C70-84A9809EC588EF21

Duval E, Hodgins W, Sutton S and Weibel S L.(2002). Metadata principles and practicalities.D–Lib Magazine, 8(4). Accessed July 14, 2002from www.dlib.org/dlib/april02/weibel/04weibel.html

Friesen N (2001). What are educational objects?Interactive Learning Environments, 9(3).Accessed July 14, 2002 from fromwww.careo.org/documents/objects.html

Friesen N, Roberts A, and Fisher S. (2002).CanCore: Metadata for learning objects.Canadian Journal of Learning and Technology,28(3), 43–53.

Geroimenko Vand Chen C (Eds.). (2002).Visualizing the semantic web: XML-basedInternet and information visualization. London:Springer–Verlag.

IEEE Learning Technology Standards Committee(2002). Draft standard for learning objectmetadata. Accessed July 14, 2002 fromhttp://ltsc.ieee.org/doc/wg12/LOM_WD6_4.pdf

IMS (2003). IMS Home page. Accessed July 14,2002 from www.imsproject.org

IMS Learning design specification (2003).Accessed July 14, 2002 from http://imsglobal.org/learningdesign/index.cfm

McGreal R (Ed.). (in press). OnlineEducation Using Learning Objects. London:Routledge/Falmer.

McGreal R and Roberts, T (2001, October).A primer on metadata for learning objects:Fostering an interoperable environment.E-Learning, 2(10). Accessed July 14, 2002, from http://elearningmag.com/ltimagazine/article/articleDetail.jsp?id=2031

SingCORE (2003). Home Page. AccessedJuly 14, 2002 from www.ecc.org.sg/eLearn/MetaData/SingCORE/index.jsp

Wiley D A (2001). The Instructional Use ofLearning Object. Accessed July 14, 2002 fromhttp://reusability.org/read/


AbstractThis paper aims to describe how to maketeaching a large class in higher education livelierand more participative by using wireless devices.The paper briefly describes the challenge ofteaching a large class and presents the DRIM–APproject (French acronym for Multiple InteractiveRadio Devices and Participative Lecture Theatres)at the Ecole Centrale de Lyon (Graduate Schoolof Lyon).

Keywordsassessment, interactivity, large class,virtual classroom

1 Context and goalsLecturing to a large class is widely used as ateaching method in higher education, but is alsoknown to present some difficulties (Gibbsand Jenkins 1992) – for example, lack ofteacher–student exchanges, limited feedback,unmotivated students. Indeed, teachers are oftenchallenged by having to perform a lecture and atthe same time, wondering about the students’level of understanding. The teacher needs todevote a part of his or her attention andconcentration to analysing the students’behaviour and encouraging them to participate.On the student side, lectures are sometimesperceived as boring and participation iscomplicated by shyness and fear of lookingridiculous in front of teachers or fellow students.

Over the past few years, e-learning has hugelyincreased the use of technologies in theeducational arena, providing software solutionsto manage remotely located students throughthe use of learning platforms or virtualclassrooms. At the same time, very little hasbeen done really to integrate the use oftechnologies into face-to-face lectures.

Interactivity in a large class using wireless devices

But since the rise of wireless networks andmobile devices, all the components are nowavailable to build solutions that could helpteachers to increase participation and get morefeedback from students in lectures presented toa large class (VanDeGrift et al. 2002; Truong et al.2002). This is the topic of our research programmeat the ICTT research laboratory at the EcoleCentrale de Lyon where we are working on theDRIM–AP project (French acronym for MultipleInteractive Radio Devices and ParticipativeLecture Theatres; see David et al. 2004), the aimof which is to provide a software solution to aidteacher–student interactivity, based on the useof wireless devices and wireless (Wi-Fi)local networks.

2 The principle of an interactivelarge class using wireless devicesMobile campus and student laptop equipmentpolicies actually provide a favourable backgroundfor new wireless applications (localisation,IP phoning, etc). In teaching activities, likelarge-scale lectures, students can use theirwireless devices to communicate in real timewith the teacher, send him or her feedback, askquestions or answer in polls. The teacher projectsthe lecture slides as usual, but is also able tomonitor the class with the same laptop computerin order to identify quickly the problems that thestudents are encountering. This allows theteacher to adapt the lecture; for example,giving further explanation about a conceptmisunderstood by the class. Figure 1 (overleaf)shows an overview of the interactive large classbased on a wireless network.

Mercier David Chalon Berthet 129

Fabrice MercierBertrand DavidRené ChalonJean-Pierre [email protected]@[email protected]@ec-lyon.fr

Laboratoire ICTTEcole Centrale de Lyon36 avenue Guy de Collongue69134 ECULLY cedexFrance

3 DRIM–AP projectThe DRIM–AP project is managed by ICTT, amultidisciplinary research laboratory working onthe human and software aspects of cooperativeinteraction at the Ecole Centrale de Lyon. TheDRIM–AP project is part of a Mobile Campusresearch programme, and is funded by an HPMobile Technologies for Teaching Grant. Our aim isto design a tool based on interactive cooperationand wireless technologies for use in a face-to-faceapproach to teaching and to evaluate the usabilityof this kind of tool for students as well as forteachers. After a state-of-the-art overviewrealised last year, we have developed a softwareapplication and have also written an evaluationprotocol in order to validate our work.

4 FunctionalitiesOn the functional level, our solution bringstogether the main up-to-date techniques torealise interactive tests, students’ feedback,questions to the teacher, notes. We have addedcomplementary features improving monitoringtools for the teacher, slide display control andevent tracking. With DRIM–AP, students can givefeedback to the teacher about the lecture, ontopics such as pace (too slow, too fast),understanding, sound and visual ambiance. Theycan also write and submit questions or remarks,take notes, give answers to full assessments orquick polls. On his/her side, the teacher managesthe interactive class, submits tests and polls,receives and reads students’ messages on hisor her laptop computer. All events (question,feedback, slide display) are tracked by DRIM–AP.

The lecture can then be replayed by the teacherin order to identify where the main problems (egunderstanding, lecture speed) occurred.

5 Technical aspectsThe DRIM–AP application software is based onthe Java 2 Platform, Enterprise Edition (J2EE)and is hosted by an open source applicationserver. A standard web-based thin client is usedon purpose to make the deployment easier, butapplets can be used as well when needed. Sothe application does not need to be installed onthe user device, but can be directly accessed ona web server.

A web-based collaborative portal framework formsthe core of the DRIM–AP software. It was built tosupport multiple wireless devices (eg PDA,laptop, Tablet PC, Smartphones). The user devicetype is detected at its connection to the webserver and the user interface (UI) is dynamicallybuilt with portlet components (JCP 2003) that arereorganised according to the screen resolutionand display capabilities of the mobile device; forexample, HyperText Markup Language (HTML)or Wireless Application Protocol (WAP).In terms of data representation, DRIM–APintegrates Extensible Markup Language (XML)specifications based on IMS Learning Design(IMS–LD; see IMS 2003) to describe the lectureactivities and the learning objects (chapters,slides, tests, support activities). IMS Questionand Test Interoperability (IMS–QTI; see IMS 2000)is also used for the XML description ofassessments so that an authoring tool can beused to create the tests and the questions.


Figure 1 Principle of an interactive large class using wireless devices

6 EvaluationOne major purpose of our study is to evaluate theusability of wireless interactivity in a large classfrom the teacher’s and the students’ point ofview. For the teacher, we wish to establish thatthe monitoring activity and all the informationdisplayed on the screen are not creating acognitive overload that could disturb him or herduring the lecture. For the student, it seemsimportant to know if this approach is suitable forlearning in all knowledge domains or only forsome. The project’s main goal is to establishwhether the use of wireless technologiescontributes to improve the learning process inlectures presented to a large class, or if it is justanother ‘technological gadget’.

7 Conclusion and future worksFrom a technical point of view, we have proved inour work that existing wireless technologies aresuitable to bring interactivity into lecturespresented to large classes. We now have toevaluate the cognitive impact and usability ofsuch a system during teaching activities. Toachieve this, an evaluation programme withteachers and students of the Ecole Centrale deLyon will start in the autumn of 2004. The nextstep will be to release a version of the DRIM–APsoftware that we would like to make available tothe education community.

ReferencesDavid B, Mercier F, Chalon R, Delotte O andBerthet JP (2004). AMPHI INTERACTIF: état del’art et projet DRIM–AP. In Actes des PremièresJournées Francophones: Mobilité et Ubiquité,1–3 juin, Nice–Sophia–Antipolis, France.NewYork: ACM, 21–24.

Gibbs G and Jenkins A (1992). Teaching in largeclasses in higher education, how to maintainquality with reduced resources. London:Kogan Page.

IMS (2000). IMS Question and Test Interoperability: best practice andimplementation guide. Final specificationversion 1.01. Burlington, MA: IMS GlobalLearning Consortium.

IMS (2003). IMS Learning Design: best practiceand implementation guide. Version 1.0 finalspecification. Burlington, MA: IMS GlobalLearning Consortium

JCP (2003). JSR 168: portlet specification.At www.jcp.org/jsr/detail/168.jsp

Truong TM, Griswold WG, Ratto M and Star SL(2002). The ActiveClass project: experiments inencouraging classroom participation. TechnicalReport CS2002–0715. San Diego: Universityof California at San Diego.

VanDeGrift T, Wolfman SA, Yasuhara K andAnderson RJ (2002). Promoting interaction inlarge classes with a computer-mediated feedbacksystem. Technical Report 02–12–02. Seattle:Computer Science and Engineering Department,University of Washington.

Mercier David Chalon Berthet 131

Figure 2 Screenshots of the DRIM–AP user interface (UI) on several devices

AbstractThis paper explores the potential of an m-learningenvironment by introducing the concept of mLab,a remote laboratory environment accessiblethrough the use of handheld devices.

We are aiming to enhance the existing e-learningplatform and internet-assisted laboratorysettings, where students are offered in-depthtutoring, by providing compact tuition and toolsfor controlling simulations that are made availableto learners via handheld devices. In this way,students are empowered by having access totheir simulations from any place and at any time.

Keywordsblended learning, m-learning, remote laboratory

1 IntroductionThe recognition of the role that mobile devicesare likely to play in the future of learning (Keegan2002) has led us to seek for learning materialsand feasible applications for mobile devices inhigher education. To the best of our knowledge,there has not been any similar initiative, offeringremote laboratory access over the mobile device.

The starting point for this project was the GuidedWave Theory engineering course at our university,which consists of learning sessions that involve ablend of technologies. The course content materialis presented via traditional classroom lecturesand an interactive online learning environment.Additionally, a vital part of this lecture course isthe remote, online laboratory setting (iSign), aself-learning environment where students canpractise transferring theoretical knowledge intopractical experiments (Christ 2002) in the area ofelectromagnetic fields calculation.

Since the launch of the iSign online laboratory inSeptember 2001, experience has shown that thestudents are very motivated to use this blendedapproach that allows them to organise and pacetheir learning in the way that best suits theirpreferences and goals.

mLab: handheld assisted laboratoryJelena MiticMarkus FeisstAndreas [email protected]@[email protected] of Applied Sciences OffenburgBadstrasse 24OffenburgGermanyhttp://isign.fh-offenburg.de

The use of mobile devices within iSign startedwith Short Message System (SMS) notificationon the successful completion of the simulation.During the last few years, we have witnessed therapid development of wireless technologies andmobile devices, as well as the release of newand powerful Java technology for creatingapplications for mobile devices – Java 2 Platform,Micro Edition (J2ME). All this made the conceptof the handheld-assisted laboratory model afeasible project.

2 mLab: the handheld-assistedlaboratory modelThe input and display capabilities of wirelessdevices are cumbersome and limited.Consequently, the learning content for thosedevices needs to be carefully designed andselected. Furthermore, these very restrictionscan be exploited to give the learner easier accessto the most important and relevant learningmaterial (Li 2003).

In our project, learning content within the mLabapplication is designed to be a brief summary ofthe classroom course with a focus on necessaryinstructions and formulae to accomplish thelaboratory tasks. In addition, the students areable to test their knowledge with themultiple-choice questionnaire.

Using a wireless device, the student can log in tothe remote server where the simulation is carriedout, and start and stop the simulation. Thesimulation run can take anything from severalminutes to several hours. During this time, thestudent is able to check the progress of thesimulation via the mobile device. Finally, thestudent is provided with the evaluation of his orher results, where they are compared with thereference results set by the teacher.

Mitic Feisst Christ 133

The overall structure is presented in Figure 1.The web- and mobile-based applications arecompletely developed in Java. Use of the sametechnology, both on the client and the serverside, simplifies infrastructure integration andallows the use of existing knowledge, tools andcode through the network (Mitic 2003).

The iSign platform is the application server thatprovides access for different types of client throughdifferent connectors. It connects to the databasewhere all user settings and simulation data arestored. The simulation server deploys the actualsoftware for the electromagnetic fields calculation.

The mLab application is not a stand-aloneapplication; it provides a limited learning materialand simulation functionality compared to a webapplication. Therefore, the use of a web-basedand mobile-based laboratory is seen to becomplementary, in the way that it providesunconstrained and continuous access to thelaboratory settings.

3 Current state of play and further workIn the current stage of the project, we areimplementing the application to be used bystudents on the Guided Wave Theory course inorder to evaluate their motivation andexpectation when using the mLab platform. Theapplication has been tested on different devicesto assure portability and the same ‘look and feel’.The plan is to make it available to students fromthe winter semester of 2004. We expect to getvaluable knowledge that will help us to improvethe mobile learning content and the studentexperience. The goal is to offer students anefficient and innovative laboratory environmentthat is accessible at any time and from any place.

ReferencesChrist A (2002). Client-server architecture for[the] Active-Online-Learning Laboratory. Paperpresented at the Online Educa 8th InternationalConference on Technology Supported Learningand Training, Berlin.

Keegan D (2002). The future of learning: fromeLearning to mLearning. ZIFF Papiere 119. Hagen:FernUniversität. At http://learning.ericsson.net/mlearning2/project_one/book.html, accessed 13 September 2005.

Li M (2003). Practice related e-learning – theVIP framework. In Proceedings of the 9thInternational Conference of European UniversityInformation Systems, Amsterdam, 3–4 July.

Mitic J (2003). Developing wireless applicationfor simulation control. Master’s thesis: Universityof Applied Sciences Offenburg.


Figure 1 Overall architecture of the web-based and mobile learning platform

AbstractThe delivery of learning material on mobiledevices, termed m-learning, has aroused greatinterest in the research community of late. Thispaper describes a collaboration between theCLEV–R and ABITS projects to provide ane-learning system that can be adapted for use onmobile platforms. The Collaborative LearningEnvironment with Virtual Reality (CLEV–R) is aweb-based multi-user environment thatenhances a student’s learning experiencethrough the use of Virtual Reality (VR) andmultimedia. It provides a social setting forcollaborative learning where users, representedby avatars, take on roles in the virtual world. TheAgent-Based Intelligent Tutoring System (ABITS)enhances existing online tutoring mechanismswithin a higher education institution (HEI) bydelivering course material tailored to a user’spreferences, knowledge and learning style. Thispaper gives an outline of the e-learningenvironments and focuses on the methods to beused to present it on a mobile platform.

Keywordse-learning, m-learning, intelligent agents,virtual campus

1 IntroductionThe use of handheld computers such as personaldigital assistants (PDAs) has increased in recentyears in a wide variety of contexts. This increasehas been fuelled by the convenience of mobiledevices along with improved computing powerand affordability. The use of PDAs enables peopleto perform tasks while on the move, making thePDA the ideal travel companion. One rapidlydeveloping area in research at present is m-learning,where traditional e-learning systems are extendedto mobile platforms. E-learning can provide thosewho cannot attend a traditional university with awide variety of learning resources and expertisethat would otherwise be unavailable to them.

Collaborative m-learning using agents and virtual reality

These services provided on a mobile platformtherefore allow people to learn and accesscourse material ‘anytime, anywhere’. M-learningcan be used for distance learning, but can also beused to accompany traditional teaching methodswithin universities.

The Collaborative Learning Environment withVirtual Reality (CLEV–R) project, (McArdle et al.)and the Agent-Based Intelligent Tutoring System(ABITS) (Roche and Mangina 2003) both providea system for online learning. At present, theyhave been developed for use within UniversityCollege Dublin (UCD) to deliver online coursematerial to the undergraduate studentpopulation. ABITS utilises agent-basedtechnology to assist students and provide apersonalised tutoring system based on astudent’s personal needs, preferences andlearning style. CLEV–R provides a new way forstudents to learn, collaborate and socialiseonline. The use of VR in the educational domain,bringing a social aspect to learning, is a novelidea in the area of e-learning.Together, CLEV–Rand ABITS will provide a personalised VR learningenvironment augmented and managed byintelligent agents. Students, tutors and coursesare represented using the belief–desire–intention(BDI) agent paradigm in ABITS; and coursecontent is delivered using the VR environment ofCLEV–R. This paper describes the collaborationbetween CLEV–R and ABITS, with a particularfocus on their future extension for use on mobiledevices such as PDAs.

The rest of this paper is organised as follows. Insection 2, an overview of both ABITS and CLEV–Ris provided. Section 3 gives details of how thecombined system can be adapted for use onmobile devices, while section 4 discussesfurther development and gives some results ofinitial testing.

Monahan McArdle Kilbride Mangina Bertolotto 135

Teresa MonahanGavin McArdleJohn KilbrideEleni ManginaMichela [email protected]@[email protected]@[email protected]

School of Computer Science and InformaticsUniversity College DublinBelfieldDublin 4Republic of Ireland

2 Overview

2.1 CLEV–RThe CLEV–R system is a web-based multi-user VRenvironment that supports interactive e-learningfor university students through the use of severaldifferent types of multimedia. In addition toproviding an immersive environment whereusers are represented by avatars (3Name3D1997), active learning is supported throughintuitive interaction with tutors and fellowstudents. Natural methods of communicationsuch as voice and text chat are available andusers may also direct their avatars to makegestures. Synchronous learning is available in theform of lectures, tutorials and meetings wheremultiple users are connected simultaneously.Self-paced asynchronous learning is alsoprovided and consists of students downloadingand reviewing course material. The synchronousaspect lends itself very easily to studentcollaboration on group projects. The VRenvironment mimics a university setting, offeringmany of the services available on a real collegecampus. For example, lecture rooms and groupmeeting rooms are provided to assist thestudents’ learning. These rooms are augmentedwith shared whiteboards and presentationboards for delivery of learning content.Multimedia facilities, such as audio and video,can also be presented in these rooms if required.Tackling the social needs of students is one ofthe primary objectives of CLEV–R. To achievethis, informal settings have been developedwhere students can discuss course content. Thisremoves the feeling of isolation, which has oftenbeen an issue in conventional online learningapplications. Figure 1 depicts the CLEV–R userinterface (UI).

Figure 1The CLEV–R user interface

2.2 ABITSABITS attempts to supplement the existingclassroom and lecture-based teaching systemused within UCD, while also providing the abilityto run courses solely in an online environment.ABITS can be used to deliver courses online,including assignments, assessments and tests.A multi-agent system (MAS), based on the BDImodel, provides the ability to reason aboutstudent learning abilities and goals. An intelligentagent’s beliefs represent its current knowledge,which includes any type of information about itsenvironment at a particular time. This informationincludes details of courses and materials studiedby the students, supplemented with details oflearning styles and preferences for media typeand content granularity. The content of its beliefset guides the agent’s behaviour and therefore theinformation selected for presentation to students.Figure 2 depicts the interface for ABITS.

Figure 2 The ABITS interface

3 The mobile applicationSince m-learning has been receiving muchattention in the research community recently, theextension of both CLEV–R and ABITS for use onmobile devices seems like a natural progression.It is beneficial for students to be able to learnwhile on the move. Due to both the limitedmemory and screen size of mobile devices,content will be altered to ensure it is delivered inan efficient manner to the users. The GraphicalUser Interface (GUI) will also need to be adapted.

A much simpler view of the GUI will be presentedto a mobile user.The virtual campus is beingdeveloped using Virtual Reality Modelling Language(VRML 1997); support for this is presentlyavailable for PDAs in the form of Pocket Cortonaby ParallelGraphics (ParallelGraphics 2004).


This is a VRML plug-in that works in conjunctionwith Pocket Internet Explorer. It proves quiteeffective for displaying small files (up to 20kb);however, larger files are very slow both todownload and render. Consequently, we intendto reduce the amount of the VRML world thatwill be displayed on the mobile devices. Manydifferent techniques are currently being exploredto achieve this – techniques include cullingalgorithms such as view frustum culling, backfaceculling and occlusion culling (Martens 2003) andLevels of Detail (LOD; see Ames, Nadeau andMoreland 1996). These methods limit the amountof the VRML world that is rendered and also thenumber of animations that occur, and so improvethe overall efficiency of the system.

ABITS is being developed to include an intelligentuser interface (UI) which can adapt, based on anumber of criteria including available bandwidth,device type, user preferences, etc. Thisadaptation is achieved by exploiting technologiessuch as Extensible Markup Language (XML),Extensible Stylesheet Language Transformations(XSLT) and JavaServer Pages (JSPs) to separatecontent and deal with layout concerns. Onmobile devices such as PDAs, restrictions areenforced so that ABITS makes available onlythe content and options that can be displayedand performed given resolution andprocessing restrictions.

4 Results and further developmentAn initial implementation of CLEV–R has beendeveloped for desktop computers. A VRuniversity setting is provided and has beennetworked so that multiple users can connect tothe system. A GUI has been designed, andusers, represented by avatars, can browsepresentation and notice boards. Initial feedbackfrom a set of test users has proved encouraging.An initial prototype of ABITS has also beendeveloped and testing is due to commence in thenear future. At present, the VR component ofCLEV–R has been successfully tested on PDAsand further work is now being carried out toimprove its display and performance. An interfacefor the ABITS system has also been developedfor mobile devices.

ReferencesAmes AL, Nadeau DR and Moreland JL (1996).Vrml 2.0 sourcebook. New York: John Wileyand Sons.

Martens R (2003). Occlusion culling for thereal-time display of complex 3D models.Master’s thesis: Transnational UniversiteitLimburg, Germany.

McArdle G, Monahan T, Bertolotto M andMangina E (forthcoming). A web-basedmultimedia virtual reality environment fore-learning. In Proceedings of Eurographics ’04.Short presentations and interactive demossection, Grenoble, France, August 2004, 9–13.

ParallelGraphics (2004). The Pocket Cortona. Atwww.parallelgraphics.com/products/cortonace/,accessed April 2004.

Roche B and Mangina E (2003). Agent-basedframework for intelligent tutoring systems withinUniversity College Dublin. In Proceedings of theInternational Conference on InformationCommunication Technologies in Education,Samos, Greece, July 2003.

3Name3D (1997). Animation created by CindyBallreich (c) 1997. At www.ballreich.net/vrml/h-anim/nancy_h-anim.wrl, accessed April 2004.

VRML (1997). VRML 97 specification. Athttp://tecfa.unige.ch/guides/vrml/vrml97/spec/,accessed April 2004.

Monahan McArdle Kilbride Mangina Bertolotto 137

AbstractThe m-learning emergency system (mLES) is amodel developed by ENEA (the Italian NationalAgency for New Technologies, Energy and theEnvironment) to support management ofemergencies through m-learning. The modelprovides system-comprehensiblerepresentations and user-navigable access.The content needs to be developed usingrepresentations that separate content fromformat for flexible delivery, independent of thedevice used.

At the next level, we need to add to that contentmodel an architecture that provides knowledgeof the user and their context, as well asknowledge of the content available, to providethe right match. This requires a usercategorisation scheme and a mapping process tocreate an intelligent delivery environment. ThemLES provides also for the training of a new typeof professional – emergency operators whofunction as intermediaries between the peoplepersonally experiencing the high-risk oremergency situation and the m-learning-basedtechnological system. The mLES can become animportant asset for the community, with positiveimplications in the security and social domains,representing a valid tool for those who alreadywork in emergency management in the differentoperational centres in the field.

Keywordsm-learning, emergency, scientific knowledge

A new model for the m-learning emergency scenarioin risk contexts: the emergency operatorAnna MorenoSergio [email protected]@casaccia.enea.itENEA (Italian National Agency for New Technologies, Energy and the Environment)Via Anguillarese 30100060 RomeItaly

1 IntroductionThe coming of TIC (Training for Industry andCommerce) in the modern economy and thediffusion of the internet at a global level havebrought about a revolution also in methods oftraining, introducing a new educationalphilosophy – e-learning. The rising importance ofe-learning was acknowledged also by theEuropean Commission (EC) when it adopted inMay 2000 its eLearning Action Plan, whichlaunched an initiative aiming to redraw the way todeliver education in Europe, supporting trainingprocesses that use new multimedia technologiesand the internet to improve learning quality.

The use of information and communicationtechnologies (ICT) in education and training hasundergone several paradigm shifts over the lastthree decades. The concepts of e-learning(learning supported by digital ‘electronic’ toolsand media) and m-learning (e-learning usingmobile devices and wireless transmission) haveonly emerged very recently. Handheld devicesare emerging as one of the most promisingtechnologies for supporting learning andparticularly collaborative learning scenarios,mainly because they offer new opportunities forindividuals who require mobile computersolutions that other devices cannot provide.During the last few years, mobile technology hasbecome integrated into day-to-day activities.Handheld computers appear to be part of ageneral movement towards mobile technology.

The convergence of computing andcommunication is a process that is about to turnphones and mobile terminals into powerfulmultimedia units. For example, SynchronisedMultimedia Integration Language (SMIL), whichis based on Extensible Markup Language (XML),has been devised for the distribution ofsophisticated multimedia content. These formsof interactive multimedia offer new possibilitiesfor how we learn, think and communicate.

Moreno Grande 139

Mobile learning can be defined as ‘…any serviceor facility that supplies a learner with generalelectronic information and educational contentthat aids in acquisition of knowledge regardlessof location and time…’ (Lehner and Nosekabel2002). Technologically, we can now define mobilelearning as e-learning’s new frontier, andaccording to a study by Adkins (2003), by 2006,the m-learning market will be worth more than5bn US dollars. The enormous potential ofm-learning is ensured by the rising developmentof the wireless and mobile telephone market.

According to recent research on mobiletelephony by Intex Management Services (IMS),from sales of 288m mobile phones in 1999, wewill reach sales of 1bn items in 2004. The growthof this market is predicted to reach 1.34bn mobilephones in 2006, when there will be 1.79bn usersthroughout the world.

2 The mlearning emergencysystem (mLES) projectENEA, the Italian National Agency for NewTechnologies, Energy and the Environment (Enteper le NuoveTecnologie, l’Energia e l’Ambiente),has been engaged for a long time in handlingprocesses of knowledge transfer to enterprisesand schools, as demonstrated by varioustechnological transfer initiatives realised at a locallevel, and by many collaborations with schools ofall kinds. ENEA has always promoted thediffusion and transfer of its own research results,supporting and favouring the technologicalinnovation processes of small and mediumenterprises (SMEs).

Precisely because of the importance, for aLearning Economy, of training in technologicaltransfer processes, ENEA developed ane-learning platform to make sure that knowledgeand capabilities generated within the agencywould be transferred through training to thehighest possible number of individuals, andparticularly to those working in SMEs.

ENEA’s e-learning platform was launched in 1996;the project was devoted to the continuingeducation and retraining of workers. After thissuccessful experience, the mission of the ENEAe-learning platform became the promotion ofsustainable development through the diffusion ofscientific culture and technological transfer to anyperson – not just workers, but also students andunemployed people. Now there are more than12,000 users and 30 courses online, with thenumber of courses expected to double in 2004.

Some of the courses will be translated into otherlanguages to allow other Mediterraneancountries to use the ENEA e-learning platform.Many agreements with schools, universities andprivate and public training organisations areunder way. The purpose of these agreements isto build up an open database of scientific learningobjects that anyone can use.

The online training philosophy that ENEAadopted is dictated by the needs of the endusers who, being strongly motivated to learn, areasking for an absolutely unrestricted training, interms of time as well as methods. ENEA offers asystem that can be considered as self-training forits own users, so that the learners will be free tolearn what they want, when they want andwhere they want. From a pedagogic perspective,mobile learning supports a new dimension in theeducational process. The ENEA educationalprocess and philosophy are very close to them-learning educational philosophy – that mobilelearning systems should be capable of deliveringeducation content at any time and anywhere thelearners need it.

Emergency m-learning is one of the possibleapplications of m-learning, offering the possibilityof exploiting the m-learning characteristics forlearning processes in emergency cases.

An emergency is a situation that can producerisks, precisely because it is different from theevents that are normally experienced by workersor the general public. The ENEA project’s idea isto create a technological system that ensures,through mobile learning, the management ofemergency cases.

To execute this vision, we start with a contentmodel that provides system-comprehensiblerepresentations and user-navigable access.

At the next level, we need to add to that contentmodel an architecture that provides knowledgeof the user and their context, as well asknowledge of the content available, to providethe right match. This requires a usercategorisation scheme and a mapping processto create an intelligent delivery environment.Figure 1 (opposite) depicts the architecture ofENEA’s mLES system.


Knowing how to manage emergencies meansnot only knowing the procedures and technicalaspects needed to limit such risks, but alsoknowing how to be effective communicators andhow to control the emotional situations that cansurface. To be an effective communicator inemergency situations calls for attitudes andcapabilities different from those needed inordinary communication: authoritativeness,emotional strength, the capacity to adjust, etc.

The mLES, then, provides for the training of newprofessional figures – emergency operators whofunction as intermediaries between the peoplepersonally experiencing the emergency orhigh-risk situation and the m-learning-basedtechnological system.

So, in the case of an emergency, the subject atrisk contacts the emergency operator through amobile device (mobile phone, palm computer,etc) and the emergency operator uses mLES toextrapolate the emergency learning object thatrepresents the most appropriate solution, andthen forwards it, via the mobile system, to theperson at risk.

The emergency operator is a key figure, since aswell as knowing how to interpret and to manageemergency calls, he or she must also be capableof managing the possible panic of the person atrisk. To ensure that the system will be efficient,we must obviously have an articulate contentrepository containing all the examples of likelyemergency events and the respectiveemergency learning objects with which torespond to the various help and emergency calls.

Figure 1 Architecture of ENEA’s m-learning emergency system (mLES)

3 ConclusionIn Italy, the use of mobile phones to report anemergency has a yearly traffic of 33m calls. Thisdata allows us to claim that an emergencym-learning system can become an importantasset in Italian life, with positive effects in thesecurity and social sectors. It represents a validtool to support those who already work inemergency management in different localoperation centres; for example, law enforcementagencies, public service agencies, localemergency services, health centres, schools andcivil protection organisations.

In 2004, ENEA designed and created aninnovative platform called MATRIX 3, whichincludes videoconference functionality for onlineclasses and seminars, using Active Server Pages(ASP), web-based and Real technologies.

Furthermore, the ENEA Usability Lab testedadvanced visual interfaces to interrogatedatabases and to search on the Web in Javalanguage. In this context, it was decided toproceed with Java-realised modules, and to passlater to a full Java platform for use on intranets aswell as the internet. The outcome of this strategywas the implementation of mLES with itsmodules JNetseminar and JNetLesson, whichaims to combine the functionality of web-basedmodules with other more innovative andcomplete functionalities.

Moreno Grande 141

Thanks to its experience in the e-learning sectorand in research on advanced computer systemssupporting training and video communication,ENEA is ready for this challenge in the mobilelearning field, and is open to agreements andcollaborations aimed at making an m-learningemergency system a well-established reality atthe service of the local community.

References and further readingAdkins SS (2003). Market analysis of the2002 US e-learning industry. Available viabrandon-hall.com

Kinshuk (2003). Adaptive mobile learningtechnologies. Palmerston North: Department ofInformation Systems, Massey University.

Lehner F and Nosekabel H (2002). The role ofmobile devices in elearning – first experiencewith [the] elearning environment. In M Milrad,HU Hoppe and Kinshuk (eds) Proceedings of theIEEE International Workshop on Wireless andMobile Technologies in Education. Los Alamitos,CA: IEEE Computer Society.

Quinn C (2002). Flexible learning: mobilelearning objects. White paper. Bellevue, WA:Knowledge Anywhere. Available viawww.knowledgeanywhere.com

Vavoula GN and Sharples M (2002). KleOS: apersonal, mobile, knowledge and learningorganisation system. In M Milrad, HU Hoppe andKinshuk (eds) Proceedings of the IEEEInternational Workshop on Wireless and MobileTechnologies in Education. Los Alamitos, CA:IEEE Computer Society.


AbstractThe MOBIlearn project, co-funded by theEuropean Commission (EC), the US NationalScience Foundation (NSF) and the AustralianDepartment of Education, Science and Training inJuly 2002, is strategically positioned to providerelevant research outcomes in the area ofinnovative use of mobile environments to meetthe needs of learners, both working by themselvesand with others. During the project, we had tofind ways of harnessing all the available expertiseto feed into this vision of the future, andscenarios for ‘envisionment’ are an effectivemeans to achieve this. The paper introduces theMOBIlearn museum scenario and the results ofthe preparatory work before the final user trial.

Keywordsm-learning, user scenarios, mobile learningarchitecture, user trials, evaluation planning

1 IntroductionThe integration of new technologies in educationand training is in essence a culturally drivenprocess, with the need to bring about change notonly in people, but in the entire learningenvironment. In recent decades, political andsocial progress has emphasised the need for freecirculation of knowledge, as the most advancedanswer to the increasing need for new skillsrelated to new technologies and newsocio-economic models created by theinformation society.

The e-Mobility 2001 EU–Information SocietyConference on Mobility in the KnowledgeEconomy held in Göteborg, highlighted thepriorities that should be explored – the need todefine new work paradigms (eg mobile worker)together with innovative models for their social,economic, cultural and environmental deployment,while preserving the local nature of content(national and regional) and cultural heritage.

MOBIlearn user trials on the museum scenario:the first results of preparatory work

At the same event, it emerged that thesustainable social and economic deployment ofsuch models within the information society ofthe third millennium will see new technologiesfor mobile access to knowledge taking a key role.

On these social and technological premises, theMOBIlearn project (Da Bormida et al. 2002) aimsto improve access to knowledge for selectedtarget users, giving them ubiquitous access toappropriate, contextualised and personalisedlearning objects, by linking to the internet viamobile connections and devices, according toinnovative paradigms and interfaces.

The Göteborg Conference also underlined theneed for pilot experimentations and technologicalapplications to ensure the fast spread and uptakeof planned models and related services in order topreserve Europe’s leadership in the exploitationand creation of new mobile technologies. Theneed for this is becoming urgent. Thus, theMOBIlearn project is justified in two ways: itspioneering research and development (R&D)directly targets priority areas for the informationsociety; and its exploitation directly addressesthe need for Europe to stay dominant in theimportant area of mobile applications.

The project includes studies of conceptualmodels and new methodologies, with prototypesto implement them. These will be evaluated intrial application fields set up and managed byinternational partners participating within theMOBIlearn project. The objective is to improvethe knowledge level of individuals by optimisingthe cost and time of learning processes. Thismaximises the opportunities of threerepresentative groups (Taylor et al. 2002):

p workers – to meet their job requirements and toupdate their knowledge continually

p citizens as members of a culture – to improvetheir learning experience while visiting a culturalcity and its museums

p citizens as family members – to have simplemedical information for everyday needs.

Murelli Da Bormida 143

Elena [email protected]à Cattolica del Sacro CuoreVia Emilia Parmense 8429100 PiacenzaItalywww.mobilearn.org

Giorgio Da [email protected] Interactive LabsVia PortobelloAbbazia dell’Annunziata16039 Sestri LevanteGenoaItalywww.mobilearn.org

2 Scenario definition and functionThe function of scenarios in this context is to setthe scope of user goals, to explain to otherpeople about possible activities in the mobilelearning domain, and for others to explain to uswhat they envisage. Various methods can beused to generate scenarios for this purpose,including storyboards, video clips, diagrams,textual stories and so on.

Scenarios (Taylor et al. 2002) are tools forordering one’s perceptions about alternativefuture environments in which today’s decisionsmight be played out. In practice, scenariosresemble a set of stories, written or spoken, builtaround carefully constructed plots. Stories are anold way of organising knowledge; when used asstrategic tools, they confront denial byencouraging – in fact, requiring – the willingsuspension of disbelief. Stories can expressmultiple perspectives on complex events;scenarios give meaning to these events.

Scenario planning is a method for learning aboutthe future by understanding the nature andimpact of the most uncertain and highlyinfluential forces that will affect it. So far, thescenarios we have produced have beendeveloped by a relatively small subset of thepartners associated with MOBIlearn.

At the first MOBILearn meeting, we agreed aworking definition of ‘scenario’ as follows:

Workpackage 2 scenarios will be constructed atan appropriate level of detail which will be quitehigh. They will contain a narrative, describing asituation where a user is engaged in a task. Thescenario will be relatively informal, and have thecharacter of a true story. Scenarios will becreated by members of WP2 (in the firstinstance). We will invite other members of theMOBIlearn project to contribute scenarios.

These scenarios will form the basis of the firstengagement with users and stakeholders. Usersand stakeholders will also be asked to constructtheir own scenarios.

The function of scenarios is to set the scope ofuser goals, to explain to other people aboutactivities within the mobile learning domain, andfor others to explain to us what they want.Various methods can be used to generatescenarios, including storyboards, video clips,diagrams, textual stories etc.

(MOBIlearn WP2 2002)

3 Developing themuseum scenarioMuseums are the means whereby we research,interpret and present our insights into the naturaland cultural worlds. They represent our beliefsystems concerning cultural interrelationships,our relationship with the environment and ourplace in the universe. They give each person aspace in which to dream.

Wireless technology is becoming a part of themuseum experience. In an effort to bring art andscience to life for a new generation of technicallysophisticated patrons, an increasing number ofmuseums are experimenting with advancedmobile technologies to make going to museumsmore interactive, more educational – and more fun.

Newly emerging portable device and wirelessnetwork technologies have the potential toenhance significantly the experience of amuseum visit. On the exhibit floor, visitorscarrying wirelessly connected devices can begiven opportunities for exploration, sharing,explanations, context, background, analyticaltools and suggestions for related experiences.

In addition, conventional desktop and internettechnologies can help to extend the visit – both inadvance, through activities that orient visitors;and afterwards, through opportunities to reflecton and explore related ideas. In the followingsection, we give some examples of projects thathave been undertaken in this area.

The San Francisco Museum of Modern Art(SF MOMA) launched ‘Points of departure:connecting with contemporary art’, anexperimental programme that offers visitorshandheld devices connected to a wirelessnetwork to provide them with video clips aboutworks on display. ‘In many cases, you have theactual artist telling you about the work you’relooking at,’ said Peter Samis, Program Managerfor Interactive Educational Technologies at themuseum. Using handheld devices such as HPiPAQ Pocket PCs, visitors will use the beacons totap into web-based information about an exhibitwhile at the museum, or access the content lateron from their home or school PC. ‘For us, it’s notjust about enhancing the experience during thevisit,’ said Dr Robert Semper, the museum’sexecutive associate director. ‘It’s also aboutoffering content that extends learning beyond themuseum’s walls by giving you access toinformation about activities or experiments youcan try on your own at home or at school.’Of course, building the infrastructure tosupport such a system is a massive endeavour.


‘The interface, content and devices have all gotto work seamlessly in a way that enhances thevisitor experience and brings value,’ said MirjanaSpasojevic, a project manager at HP Labs. ‘It’sabout bringing pervasive computing out of thelab into the real world.’

As early case studies, museums will help toanswer questions that will shape the future ofpervasive or ubiquitous computing. ‘We’re goingto learn what types of device people want to use,what kinds of graphical interfaces they like, whatcontent they prefer – text, audio, visual – andhow they like to use these devices withoutfeeling like a bunch of cyborgs,’ said SF MOMA’sSamis. HP iPAQ Pocket PCs are in use inmuseums – ‘It’s the way of the future’.

Wireless is a great way to give visitors access tocontent that provides context to exhibitions.‘Another key will be finding out how thesetechnologies impact [on] informal learning,’ saidSpasojevic. ‘We’re finding that kids adapt verywell to the technologies, while older people tendto want to share the devices and discover thingstogether.’ For the museums themselves, thechallenges of mobility are not limited to thetechnology. HP and the Exploratorium inSan Francisco (a children’s science museum)launched a large-scale test of wirelesstechnologies with around 100 participants, todevelop models and methodologies for studyingusers in mobile computing environments andthen report their findings to the museum andtechnology communities.

‘There’s no telling what we’ll learn,’says Spasojevic. Yet those involved withimplementations at other museums have theirown thoughts. ‘The first thing you learn is [that]the technology changes so fast that you shouldn’texpect your first implementation to be the finalsolution,’ said Samis. ‘All experiments are good,but they’re really just grist for the mill.’

3.1 Requirements in themuseum settingConsidering that we are going to create a mobilesystem in a museum, our system design startedwith the following assumptions.

p The system users are a diverse group, covering alarge range of ages, levels of scientificsophistication and familiarity with computers andhandheld devices. We would like to providepersonalisation and user control of theinformation and services through customisationat the level of an individual or of a group. Thesystem should also support communicationbetween visitors.

p The system needs to support both synchronousdelivery of content to handhelds at the exhibitand asynchronous delivery for users who maynot want to absorb a large amount of informationon the spot, but prefer to ‘bookmark’ interestingartifacts and browse the web pages later.

There are issues as to how users can carry anduse handheld devices, and how well the deviceswill survive in a somewhat hostile environment.The physical environment can be noisy andbustling. Many of the exhibits at theExploratorium, for example, require physical actioninvolving the whole body (eg spinning around onturntables or chasing balls across the exhibitarea) and/or manipulation using both hands.

In addition, the specific requirements to createan appropriate m-learning system for museumexperiences are the following:

p personal profile

p virtual contextualised maps

p useful information

p simple and complex queries search

p customised information according to users’preferences in terms of interaction modalities,delivery media and personal interest inspecific topics

p the system must allow groups visiting themuseum to stay in contact with each otherremotely while maximising each member’smuseum experience

p control of traffic flow for themuseum administrator

p display or exchange of content

p linguistic and contextualised supporting services

p the possibility of making notes and commentsat every moment of the museum visit and oneach painting seen.

How can these requirements fit into a scenario?You can imagine: you bring your own WirelessApplication Protocol (WAP) device to themuseum: it could be a Palm Pilot; it could be amobile phone; it could be some combination ofboth things, since we seem to be moving in thatdirection. As you walk through the museum,because of the sensory technology that isembedded in the tombstone signage beside thepictures, or in the floor next to the pictures, yourdevice knows where you are in the museum andit knows what you have looked at. The devicebuilds a record of the objects you look at asyou move around. At the end of that museumexperience, or when you get home, youcan look at the record and say ‘Actually,I want more information on all of them’.


Your device finds the information and delivers itto the platform in your house that is mostappropriate for you. I can imagine anotherscenario. The museum has a profile of you. Oneday you are passing the museum and your devicecalls you, saying, ‘Hello. We have a great exhibiton Native Americans and we really think youwould like it. Why don’t you come in?’ Thistechnology beckons you into the museum. Youcan imagine, then, the rewards of being a regularpatron: the museum knows you better and itknows both how to engage you and surprise you.

An ideal electronic guide to a museum is onethat you take at the entrance, put in your pocketand forget you have. It should fully support afree, natural visit, providing the most appropriateinformation at the right time and place. The onlyactivity required of visitors is to enjoy theexhibition: the interaction is with the(augmented) museum, no longer with the guide;the guide analyses the context and composespresentations adapted to the current situation. Ofprimary importance in an augmented museum isthe interpretation of both visitors’ behaviour (egstops, long stay, coming back) and the context inwhich the action takes place. The places of use(a museum, the open air), the device (a PDA, aphone) or the available infrastructure [networks,Global Positioning System (GPS), infrared] are asimportant as social aspects (eg being alone ornot, who the others are, if this brings pressure)and personal traits (eg attitudes, preferences,interests). This interpretation involves bothmaterial and social aspects.

Starting from these considerations, we havecreated a useful questionnaire and interviewed600 visitors in museums in Florence and at theDuchy of castles in Parma and Piacenza (Italy)to establish their real requirements, needsand expectations.

The visitors interviewed are part of a relativelyyoung public (31% between 18 and 25 years old;23% between 25 and 30 years old); they visitmuseums and castles two or three times a yearin their spare time or during holidays; almost allof them have a mobile phone and 55% of thosehave a Nokia; relatively few people (15%) havea palmtop device.

The results of the first part of the questionnaireare satisfying: the tourists find the idea of mobileservices that are accessible before entering themuseum very interesting and useful. In particular,they are interested in booking tickets in advance,in receiving information about the museumopening times and any related changes, andin viewing the shortest route to reachthe destination.

These services are appreciated by touristsbecause it will help them find their way aroundan unknown town and they want to avoid anyinconvenience (eg from changes in thetimetables), especially during their holidaysor a weekend.

In the second part of the questionnaire,dedicated to the specific requirements that ariseduring the visit inside the museum, the analysisof results highlighted the following issues.

p The services: ‘viewing museum plan’, ‘planningthe visit according to personal interest andrequirements’ and ‘reaching the area of interestas quickly as possible’ all made a goodimpression on tourists. It should be noted thatseveral people expressed doubts about thetransferability of a museum plan onto a PDA ormobile phone as mobile device screens are smalland maps are usually large.

p Opinions are positive about ‘viewing a catalogueof works by a particular artist and understandingthe artist’s techniques’. Few people want or havetime to make notes or comments during theirvisit except for those that visit museums forwork or for study purposes.

p Opinions are very positive on the use of an audioguide [32% of the people interviewed gave this amark of 4 (out of 5) and 32% gave it 5], as it isconsidered a good alternative to the traditionalguide, especially for those who usually visitmuseums alone.

p Generally, we can say that the touristsinterviewed expressed positive opinions on theuse of mobile technologies and their applicationsand interest in the idea.

4 Museum user trialIn parallel with the development of the technicalprototypes, the next step was to refine thescenarios. The need was identified for a moreconcrete instantiation of the scenario to providedetail for the user trials [where the term‘instantiation’ indicates that the scenario is to runin a specific place, at a given time, with realusers and real outcomes (Evans and Taylor 2005).The instantiation of the test scenarios providedthe details of users, content and devices thatwere required. A decision was made to extendthe instantiation to satisfy the needs of theevaluation process. This would include details ofthe context of use, questions to be answeredduring the evaluation and the activity or taskwhich would be used to gather responses.


The target users involved in the trial are people –men and women and representing a range ofages – who have an interest in art and would liketo have an extraordinary experience visiting amuseum, finding useful information on what theysee, and increasing their knowledge to answersome questions or satisfy a sense of curiosityraised during their experience.

How will the user trials take place in the museumstrand of the MOBIlearn project? Due to limitedresources, we require users to speak English orItalian. We can cater for people with severe speechand hearing disabilities or motor impairment, butwe cannot include people with severe visualimpairment or with manipulative difficulties.

We have divided the target group into three parts:

p students

p tourists

p art experts.

From a pedagogic point of view, for each of thetarget group established, we have identified:

p motivation

p knowledge

p skills

p learning experience

p personal knowledge improvement

p learning activity requirements.

According to these objectives, the museumuser trial should follow this plan.

The test will be introduced with a presentationof the MOBIlearn project and its system(MOBIlearn System’s Introduction) andinterviews with the participant to find out ifthey are suitable candidates for the test.

Before being selected as test candidates, thepossible candidates have to fill in a Pre-ScreeningForm (on a mobile device if possible or on paper)where they insert their information and theiruser profile.

The selected users have to fill in a Pre-TestQuestionnaire that will be completed at themuseum before the usability test.

If the user is:

p a student, the questionnaire is used to testhis or her knowledge

p a tourist, the questionnaire is used to understandhow much time he or she has for the visit

p an art expert, the questionnaire is used tounderstand the learning model that is the basisof his/her visit at the museum.

The users will then be divided into 10 groups,each composed of a maximum of five people,and they are asked to:

p do some activities as quickly as possible

p communicate and share contents with others

p find specific information on artists, works of artand the museum.

During the user trial, the user will perform someactivities and sub-activities according to thestarting point (inside or outside the museum),the type of user (tourist, student or art expert)and his/her user profile.

At the end of the test, the users have to fill intwo questionnaires.

p Post-Task Questionnaire: completed at themuseum after the activities to evaluate theMOBIlearn system and its efficiency.

p Post-Test Questionnaire: only for students,filled in at the museum after completingthe test, to evaluate their learning with thismuseum experience.

The results will be analysed to find out whichtypes of problem the users have faced, whichactivities have been performed best and whichcontent the users have found most interesting.

As stated in the Technical Annex (MOBIlearn 2002),the result of this trial is an extensive evaluationreport from external real users that will be thebasis for further MOBIlearn exploitation.

5 ConclusionWe have found that the use of scenarios inthis way has many features.

p It is multi-stranded, supporting mutuallyinforming dialogue.

p It facilitates the identification of each type andcategory of requirements.

p It performs a technical role in relationto requirements.

p It provides the missing link betweenhigher-order concepts, such as pedagogyand learning objectives and lower-levelimplementation aspects.

p It supports verification of system functionalityagainst user requirements.

p It offers a mechanism to support activities linkedto user-system interaction.

p It serves to keep the user at the centre of thedevelopment process.


AcknowledgementsMOBIlearn is a 30-month, 8m euro projectco-funded by the EC (DG Information SocietyD/3–Education and Training), under contractnumber IST–2001–37187, within the InformationSociety Technologies (IST) programme of theFifth Framework Programme of RTD (researchand development). The participation of USpartners (MIT and Stanford University) is fundedby the US National Science Foundation (NSF)under its implementing arrangement with the ECthat specifically addresses cooperative activitiesin the field of e-learning. The participation ofAustralian partners is funded by the AustralianDepartment of Education, Science and Training.

References and further readingDa Bormida G et al. (2002). MOBIlearndescription of work, May 2002. Availablevia www.mobilearn.org, accessed15 September 2005.

Evans D and Taylor J (2005). The role ofuser scenarios as the central piece of thedevelopment jigsaw puzzle. In J Attewelland C Savill-Smith (eds) Proceedings ofMLEARN 2004. London: Learning and SkillsDevelopment Agency.

Gay G, Spinazze A and Stefanone M (2002).Handscape: exploring potential use scenariosfor mobile computing in museums. Atwww.cultivate-int.org/issue8/handscape/,accessed 8 September 2005.

MOBIlearn (2002). Next-generation paradigmsand interfaces for technology supported learningin a mobile environment. Exploring thepotential of ambient intelligence. Project NoIST–2001–37187. Technical Annex.MOBIlearn Consortium.

MOBIlearn WP2 (2002). Minutes of aMOBIlearn WP2 meeting, 1 July 2002.Unpublished document.

Spinazze A (2002). Understanding visitorexpectations and museums as mobile computingenvironments: a report on hand-helds in themuseum landscape (a CIMI symposium).At www.cultivate-int.org/issue8/hssymposium/,accessed 8 September 2005.

Taylor J et al. (2002). MOBIlearn D2.1 Initialrequirements, November 2002. MOBIlearndeliverable to the EC. MOBIlearn Consortium.


AbstractWhile wireless technology provides incrediblescope for the delivery of curriculum content tocommon types of ubiquitous wireless mobiledevice – personal digital assistants (PDAs),mobile phones and laptops – a core problemexists. This concerns the inability to delivermaterial in the way that the student requires withrespect to learning styles, in a context that theyunderstand, and which is appropriate to thelearners’ various backgrounds or age groups,including those who may have encounteredbarriers to learning. Such barriers relate todexterity, computer skills, literacy skills, learningstyles, unsuitable delivery methods and badinstructional design of content. This problemis compounded by the lack of a robustassessment model in both instructor-led andcomputer-directed training. In this paper, aconceptual model called MoDCA (Mobile DeviceCollaboration and Assessment) is proposed. Thismodel defines how the material/learning objectsshould be presented and also how the pedagogyshould be tailored across the different wirelessplatforms in such a manner as to promoteeffective learning and assessment for individualswith diverse learning needs which arecontinuously changing and adapting.

Keywordsmobile learning environment, collaborationmodel, evaluation of learning objects,assessment model for learners

1 Wireless devicesThe increasing popularity of wireless devices –mobile phones, PDAs, etc (Baldzer et al. 2004) – hasmeant that the scope for the application of wirelesscommunications and services is vast (Falk 2003;Lu et al. 2003) – from m-commerce to lifelonglearning.The use of wireless devices means 24-houraccess, 7 days a week, irrespective of location.

How can one effectively assess students working in a collaborative mobile environment on an individual basis?Caoimhin O’NuallainAttracta [email protected]@nuigalway.ieNational University of Ireland GalwayUniversity RoadGalwayRepublic of Ireland

This bodes well for lifelong learning as ‘wirelesscommunication services may bring about newways of distance learning and support theconcept of learning anytime, anywhere’ (Hui,Fong and Lau 2003). In a study carried out atErskine College in the US, students were askedto answer multiple-choice questions (based onthe curriculum content just presented to them),deployed to their palm devices (Falk 2003). Theiranswers, which indicated how well theyunderstood the lecture content, were ‘fed back’to the lecturer within seconds. Being able toidentify the material with which the studentswere having problems meant that the lecturer inquestion could address these problems instantly.Meanwhile, at the Nanyang TechnologicalUniversity in Singapore, a wireless campusenvironment was established by developing apersonal communications system (known as thepersonal communicator) that ‘integrated variousservices into a unified platform, providing aone-stop source for both information access andcommunication … to support the delivery ofteaching materials and the exchange of ideasamong staff and students’ (Hui, Fong and Lau2003). In a study of this system, a sample of 20student users assessed the wireless technologyas a medium for the delivery of lecture material.

The results from this study indicated an averagesatisfaction rating of 87%, with regard to theusefulness, availability, relevance and reliability ofthe delivery medium. However, the authors havenoted that in the development of the personalcommunicator, a single user interface (UI) wasused. While this eliminated the need to designmultiple interfaces for multiple communicationdevices, it also negated the importance of andneed for user profiling, disregarding the factthat different users learn in different ways,based first on their individual learningpreferences (ie visual, auditory or kinaesthetic);and second on their age, profession, experience,gender and cultural background, etc.

O’Nuallain Brennan 149

Furthermore, while the study in Singaporeshowed high user satisfaction, the authors alsonote that neither the end learning experience northe actual system itself was assessed.

Consequently, the authors contend that while thewireless environment provides great potential fordistributed learning (and a resultant enrichedlearning experience), research to date has shownthat the pedagogical aspects of individual userlearning (with respect to learning styles and userprofiles), user interface issues, collaboration andassessment (from both a system and a learningexperience perspective) have not been addressed.Their view is supported by Lu et al. (2003) whostate that ‘while there has been an increasingamount of wireless mobile activity, little attentionhas been given in the literature to user acceptanceof (WIMD) wireless internet via mobile devices’.Wireless applications also need to take intoaccount the changing ways in which peopleinteract with and use information for work andstudy. Therefore this paper discusses theimportance of presentation and assessmentwithin the context of curriculum content deliveryacross wireless mobile devices. The authorspropose a conceptual architecture, MoDCA(a wireless device-based collaborative andassessment model), which takes into account thepedagogical aspects of learning, the psychologyof collaboration, the heterogeneous nature ofmobile devices (from a user interface perspective)and the importance of effective assessment.

2 Collaboration within awireless domainStudents learn more effectively through thesharing of ideas with other students and generalbrainstorming which occurs through the use ofcollaborative features. Collaboration can be eithersynchronous or asynchronous, and can includesuch features as:

1 talking in groups: as in a classroom discussion ortelephone conference

2 on a one-to-one basis: peer-to-peer in theclassroom or on the phone

3 e-mail: this can exist through the use of standardclients like Outlook or Eudora or can be built intothe delivery application

4 forums: online question-and-answer sessions/forums with classes (where tutors moderate)

5 texting: on SMS-enabled devices

6 voice-enabled applications

7 online folders for files: these includeshared folders on the network or on individualpersonal computers

8 videoconferencing

9 mentoring: this occurs when a subject-matterexpert engages with the class or individualstudents in the resolution of issues that thestudent may have encountered.

The advantages (the appeal of which will vary,depending on the device type, the learning styleof the user involved and the capability orbandwidth available) associated withcollaboration are multifarious. Some of theseadvantages include (Nikana 2000):

1 increased understanding of the material/curriculum content: through the differentcollaborative methods and initial deliveryapproaches, the student is provided with anincreased understanding and depth of knowledgeregarding the material/curriculum content

2 increased motivation through discussion:through group discussion and dialogue, the useris motivated to learn more

3 quick and effective feedback: discussion and therepetition of ideas and material (through differentpresentation formats) reinforces learning,thereby increasing memory retention

4 cost effectiveness: by this, we mean that oneshould not always be ‘eager’ to develop newinterfaces, but instead should use what isavailable and effective

5 a good assessment tool for the student

6 a good means for the tutor to identify the level ofthe student’s knowledge

7 reinforcing: it acts as a means of reinforcingexisting material

8 alternative views: it allows for differentperspectives to be examined and tested foreffectiveness, through discussion

9 bonding: increased tutor and student bonding, tohelp shy and less enabled students to obtain thetutor’s attention without embarrassing themselves

10 student retention: increased student retentionas a result of increased student motivationand understanding.

While the terms collaboration and cooperationare frequently used interchangeably, Table 1(Nikana 2000) helps to distinguish differencesbetween these terms.


3 The MoDCA architectureThe MoDCA architecture has been designedto address the following question: ‘How canone effectively assess students working in acollaborative mobile environment on an individualbasis?’ The main components of the MoDCAarchitecture (as outlined in Figure 1 overleaf)are as follows.

p Pre-test: consists of two parts – first, a userquestionnaire; and second, some samplecurriculum/learning objects presented to theuser. This curriculum is used to trace usermovements and thinking, based on his/herquestionnaire, to confirm, correct or update theinformation that exists about him/her. Allcaptured information (ie social background, age,experience, gender, layout preference, etc) isinput into the user profile and evolves as thesystem learns about the user.The profile is madeup of 15–20 categories of parameter which takeinto consideration every aspect of the studentand his/her learning. Similarly, the system will beaware of changes to either the profile of the userand/or the devices. Based on this information,the system will then present the coursematerial/learning objects in the manner mostappropriate to the user and to the device beingused, in order to promote effective learning(ie the material will be presented according tothe principles of blended learning, in a mannerwhich ensures that both the left and right sidesof the brain are kept sufficiently stimulated inorder to provide optimum student responses).

p Mobile devices: at this point, the specifics ofthe device are collected and confirmed, thelearning objects are displayed based onthe characteristics of the specific mobile deviceand user preferences, and usage is monitoredand saved in the user profile.

p Assessment model: three models of assessmentare used. First, the user’s interaction is assessed –that is, are the learning objects presented basedon their preferred style of learning? Second, thesoftware is assessed for flaws (which becomeapparent during learner usage) correspondingmainly to human–computer interaction (HCI) andpedagogical issues. Third, the user’s learning isassessed. All information feeds back into theprofile and generates a more accurate livingprofile and living assessment model.

p Actual assessment: the profile ‘knows’ and candetect which device types are being used by thelearner. This is critical as the device type beingused has a big impact on how the learningobjects are both presented and assessed.

p Learning experience: based on the userinformation collected and compared againstother users, it will be possible to validatewhether or not learning is occurring. If it is foundthat learning is not occurring, this will feed backinto the system and aspects of the way in whichthe curriculum content is presented andassessed will change dynamically.The systemcan change the presentation style available,depending on the device and the type and levelof assessment. Such presentation styles couldinclude audio output and input, but will definitelyallow for collaboration between users, which hasbeen proved to enhance the possibility forlearning and the learning experience itself, as itfacilitates high-order learning.

O’Nuallain Brennan 151

Table 1 Differences between collaboration and cooperation

Collaborative learning Cooperative group work

The teacher forms the heterogeneous group Often voluntary group formation

Group members organise joint efforts and negotiate Less formal group activities – at times,who will perform group maintenance roles individuals are assigned specific roles

Less active monitoring; ie the group resolves issues The teacher observes and interacts. Small-group social skills training

The group is responsible for each individual’s learning; Individuals are responsible for their own learning ie group members have to ensure that everyone in with group support; ie teaching and learning is the group understands the work a shared experience between teacher and learner

Most commonly used for learning content Most commonly used for group discussion

Various group and/or individual assessment models Usually no assessment of formal learning

p Learning objects: a learning object represents thesmallest component of a learning item. Theseobjects can be reassembled in different ways andwith different presentation styles so as to besuitable for any device. It is also essential, due toproblems that currently exist with the delivery ofe-learning, that in the delivery of the curriculumcontent, the available bandwidth be taken intoconsideration. In that way, the curriculum contentis delivered seamlessly and coherently to theuser. This is essential in order to maintain thestudent’s attention and confidence in both thedelivery mechanism and the curriculum.

Figure 1 The MoDCA Architecture

The MoDCA model will be implementedusing software agents to ‘push’ and ‘pull’information from the wireless devices to thedistributed servers.

4 ConclusionsWireless technology provides the potential forthe delivery of curriculum content to commontypes of ubiquitous wireless mobile device.However, the manner in which the material/curriculum content is first delivered and thenassessed needs to be addressed. In this paper, aconceptual architecture called MoDCA is presentedwhich takes into account the pedagogical aspectsof learning, the psychology of collaboration, theheterogeneous nature of mobile devices (from auser interface perspective) and the importance ofeffective assessment. In the assessment module,three modes of assessment will be carried out.First, the user’s interaction is assessed – that is,are the learning objects presented based ontheir preferred style of learning? Second,the software is assessed for flaws,corresponding mainly to HCI and pedagogicalissues. Third, the user’s learning is assessed.

All information feeds back into the profile andgenerates a more accurate living profile and livingassessment model. The profile ‘knows’ and candetect which device types are being used by theuser. This is critical as the device type being usedhas a big impact on how the learning objects areboth presented and assessed. The authorsbelieve that the MoDCA model, which applies auser-centred learning and a user-centredassessment approach, addresses many of theproblems which exist in current (standard and)mobile learning device applications.

References and further readingBaldzer J, Suzanne B, Palle K, Jens K, Jochen M,Norbert R and Ansgar S (2004). Location-awaremobile multimedia applications on theNiccimon platform.

Falk H (2003). Electronic campus. The ElectronicLibrary, 21(1), 63–66.

Hui C, Fong ACM and Lau CT (2002). Unifiedpersonal mobile communication services fora wireless campus. Campus Wide InformationSystems, 19(1), 27–35.

ISO/IEC JTC1 SC36 (2001). Informationtechnology for learning, education and training.Participation in SC36/WG2 collaborationtechnologies. Geneva: InternationalOrganization for Standardization/International Engineering Consortium.

Lu J, Chun-Sheng Y, Chang L and Yao JE (2003).Technology acceptance model for wirelessinternet. Internet Research: ElectronicNetworking Applications and Policy,13(13), 206–222.

Nikana (2000). Co-operative group work.Collaborative Learning, January 2000.

Pailing M (2002). E-learning: is it really thebest thing since sliced bread? Industrial andCommercial Training, 34(4), 151–155.


AbstractWe describe the application of the Delphimethod in studying the future options for mobileworkplace learning in one multinational Finnishcompany. In the first round of applying the Delphimethod, the mobile learning experts answered asurvey based on six scenarios of mobile learning.The second round took the form ofsemi-structured interviews, in which theconflicting issues were examined. The aim of thestudy was to examine the present situation andlook at future options for mobile learning whichwould be feasible and beneficial for the company.More generally, the goal was to explore theconnections between mobile workplacelearning, personnel development andknowledge management.

Keywordsmobility, workplace learning, futures studies,Delphi method, scenarios

1 IntroductionIn this case study, we explore the conditions forthe implementation of mobile learning in theoperational environment of one multinationalFinnish company. In order to discover feasibleand beneficial working methods in the future(see eg Bell 1997), we have applied the Delphimethod in studying the future options for mobileworkplace learning. Mobile learning isunderstood here as the option to learnindependently of place and time, and weapproach it from the viewpoint of flexible andlifelong learning. In different situations, learningactivities can be supported with different media,and we aim to find out which mobile solutionsserve the learner – or worker – best. Inconsidering an individual’s learning in theworkplace, we focus mainly on non-formal andinformal learning activities. On the other hand,learning is seen very much by the company asactivities connected to its service processes andpersonnel development.

A case study on the future options for mobile workplace learningMarika PehkonenHanne Murto (née Turunen)[email protected]@uta.fiHypermedia LaboratoryUniversity of TampereFIN–33014 Tampereen yliopistoFinland

2 The Delphi methodThe idea behind the Delphi method is to explorepossible worlds and their realisations throughcollecting and iterating expert knowledge.Linstone and Turoff (1975) define Delphi as amethod for structuring a group communicationprocess. Basically, the group has at least oneopportunity to evaluate and revise its views, andthere is always some degree of anonymityattached to the individual responses (Linstoneand Turoff 1975; Kuusi 1999).

Delphi studies have been criticised for thesubjective and narrow views of the experts andan intention to reach an artificial consensus (seeeg Linstone and Turoff 1975; Mannermaa 1991;Kuusi 1999). However, Linstone and Turoff (1975)advocate the usefulness of exploring differentpoints of view and Turoff (1975) has introducedPolicy Delphi, which pursues severalwell-founded views of the future. In the casestudy, we applied the Delphi method as ahermeneutic and heuristic method of collectingand interpreting the expert views, andconcentrated mainly on the deviating answers aspossible ‘weak signals’ (ie perceived symptomsof change) in the external and internaloperating environment.

2.1 Scenarios of mobile learning asa starting pointSix scenarios depicting different situations ofmobile learning were used as stimulatingmaterial in the first round of the Delphi method.Instead of the wide scope of futures studies, thescenarios focused on the organisational,psychological and cultural, pedagogical andtechnological aspects of mobile learning. Theseaspects represent the contradictory dimensionsof mobile learning that may produce so-called‘wild cards’ and uncertainties (also ‘germs’ and‘weak signals’; see eg Kuusi 1999) in the future.

Pehkonen Murto 153

These opposing dimensions (Ahonen et al.2002a) are illustrated in Figure 1. This approach isbased on earlier studies of future learningpractices and mobile learning and teachingtechnology (eg Gokhale 1995; Bonk andCunningham 1998; Caldwell and Koch 1999; Bork2000; Väänänen-Vainio-Mattila and Ruuska 2000;Collis and Moonen 2001; Ahonen et al. 2002a,2002b, 2003).

Figure 1Possible development trends (and uncertainties) in mobile learning

The scenarios were presented in pictorial andnarrative text forms. Six different models ofaction were depicted using actors, theirsituations, and their actions with materials atcertain times and in certain environments(Mannermaa 1991). Each scenario was followedby an open-ended question about its strengthsand weaknesses. The potential strengths,weaknesses, opportunities and threats (SWOT) inrelation to mobile learning were also summed upin statements after each scenario.

2.2 The second round of Delphi:expert interviewsThe disagreements arising from the first-roundresponses were further explored insemi-structured interviews with the aim ofbringing out the underlying reasons for thediffering points of view.The goal was to producesome plausible future views of mobile learningwhich would conform to the company’s needsand its scope for action in different contexts – indifferent units and countries. The interviews inFinland were conducted face-to-face; while inother countries (France, the UK, the US),teleconferencing was used, supported by aweb-based working environment.

3 FindingsIn our case-study organisation, there aredifferences between countries, cultures andunits as regards technology acceptance andadopting new ways of working and learning.Also, views about workplace learning and aboutthe necessity of evaluating its outcomes differ.We cannot see one future scenario forconverging mobile workplace learning, personneldevelopment and knowledge management in themulticultural environment. Instead, there seemto be many different options for the future.

According to both the survey and the interviews,the main reason for applying mobile working andlearning practices in our case-study companywas efficiency. Working in virtual teams cutstravelling expenses and enhances timemanagement. Accompanying working methodsare perceived to be effective and this facilitatestheir spread across the company. At their best,mobile solutions connected to the operationalsystems help to streamline the businessprocesses; for example, they enablecommunication of real-time information fromthe field of operations.

As a starting point, the assured status of the unitwithin the company and its level of integrationinto the organisational culture are thepreconditions for adopting different types ofmobile operation. Culturally, digital presenceadds a new global sense of community andbelonging to the same organisational culture. Thebasic condition is still the level of IT infrastructureneeded for mobile operations. Vital steps inlaunching new services are the maturing of thetechnologies, the usability of the systems andthe competitive advantage gained with thedifferent forms of mobile work and learning.Instead of approaching the human resources(HR) development in the spirit of lifelong learning,more weight is still put on recognising key skillsfrom the viewpoint of organisational activitiesand developing competences. Mobility is appliedto the actual needs of the company in terms offlexible learning and working.

AcknowledgementsWe want to thank Mr Juha-Matti Arola for hisinvaluable help and comments on the study.Thestudy is part of the Digital Learning 2 project andis funded by TEKES (the National TechnologyAgency of Finland), the European RegionalDevelopment Fund (ERDF) and severalcompanies in the Häme region of Finland.


ReferencesAhonen M, Joyce B, Kotala O, Peltola K andTurunen H (2002a). Scenario work in [the] DigitalLearning project. Working paper (unpublished).

Ahonen M, Koponen H, Syvänen A and Turunen H(2002b). Uudet mobiili-innovaatiot oppimisentukena. In Digital Learning-tutkimusprojektinosaraportti. Hämeenlinnan ammattikorkeakoulunjulkaisusarja A: 4/2002.

Ahonen M, Joyce B, Leino M and Turunen H(2003). Mobile learning – a different viewpoint. InH Kynäslahti and P Seppälä (eds) Mobile learning.Helsinki: IT–Press, 29–39.

Bell W (1997). Foundations of futures studies:human science for a new era. Vol. 1. History,purposes, and knowledge. New Brunswick (NJ):Transaction Publishers.

Bonk CJ and Cunningham DJ (1998). Searchingfor learner-centered, constructivist andsociocultural components of collaborativeeducational learning tools. In CJ Bonk andKS King (eds) Electronic collaborators:learner-centered technologies for literacy,apprenticeship, and discourse. Mahwah, NJ:Erlbaum, 25–50.

Bork A (2000). Four fictional views of the futureof learning. Internet and Higher Education, 3,271–284. At www.ics.uci.edu/~bork/fiction.pdf,accessed 25 November 2003.

Caldwell D and Koch J (1999). Mobile computingand its impact on the changing nature ofwork and organizations. Working paper 24.Santa Clara University. At http://sts.scu.edu/research/MobileComputing.pdf, accessed25 November 2003.

Collis B and Moonen J (2001). Flexible learningin a digital world: experiences and expectations.London: Kogan Page.

Gokhale AA (1995). Collaborative learningenhances critical thinking. Journal of TechnologyEducation, 7(1).

Kuusi O (1999). Expertise in the future useof generic technologies. Epistemic andmethodological considerations concerning Delphistudies. Research Report No 59. Helsinki:Government Institute for Economic Research.

Linstone HA and Turoff M (1975). Introduction.In HA Linstone and M Turoff (eds) The Delphimethod. Techniques and applications. Reading:Addison–Wesley, 3–12.

Mannermaa M (1991). Evolutionaarinen tulevaisuudentutkimus.Tulevaisuudentutkimuksen paradigmojenja niiden metodologisten ominaisuuksientarkastelua. Acta Futura Fennica 2. Helsinki:Tulevaisuuden tutkimuksen seura.

Turoff M (1975). The Policy Delphi. In HA Linstoneand M Turoff (eds) The Delphi method.Techniques and applications. Reading:Addison–Wesley, 84–101.

Väänänen-Vainio-Mattila K and Ruuska S (2000).Designing mobile phones and communicators forconsumers’ needs at Nokia. In E Bergman (ed.)Information appliances and beyond.San Francisco: Academic Press.

van der Heijden K (1997). Scenarios, strategiesand the strategy process. Breukelen: NijenrodeUniversity Press. At http://library.nyenrode.nl/NL/publicaties/NIJREP/nijenrode.html, accessed17 November 2003.

Pehkonen Murto 155

AbstractThe INLET project (Lingua 1)1 demonstratestechniques for promoting the contextualisedlearning of introductory Greek language at theOlympic Games in 2004. This paper will reviewthe implications for understanding motivation inlanguage learning, as well as more generalunderlying principles for approaches to providingubiquitous just-in-time knowledge that involveShort Messaging System (SMS), including anSMS-searchable database.

KeywordsGreek language, Olympic Games, languagesupport, motivation, ubiquitous, just-in-time, SMS

Approaches to just-in-time learning with mobile phones: a case study of support fortourists’ language needsAnita [email protected] of EducationUniversity of London20 Bedford WayLondon WC1H 0ALUnited Kingdomwww3.ellinogermaniki.gr/ep/inlet/

1 Project overviewThe project entitled InNLET (IntroducingLanguage Enhancement Techniques) wastendered under the European Union’s (EU)Socrates Programme as Lingua 1 to EleniMalliou, Savvas Stavros and Sofoklis Sotiriou ofthe Ellinogermaniki Agogi in Chalandri, Greece;and to Antonis Miliarakis and Manolis Stratakisof FORTHnet SA, in Heraklion, Crete.

Technically, it is relatively straightforward.Tourists coming to Greece for the OlympicGames will receive:

p booklets at airports, hotels and sports venuesexplaining the language help available

p some useful Greek phrases in the booklets

p a minidisc with the same information as inthe booklets

p a phone number to register their interest

p SMS messages sent regularly to their mobilephone with useful phrases

p the facility to request and receive SMStranslations from or into Greek.

Figure 1 (overleaf) shows the InLET system,with InLET-initiated and client-initiated supply.

Pincas 157

Figure 1 InLET system, with InLET-initiated and client-initiated supply


2 Educational aspects of INLET

2.1 Features of ubiquitous supportThe parameters of the project fit with generallyagreed principles for ubiquitous support asshown in Table 1.

Table 1 Principles of ubiquitous support

Advantages Limitations

Portable Physical limitationsPersonal Psychological limitationsAccessMulti-device deliveryMulti-userMulti-languageUser management

Table 2 lists some sample application categories(see Clark 2003).

Table 2 Some sample application categories

Sample applications

DiagnosisSelf-assessmentPerformance supportLanguage learningCollaborationOnline mentoringReinforcement

InLET adds another application to the list, namelyinformation supply. We can consider the outputof the project under two headings.

p Information supply, where we will find it linkswith performance support

p Language learning, though we would revise thisto language supply. There are different factors toconsider under each heading.

2.2 InLET information supplyInLET recognises the common fact that peopleon the move need information in three ways – indifferent spaces, in different areas of life, and atmany different times (McLean 2003). But thisdoes not necessarily follow an m-learningparadigm where the ultimate assumption is atransformation of learning styles (Vavoula andSharples 2002). In that paradigm, the principalpedagogical considerations to be taken intoaccount would be (Singh 2003):

1 urgency of learning need

2 initiative of knowledge acquisition

3 mobility of learning setting

4 interactivity of learning process

5 situatedness of instructional activities

6 integration of instructional content.

But we revise this list to fit information supply,without assumptions of learning on the part ofthe user:

1 urgency of information need

2 initiative in information acquisition

3 mobility of setting

4 interactivity of process

5 situatedness of needs

6 integration of content.

Chen et al. (2002) noted four trends inmobile applications.

p The first is a move from courseware to‘performanceware’ – that is, the stand-alonelearning content model needs to be transformedinto a context-driven, task-sensitive,performance-support model. In our model, thereis performance, but not necessarily learning.

p The second is a move from instructional designto performance-based design, where compilingcontent and courses is transformed into job,task, activity and business application contextanalysis. Our model is based on activity andapplication context analysis.

p The third is a move from course management tobusiness workflow, where business workflowand processes become the delivery platform formobile learning and performance support. Ourmodel exemplifies a move towards activity flowand performance support.

p The fourth is from mouse-and-click topen-and-voice interface, which our modelclearly follows.

These trends are exemplified in relation tolanguage provision in the next section.

Pincas 159

2.3 INLET language provisionGiven the well-known current obstacles oflimited memory storage, small screens andintermittent connectivity (Vavoula and Sharples2002), we agree with the recommendations ofMcLean (2003) – that any initiative must eitherimprove on existing practices or do somethingthat cannot be done using existing technologiesand practices, and that there should be noattempt to provide a total solution to any learningprogramme. Thus, in considering the languageissue for tourists at the Athens Olympic Gamesin 2004, it was clear that learning evenelementary Greek would not take place in aperiod of 1–3 weeks. Clearly, it was linguisticperformance activity of some kind that touristswould require. We therefore considered whichwere likely to be the most useful linguisticactivities that a tourist would normally wish toperform; and then, which would be most usefulto our specific subset of tourists.

An inspection of a typical pocket Berlitz dictionaryand phrase book for travellers shows 2500dictionary items, seven pages of basic grammarand pronunciation, and 140 pages of phrases inthe categories: arrival, hotel, eating out, travellingaround, sightseeing, relaxing, making friends,shopping, money, post office and doctor. Thisis clearly more than information supply; it is anaid to learning as well as just-in-timeperformance support.

To follow our goals meant that we needed toselect a small subset of the categories. Therewas no ready research information to guide thatselection. We therefore used our judgement thatthe following categories of words and phraseswould fulfil the most urgent Greek languageneeds for a tourist spending only a few days inthe country.

p Basics: greetings and other polite phrases,naming, numbers, and words like ‘come’,‘need’, ‘know’, ‘help’, ‘wait’, ‘have’, ‘like’, ‘want’,‘is’, ‘am’, ‘are’, ‘very’, ‘passport’.

p Where: ‘Where is?’, ‘here’, ‘there’,‘right’, ‘left’, ‘in’, ‘to’, etc; ‘by taxi’, ‘bus’,‘subway/underground/metro’, ‘train’, ‘on foot’,‘street’, etc.

p When: ‘now’, ‘later’, ‘today’, ‘tomorrow’,‘What is the time?’, days of the week, etc

p Olympic sports: ‘Olympic Games’, ‘archery’,‘athletics’, ‘fencing’, etc

p Buying: ‘money’, ‘credit card’, ‘How much?’,‘expensive’, ‘cheap’, ‘sales’, ‘buy’, ‘I would like’, etc.

Languages used in the InLET system are Greek,English, German and Slovenian, based on theperception that English is the world’s lingua franca,and the other two were languages of a priorexisting partnership, but also covered a range ofpeople likely to attend the Olympic Games.

We feel that our language choices supplyinformation in terms of the categories listed insection 2.2. Clearly, all the needs we cover wouldbe strongly ‘situated’, with an urgent need for theGreek language information in the specificsituation, which would vary according to themobility of setting. Our system is designed insuch a way that some of the dictionary items aresupplied to the mobile users by SMS on a regularbasis several times every day. Others are availableby request, so that users take the initiative inacquiring the information through the mobileinteractivity process, in which the users themselvesintegrate the situation and the content.

In other words, we fulfil the main goals ofjust-in-time learning, even though we do notassume that users will necessarily retain theinformation as language knowledge.Nonetheless, we are offering information that isavailable at any time and in any place, and alsoinformation on demand. Finally, the usersthemselves determine whether and how theywill engage in developing short-term skillslearning when they attempt to understand or saythe Greek words and phrases.

2.4 Why language provision by SMS?The system as described actually provides whata book does. So the question arises: whydevelop our system rather than assume touristscould buy their own phrase books?

First, we could not assume they would comewith helpful books, so booklets with some usefulGreek phrases to be picked up at airports, hotels,and sports venues would fulfil a need. Second,for those with minidisc players, we offer thesounds of the language as well as the writtenform. Third, the SMS messages sent regularly totheir mobile phones with useful phrases wouldbe a constant reminder of the real possibility ofcommunicating – even minimally – withmembers of the local community. Fourth, weincorporated items that are more targeted than astandard tourist phrase book would be. Fifth,even though learning Greek is not an aim weexpect in a tourist to the Olympic Games, all ourmessages could be saved by the users andtherefore reused as often as needed, andperhaps ultimately learned.


However, perhaps the strongest motivation forthe project was not to fulfil the needs of thetraveller to the Olympic Games, but the needs ofGreece as a minority-language state within theEU. We recognise the current limited interest inother languages, and that we are dealing withpeople with a very low motivation to learn thatlanguage. Why, then, should they use any Greekat all? Our answer is that we want them to ‘feel’a connection to Greece and its people. ForGreeks, this has to be linked to the language,even though there are many people who aredeeply interested in Greek history and culture,especially that of Ancient Greece, with noknowledge of, or interest in, the language. Butwe feel that by enabling visitors to say andunderstand at least some very minimal, basicwords, a rapport with Greek people might beestablished in ways that would not happen justby using English. This is what we often call phaticcommunication; it does not really conveyanything informative (ie facts), but it establishesan emotional bond.

We need to motivate them to make the effort totry to use Greek; and, as we all know, a smalleffort is usually more likely to be undertaken thana bigger one, especially if it attracts attentionby being in some way unusual or unexpected.People are less likely to reject an offer if it iseasy to cope with and it does not cost themanything. The InLET project design meets theserequirements.

It has never been done before in quite this way.People with mobile phones may have learned toreceive unsolicited advertising SMS messages,or messages from their connection provider,but not free material that would be of genuineuse to them (although, admittedly, someadvertisements are useful). The InLET systemis therefore likely to arouse some curiosity. It iseasy to cope with, even while the user is doingother things – for example, travelling withdistractions in the background and therefore witha lower level of concentration. The SMSmessages will be short – no more than 160characters. Use of the minidisc is also quitesimple for anyone who has a MiniDisc Walkman.Finally, registering the phone number andopening incoming SMS messages is not difficultand will be free.

The author did a small but international survey ofmore than 200 mature adults studying in theOnline Education and Training course at theUniversity of London,2 to discover whetherpeople would want the help we are offering. Theresults are surprising, perhaps, and disappointing.

By definition, being on this course, thesestudents were alI ICT-literate people, and hadchosen to improve their skills even more. Yet halfof them said they did not want to receive theSMS help. One respondent was very violentlyopposed, claiming he could learn the languagebetter in a person-to-person way in the tavernas,playing cards and drinking ouzo with the villageGreeks. He misunderstood InLET’s goals, whichwere not to teach the language. However, suchresponses to a hypothetical question posed out ofcontext need not, perhaps, be taken too seriously,since the same people could respond quitedifferently if asked while actually in Greece. Thisremains to be verified during the Olympic Games.

Overall, the project highlights contemporary usesof ubiquitous technology that are highly likely tohave an impact upon more learning and teachingmethods as palm devices are integrated withinternet-connected mobile phones and suppliedwith fold-away keyboards. It clearly has widereducational implications for any learning thatinvolves no teacher and is oriented towardsshort-term skills without the implication oflonger-term learning.

ReferencesChen YS, Kao TC, Sheu JP and Chang CY (2002)A mobile scaffolding-aid-based bird-watchinglearning system. In M Milrad, HU Hoppe andKinshuk (eds) Proceedings of the IEEEInternational Workshop on Wireless and MobileTechnologies in Education. Los Alimatos, CA:IEEE Computer Society, 152–156.

Clark D (2003). M-learning. The best place to learn? No place. At www.epic.co.uk/content/resources/white_papers/m-learning.htm,accessed 12 September 2005.

McLean N (2003). The m-learning paradigm: an overview. At www.oucs.ox.ac.uk/ltg/reports/mlearning.rtf, accessed 13 September 2005.

Singh H (2003). Leveraging mobile and wireless internet. At www.learningcircuits.org/2003/sep2003/singh.htm, accessed 13 September 2005.

Vavoula GN and Sharples M (2002). KleOS:a personal, mobile, knowledge and learningorganisation system. In M Milrad, HU Hoppe andKinshuk (eds) Proceedings of the IEEEInternational Workshop on Wireless and MobileTechnologies in Education. Los Alimatos, CA:IEEE Computer Society, 152.

Pincas 161

Endnotes1 The InLET project (EU project reference no

109926–CP–1–2003–1–GR–LINGUA–L1)was originated by Eleni Malliou, Savvas Stavrosand Sofoklis Sotiriou (Ellinogermaniki Agogi,Chalandri, Greece) and Antonis Miliarakis andManolis Stratakis (FORTHnet SA, Heraklion,Crete). Partners in the project are:

p Anita Pincas, Institute of Education, Universityof London, UK

p Hellenic Broadcasting Union, Greece

p Eleftherios Venizelos International Airportof Athens, Greece

p Milos Educational Women’s Collaboration forActivities in Tourism, Greece

p Community Service Volunteer Media, UK

p Radiotelevizija Slovenia

p Nederlandse Christelijke Radio Vereniging.

2 For information on the Online Education andTraining course for professionals involved ineducation, see www.ioe.ac.uk/english/OET.htm,accessed 10 September 2005.


AbstractWe have coined the term ‘mobile lesson’ toexpress the concept of lessons carried outdirectly on site. Mobile is the keyword to beemphasised because all actors (teachers, tutors,students) in a mobile lesson are free to movewithin sites of educational relevance, and bymeans of mobile devices – Tablet PC, personaldigital assistant (PDA), smartphone equippedwith a Global Positioning System (GPS) – andgeo-referenced applications, are able to performeducational activities on the spot. Suitabledisciplines for mobile lessons include history,geography, physics, biology and linguistics andideal scenarios are archaeological sites andnatural parks. Each point of interest for thelesson is expressed in latitude–longitudecoordinates and is called a ‘hotspot’.

The technical infrastructure for mobile lessonsconsists of a web browser augmented with GPScapabilities and a web service able to generatedynamic content according to the actual positionof the end user.

We have chosen a web-based architecturebecause web browsers are commonly availablefor a large number of platforms such as PCs,PDAs and smartphones.

The browser has been extended with acustomised add-on which is able to read thecurrent position from GPS hardware and adds a‘user-location’ header to each http request. Theweb applications can return web pages createdon the fly and tailored for the end-user position.

Keywordsmobile lesson, GPS, web

1 IntroductionWe coined the term ‘mobile lessons’ for lessonsheld outside the classroom. During a mobile lesson,all actors (teachers, tutors, students) are mobileand have to move to perform the required tasks.

Mobile lessons: concept and applications for‘on-site’ geo-referenced lessons

Topics and disciplines that may benefit from themobile lessons format are geophysics andmineralogy in geography, the study ofmonuments in history or trees and ecosystemsin biology, distance measuring in physics andgeometry, and dialects in linguistics.

Mobile lessons are not a new teachingtechnology or methodology (Du Boulay andLuckin 2001), but they are a way of makinglessons more efficient and more attractive usingnew communication technologies and mobiledevices. Students can improve their knowledgedirectly on site, looking for information, observingthe real environment, and feeling more involvedbecause they are able to behave autonomously(Rogovin 1998). To integrate mobile devices intoour ‘on-site’ approach, we designed andimplemented software that helps teachers tocreate and edit lessons that will later be deliveredto students and accessed through mobiledevices. Finally, the mobile lesson format allowsteachers to monitor students’ activities whilethey are on site (Giroux et al. 2002).

2 Organisation of a mobile lessonA mobile lesson comprises four main steps.

The first step concerns the lesson design:teachers and/or tutors choose the scenario (eg anarchaeological site or a natural park), go to thesite, choose particular points of interest (calledhotspots), and mark them with their GPScoordinates. Then, either in the hotspot or later inthe school, they prepare the pedagogical materialsuch as historical information, hotspotdescriptions, observation instructions, help andtips. They can also prepare quizzes and tests,both relating to the whole scenario and toparticular hotspots. This entire step is performedby means of the mobile lessons authoring tool.

Pintus Carboni Paddeu Piras Sanna 163

Antonio PintusDavide CarboniGavino PaddeuAndrea PirasStefano Sanna

[email protected]@[email protected]@[email protected]

CRS4Center for Advanced Studies, Research and Development in SardiniaEdificio 1Loc. PiscinamannaPolaris 09010Pula (CA)Italy

The second step is a ‘traditional’ lesson in theclassroom to impart to the students thebackground knowledge required to appreciatefully the content of the on-site experience. Thelesson should also explain the technicalequipment they are going to use.

The third step is the on-site experience. Studentsand teachers are free to move around thescenario with their mobile devices. Studentshave to discover the hotspots previously chosenby teachers and the software provides them withevery geo-referenced piece of informationprepared by the teachers in the first step. Themobile lessons can be like a treasure hunt,where students receive a score for every hotspotfound and for every quiz question answered.

The fourth and last step is performed back in theclassroom. Teachers conclude the lesson byassessing the results of the experience andfocusing their attention on pedagogical correctiveactions for those students who have achievedpoor results. Figure 1 shows the four main stepsof the mobile lessons concept.

Figure 1 The four main steps of themobile lessons concept

3 Design goalsThe three main design goals for the mobilelessons system are as follows.

p Devices must be online and communicate with acentral server containing all relevant data. Wedisapprove of the use of offline devices withpreloaded data because keeping every device upto date is a tedious task.

p We do not want to write from scratch a brandnew multimedia interactive application for thedevice. We want to reuse as much material aspossible from existing software.

p We do not want to define and implement a newclient-server protocol, but instead we prefer touse a reliable implementation of the HTTPprotocol which is well tolerated by firewall/routersecurity policies.

The design goals stated above lead to anarchitecture based on a web browser,augmented with a GPS interface and a webservice able to generate dynamic contentaccording to the actual position of the end user.

The ‘augmented browser’ is part of the GPSWebProject (Carboni et al. 2004). We have chosen aweb browser for three reasons.

p It is commonly used by a large number of users.

p It is included in a wide set of devices, such asPCs, PDAs and smartphones. Although eachbrowser provides its own user interface (UI)(sometimes customised with original solutions,such as tab panes or third-party toolbars),common basic functionalities are easily foundand accessed by users.

p It allows developing applications using commonscripting languages (PHP, ASP and so on) and theexisting ones can be adapted with few changes.

The GPSWeb module is available for Mozilla Firefoxand Microsoft Internet Explorer. The module readsthe user’s current position from the available GPShardware and adds the ‘user-location’ HTTP headerto each browser request. The web applicationsreturn web pages created on-the-fly which arespecific for the position given in the header.

The clients are notebooks and Tablet PCs withGPSWeb-enabled browsers that are connectedthrough serial or Bluetooth ports to GPS devicesand able to access the internet via GeneralPacket Radio Service (GPRS) connections –although it can be used on small areas withwireless local area network (WLAN) coverage,the application does not rely on a specificnetwork infrastructure.

The web application is developed in Java andPHP and a MySQL database – a popularopen-source database – is used to store anyinformation that is relevant to a mobile lesson.


4 The experimentThe archaeological site of Nora, in southernSardinia, has been selected as the scenario forour first mobile lessons experience. The site has13 hotspots, such as the Temple of Aesculapius,the main Roman street, the theatre, the thermalbaths and so forth.

For each hotspot, the teachers specified thename, a brief description, quizzes, pedagogicalmaterial and the radius of the area of interest(AOI) with the hotspot at its centre. When thestudents were on the site with theirGPSWeb-enabled clients, they asked to start thequiz. If their position was inside the AOI for thathotspot, the quiz was presented to the students.Otherwise, no quiz was shown and the studentlost marks. All questions and answers werestored in the mobile lessons database.

The experiment has revealed several positiveaspects of the concept and applications ofmobile lessons. The teachers and studentsinvolved were very excited about the on-siteexperience, confirming the effective pedagogicalvalidity of the innovative approach of our project.Students find it more interesting to have a directcontact with what they are studying and thecompetition among classmates is an excitingincentive. The teachers find software tools usefulin helping them to create lessons, to preparequizzes, and to analyse student answers.

5 ConclusionsWe have presented the mobile lessonsexperience, based on web technologies such asthe GPSWeb tools. Connected mobile devicesprovide valid support for instant informationaccess and the GPS-enabled browser improvesdata processing on the server side. At the sametime, on-site lessons are a very excitingexperience for both teachers and students. Bymeans of a standard web browser, teachers candefine and modify the lesson whenever theywant (even during the on-site lesson, if needed),while only information or quizzes related to aspecific area are presented to students. Testscarried out at the archaeological site of Nora inSardinia have shown the practicability of this newmobile learning experience. Since wirelesstechnologies are rapidly evolving, mobile lessonswill have several benefits and we can imaginereplicating the same experience in the futurewith Universal Mobile TelecommunicationsSystem (UMTS) phones or wireless networked(Wi-Fi) PDAs.

ReferencesCarboni D, Giroux S, Piras A and Sanna S (2004).The web around the corner: augmenting thebrowser with GPS. In Proceedings of theWWW2004 Conference, Alternate Track Papersand Posters. New York: ACM Press, 318–319.

Du Boulay B and Luckin R (2001). Modellinghuman teaching tactics and strategies fortutoring systems. International Journal of AIin Education (IJAIED), 12, 235–256.

Giroux S, Moulin C, Sanna R and Pintus A(2002). Mobile lessons: lessons based ongeo-referenced information. In Proceedingsof the World Conference on E-Learning inCorporate, Government, Healthcare and HigherEducation. Charlottesville: Association for theAdvancement of Computing in Education (AACE),331–338.

Rogovin A (1998). Learning by doing: homeand school activities for all children. Nashville:Abingdon Press.

Pintus Carboni Paddeu Piras Sanna 165

Abstract‘Dealing with size’ is one of the critical challengesin developing mobile learning elements.

How can we fit all we need onto such asmall screen?

How can we support a wide range of mobiletechnologies, yet not get lost in the ‘lowestcommon denominator’?

During the past 3 years, the m-learningdevelopment teams have found various ways ofaddressing these issues, both technologicallyand pedagogically. This paper will focus on twozones of this development, and solutions wehave found will be presented and discussed.

KeywordsXHTML, mobile learning, screen design, interface

1 The m-learning projectThe m-learning project is a 3-year pan-Europeancollaborative research and development (R&D)programme supported by the European Union(EU). The project has developed prototypeproducts and innovative approaches designed tosupport learning, particularly literacy, numeracyand life and survival skills, using handhelddevices such as mobile phones and palmtop orpocket computers. A key objective is to engagewith and motivate young adults who are notengaged in education or training, particularlythose who are unemployed or homeless.

The consortium is a partnership of organisations –the Learning and Skills Development Agency (LSDA),Cambridge Training and Development (CTAD) andUltralab from the UK, Lecando from Sweden andCentro di Ricerca in Matematica Pura ed Applicata(CRMPA) from Italy – combining skills inpedagogy and technology. The project hasfunded 100 devices (SonyEricsson P800s andP900s; O2 XDA2s) which are currently being usedby 300 learners in the three partner countries.

m-learning via the web: the challenge of size

To deliver a rich learning experience, we areusing a range of technologies including:Multmedia Messaging System (MMS), ShortMessaging System (SMS), Extensible HyperTextMarkup Language (XHTML) (PHP, ASP, JSP, javaservlets), Wireless Application Protocol (WAP),Java 2 Platform, Mobile Edition (J2ME),JavaScript and Flash.

This paper will focus on the development anddesign factors that were raised in the use ofmini-browser technologies. Although size mightappear to be an overriding factor, there were otherconsiderations that were equally, if not more,important. This paper addresses some generalpoints and discusses some possible solutions.

2 Development and design factorsSystems and content were developed primarilyfor mini-browser technologies. This majortechnological choice was influenced by researchcarried out at the beginning of the project by theproject partners and by external indications ofdeveloping trends within the mobile technologymarket. A number of factors were consideredand these influenced the design.

The overarching factor, of course, was the focusof the project as described above, whichinfluenced all designs and choices. Beyond this,the profile of the target group, both in terms ofage and potential problems with literacy, stronglyinfluenced both the way that the materials andsystems were presented and the content of thematerials. An important point that needed to beconsidered was the fact that the m-learningproject lasted for 3 years, during which timethere were many developments in the featuresand services which mobile phone manufacturersand providers offered. This forced us to adopt aflexible approach which would be able to assimilatethese new technologies as they appeared.

Popat Stead 167

Kris [email protected] Polytechnic University (APU)Victoria Road SouthChelmsfordEssex CM1 1LLUnited Kingdomwww.m-learning.org

Geoff [email protected] Training and DevelopmentLincoln HouseThe Paddocks347 Cherry Hinton RoadCambridge CB1 8DHUnited Kingdomwww.m-learning.org

The main foundation of this was the adoption ofstandards which were firmly embedded in thetechnology world and which extended outsidethe mobile world: hence, we chose web deliveryas the primary means of delivering content andsystems, especially via the emerging XHTMLstandard with its profiles for different platforms.

The limitations of using mobile devices for webdelivery introduced barriers which stronglyinfluenced designs, content and methods ofdelivery. Arguably, the major barrier wasscreen size. Having said that, the devices alsointroduced possibilities in terms of their implicitcommunication and multimedia features.

3 Zones of developmentDevelopment in the m-Learning project can bedivided into several areas including content,navigation structures, a portal, a page-creationtool for user-generated content, a mediaboard,a learning management system (LMS) and aprofile-based course management agent. Thissection presents some of the issues that arisewhen designing web-based material andservices for devices with small screens.

3.1 Learning objectsThe learning objects are focused learning materials,including test items and materials for practisingspecific skills. There is a strong emphasis onvisual richness, colours, animation and soundeffects, based on feedback from earlier trials.These are some of the lessons we learned.

p Learning goals: both Virtual LearningEnvironment (VLE) and stand-alone delivery hadto be considered. Short, sharp pieces of learningthat focused on specific sets of skills and levelswere well received by learners. A ‘quick reference’or ‘focused skill update’ type was more effectivethan a ‘complete package’ approach.

p Device-specific development: despite the use ofdevice-independent standards (eg XHTML) wewere unable to create a rich learning experiencethat could span all our devices. Instead wecustomised our materials for specific device‘families’. Thus richer media was available to ourlearners who were using a Pocket PC (O2 XDA2)than those using the Symbian operating standard(SonyEricsson P900).

p Scripts to span technologies: writing effectivematerials is costly. Although the finishedmaterials did not span devices, the masterscripts did. We focused on specific learningthemes and skills, then wrote the scripts sothat they could be developed into a slightlydifferent learning experience for each technology‘family’. For example, a text-message quiz (SMS),a mini-web quiz (XHTML) or a rich mediamini-web quiz (Flash) could all convey the samelearning message.

p Different solutions for different gadgets:our learners demanded high graphic values,fast download, rich interactivity. These seemimpossible with a tiny screen, a minimalistbrowser and low bandwidth. The solutions wefound were different for each handset, butinvolved combinations of pre-cacheing,Stylesheets, client-side scripting, XHTMLcompression and Flash Apps loading externalXML (for the XDA).

p Browser quirks: mobile web browsers bear astrong resemblance to the early days ofcomputer-based web use, when nothing wasstandard and every browser was full of quirks.XHTML standards are well established, but thereis a wide range of differences in the way that thedifferent handsets can render the pages. Somemanage full Dynamic HTML (DHTML) andclient-side scripting. Others look more like a WAPpage. It was necessary to test everything aswidely as possible, especially when includingclient-side functionality with the page.

3.2 Page-Builder: a case studyGiven the length of this paper, we cannotaddress all aspects of the project, so we havechosen to present a case study of thepage-building software to illustrate some of theissues. The m-learning Page-Builder systemaddressed one of the VLE aspects of thedelivery. It was designed to allow ‘throughthe web’ creation of web pages in apassword-protected environment. It suppliestools which allow the users to make pages basedupon templates which themselves are created inthe same environment. It has a number ofpredefined elements which can be put on pagesto encourage collaboration within theenvironment including lists of links, web logs(blogs) and one-thread debates. These elementscan sit alongside text, images, animations andother ‘static’ page elements.

The system makes use of ‘browser steering’ toformat automatically the pages for a particulardevice. It does have a generic set of configurationswhich try to accommodate most browsers,


the principle being the use of percentage widthson all elements, so users on screens which aretall and thin still only have to scroll up and downand not side to side. Sets of configurations canbe quite easily created for specific browsers/devices and the user agent string of a particularbrowser can be associated with that set. If anunknown user agent string is encountered, thenthe generic is used and a message is sent towhomsoever administers the system.

In many ways, this differs quite radically from theprinciples adopted for the development of thestand-alone learning materials as described insection 3.1. The Page-Builder was created to tryto accommodate any browser and system,building pages ‘on the fly’ for that browser. Insome ways, the challenge of screen size waspassed on to the users of the system when theycreate their own content. However, this is nottotally true. Considerable thought was put intodesigning the tools which allowed the users tocreate the pages and the elements whichinhabited them. Special effort was given todesigning small but meaningful icons for actionssuch as editing, creating, deleting, linking,moving and sending elements/pages. These alsoneeded to be used throughout the system in aconsistent way. The types of operation and thetypes of object had two distinct sets of icons thatcould be amalgamated. So there was an icon fora page, a picture, a debate, etc; and an icon foredit, create, delete, etc. These could then becombined to form edit–page, delete–page,edit–text and so on. In effect, we were creating asmall iconic language for users to manipulate thesystem, which would be, it was hoped,meaningful and also not language-specific.

The main problems we encountered, however,were with screen size. It transpired that the iconshad to be made so small that meanings werehard to convey, especially as it proved verydifficult to combine them at this size. The finalsize which allowed the best fit on most handheldsystems was 12 x 12 pixels. At this size, theamalgamations were so small that they couldbarely be identified. As a result, it was decidedthat they should be used alone, allowing thecontext to convey the meaning in terms of whatthe icon was operating upon. So if the ‘delete’icon was found next to some text, you couldassume that you were deleting that text. Thisproved to be less than successful as the usersfound it quite hard to understand what the iconsmeant when they were isolated in this way.Also it proved quite difficult to create effectiveicons that convey conceptual tasks whichdo not, as yet, have standard accepted imagery –for example, ‘create’, ‘new’ and ‘edit’.

In the version which is currently being used byour research population there is a mix of wordsand icons, depending on where they appear onthe page. It is hoped that we will be able to getfeedback as to the comparative effectiveness ofthese two approaches from the learners at theend of the project.

4 Lessons learnedThis is an ongoing project, and we are continuingto address the points that have been raised in thispaper. The issues that are listed in section 3.1 arethe central focus of our research at this stage,and we are undertaking further investigationsand development work in response to theseconsiderations. Interaction with the learnersthemselves necessarily forms a major part of thisinvestigative work. The creation of complex,media-rich systems on small screens is a challengefor all projects that incorporate mobile devices,as is the breadth and disparity of devicescurrently available. Our research is likely to haveimplications for the market both within andbeyond educational and learning technologies.

ReferencesATG (2002). Microportals: the big need for little portals. White Paper. Athttp://whitepapers.zdnet.co.uk/0,39025945,60041872p-39000536q,00.htm,accessed 14 September 2005.

Childs B (2003). Web development guidelines[Smartphone 2002]. Orange MultimediaOperations.

Collis B and Moonen J (2001). Flexible learningin a digital world. London: Kogan Page.

Lindgren M, Jedbratt J and Svenson E (2002).Beyond mobile. Palgrave.

Melton RF (2002). Planning and developingopen and distance learning. RoutledgeFalmer.

Traffic (2004). Website. At www.traffick.com/,accessed January 2004.

W3C (2004). World Wide Web Consortiumwebsite. At www.w3.org/, accessedJanuary 2004.

WAPF (2001). XHTML mobile profile. La Jolla,CA: Wireless Application Protocol Forum Ltd[now part of Open Mobility Alliance (OMA)].

Popat Stead 169

AbstractThis research paper explores the potential forusing learning objects to implement contentfor mobile learning, and the viability ofreuse of existing e-learning materials ina mobile environment.

By drawing on the results of this evaluationand using tools provided by MOBIlearn (2005) –a pan-European research programme exploringthe use of mobile learning – a viable model fordevelopment and reuse of learning objects isdescribed and a process is proposed forconverting existing e-learning content formobile devices.

Developing new course content specifically form-learning is an expensive and time-consumingprocess which will require at least a reasonableknowledge of the learning content, learning andteaching issues and software engineering. Thereis a multitude of e-learning content in the currentmarketplace and vast sums of money have beenspent on such content development; thereforeit makes sense to look at the possibilities ofconverting existing content into a format thatcould potentially operate on mobile devices,rather than reinventing the wheel.

Much e-learning content is well establishedand has been developed by educators andtechnologists exploiting current researchoutcomes in learning and software development.In order to benefit from this knowledge andexpertise, a research project was undertaken toexplore the ease of this conversion and the stepsrequired in order to convert existing e-learningcontent into a format that would operate on amobile device. To this purpose, the concept oflearning objects was used to implement contentfor mobile learning. Learning objects have thepotential to simplify the conversion process byallowing existing materials to be broken downinto discrete ‘chunks’. Once the appropriateobjects had been chosen, the possibilities forconversion were evaluated.

E-learning to m-learning: an investigation into the potential for content conversionLeon [email protected] to Ufi Ltd16 Foyle RoadLondon N17 0NLUnited Kingdom

As the project results indicate, the majority ofthe e-learning content used did not worksuccessfully on mobile devices. The onlysuccessful example of conversion of e-learningcontent to m-learning was one involving a coursethat had been re-engineered completely tofunction on a mobile device. However, this kindof conversion can be almost as costly as buildinga new course from scratch, so it potentiallydefeats the object of converting existingmaterials for cost-saving reasons. To achievecost-saving, one version of the content shouldfunction on any device used for access.

Although the results of this research did notmeet the initial expectations, the evaluationprocess and tools used allowed some usefulconclusions to be drawn. The actual process thatwas used offers a framework that could befollowed when looking to convert e-learningcontent for a mobile device. The process allowsfor the identification of possible problems(eg whether the content is suited to the mobileformat) and also identifies potential opportunitiesthat could guide further development of thecontent for the mobile device.

Keywordslearning objects, m-learning, mobile devices,reuse of legacy content

1 Learning objectsBefore any actual substantive evaluation ofcourse content was undertaken, it was essentialto establish the parameters for the selection ofthe content and to understand what a learningobject is. Many definitions of learning objectscurrently exist, making the task of defining alearning object – for the purpose of reviewing itsapplicability to support conversion of learningmaterials to a format suitable for a mobile device –practically insurmountable.

Rodin 171

Following a literature review of learning objects,learning object research and the use of learningobjects for mobile learning, several differentdefinitions were explored, but all were slightlydifferent. This made the task of selecting adefinition to support the main purpose of myresearch very difficult.

Welsch (2002) discusses the varying definitionsof a learning object and suggests that it dependson whom you are asking. The research for thisproject required a definition which offered a basisfor deciding how the e-learning content to bereviewed could be broken into more manageablechunks that would potentially be suitable forconversion to function on a mobile device.Consequently, a combination of severaldefinitions appeared to be more suitable for thisproject than any one definition proposed thus far:some definitions were far too broad and couldnot therefore make the task of selecting potentialobjects straightforward; and others were toonarrow, which would have excluded somecontent that had the potential for reusability.The combination of the definitions from Rehakand Mason (2003) and Brown and Hoyle (2003)offered a definition (Rodin 2004) that wassuccinct, but not too restrictive: ‘Discreteelements of learning content that meet adefined learning objective, and are possiblyindependently assessable, which may take theform of text, graphics, video, stills, animations,diagrams or audio’.

2 Research undertaken

2.1 Research questionThe conclusions drawn from the literature reviewintimated a more fundamental question: aremobile devices suitable for learning purposes?Although the project is not structured to explorethis question in great depth, the conclusionsof the research outlined below may help toexplore some of the issues involved and offersome helpful pointers on the use of mobiledevices for learning.

The analysis that was proposed following theliterature review offered a focused objective forfurther research; by applying the concepts oflearning objects to existing e-learningcourseware, is it possible to migrate coursewareto mobile devices? So the research question thatwas explored can be described as: in migratingexisting e-learning courseware to mobiletechnologies, can learning objects help?

2.2 Device and tools used to supportthe researchWith the support of MOBIlearn and Ufi Ltd,10 pieces of e-learning content were selectedfor evaluation. The intention was to review theprocess of converting the materials from aconventional PC-based e-learning environmentto a mobile m-learning environment on theHP iPAQ. The reason for selecting the iPAQ wastwo-fold. First, the iPAQ was equipped with thecurrent standard Microsoft Windows mobileoperating system and also appeared to havesoftware available that matched the format ofsome of the selected materials, such asMacromedia Flash. Second, Keegan (2001),describing the results of an EU-funded project‘From e-learning to m-learning’ notes that aniPAQ was cited by learners testing somem-learning applications as their preferred device.It scored highest in nearly all of the evaluationquestions, suggesting that it was a more usefuland functional device for use in learning than amobile phone or a Smartphone.

MOBIlearn had also made available severalmatrices that had been developed to support thedivision of learning content into learning objectsand the elicitation of metadata for the identifiedobjects. The first matrix, the Learning ObjectDescription Matrix, was designed to assist in theprocess of identifying any potential learningobjects. The second, the Metadata Matrix, listedspecific potential metadata elements that coulddescribe a learning object, and these elementswere developed as a subset of the IMS LearningResource Metadata, which is now effectively astandard (Mangiavacchi 2003). The subset wasagreed by MOBIlearn and included only thosepieces of information that were considered to beuseful for m-learning. The MOBIlearn matriceswere applied separately, as each attempted tosupport different aspects of learning objects.

2.3 Content review processBy working through the process of convertingthe content from an e-learning to an m-learningenvironment for the first piece of content, thevarious stages of the evaluation process wererecorded. The capturing of the flow of tasksenabled a flow diagram to be developed, and thisthen gave a foundation and framework for theevaluation of the remaining content.

It was crucial to focus the evaluation onthe necessary tasks to ensure that therewas no deviation from the project plan.Any additional tasks or research that wereidentified were assessed to establish theirappropriateness to the overall research.


If the additional tasks added value to the research,then they were explored on an ad hoc basis; if,however, it was decided that no value would beadded, the task was recorded along with thereason for the decision not to continue with it.

Seven stages were identified, together comprisingthe required process for evaluating whether thee-learning content could be transferred tofunction correctly on a mobile device.

p First, the content was loaded onto a PCand the ‘look and feel’ explored to identifyhow it functioned in a conventionale-learning environment.

p The second stage was to evaluate the actualcontent, identify possible objects, review thestructure of the content and the media used, etc.

p The third stage was the application of theMOBIlearn matrices to the content.

p Stage 4 focused on the structure of the contentand whether the file directory, coding and designled to the simple elicitation of individual objects.

p Stage 5 comprised the attempt to load the contentonto the iPAQ using the Microsoft synchronisationsoftware and any plug-ins that were thought tobe appropriate (eg Macromedia Flash 6) if thecourse was built using this technology.

p The sixth stage consisted of recording anydifficulties, observations or suggestions forimprovements to the content; if the conversionwas successful, the content was tested and itsfunctionality evaluated using a series of setquestions to assess the suitability of the contentfor the mobile device.

p Stage 7 ensured that any outstandingobservations or issues were recorded, beingintended as a review of the previous six stages.Any final conclusions or thoughts were then noted.

3 ResultsThe results were divided into each of the sevenstages and the content was evaluated at eachstage. None of the content that was not divisibleinto learning objects (as defined previously)would convert to or function on the mobiledevice. Several pieces of the content that wereevaluated had been originally designed tofunction on a mobile device, or had beenspecifically re-engineered to function on a mobiledevice and therefore converted without difficulty.The evaluation of such content was, however,useful to review the suitability of the seven-stageconversion process and also acted as the controlcontent for the experiments.

The first matrix used, the Learning ObjectDescription Matrix, did help with theidentification of learning objects, but theMetadata Matrix was not really useful for thisresearch, as it was designed to help in specifyingparticular metadata for each learning object, andthis was outside of the scope of this project.

The results were disappointing in that the onlycontent that successfully and fully converted tofunction on the mobile device was that designedspecifically for mobile devices; the only exceptionto this rule was a a single piece of content thatwas, in fact, a video clip that operated throughthe Windows Media Player software on the PCand the iPAQ.

4 Conclusion and discussion

4.1 The main researchThis project has reviewed the current researchinto learning objects, their use in mobile learningand the use of mobile devices for learning. Thesubstantive research evaluated the potential ofconverting e-learning course materials,developed for conventional PC-based learning, sothat they could function in a mobile environmenton a mobile device, the HP iPAQ.

Although the results of this research did notmeet the initial expectations, the evaluationprocess and tools used allowed some usefulconclusions to be drawn. The actual stagedprocess that was used offers a framework thatcould be followed as an initial step when lookingto convert e-learning content to a mobile device.The seven-stage process allows for theidentification of possible problems, such aswhether the content is suited to the mobileformat; it also identifies potential opportunities thatcould guide further development of the contentfor the mobile device. Following the use of thisprocess, it was found that two of the stages didnot generate any tangible results; based on thisexperience, it may be more appropriate to reducethe whole process to five stages:

p Stage 1: load content into the e-learningenvironment, evaluate the e-learning materialsand attempt to divide into objects

p Stage 2: apply and evaluate the MOBIlearn matrices

p Stage 3: evaluate and review the code, coursedesign and file structures

p Stage 4: transfer the content to the mobiledevice, evaluate the functionality of the contentand apply the test questions

p Stage 5: review the previous stages and recordany conclusions or final observations.

Rodin 173

The results also demonstrated that the hardwareof the mobile device used and the operatingsystem were more specific than first anticipated;much of the difficulty in executing the transferredcontent was due to software incompatibility. Thevarious plug-ins that were downloaded did helpwith some of the content, but even these werenot as useful as the manufacturers’ descriptionssuggested. The Macromedia Flash 6 plug-inworked for some of the content, but it wasnecessary to download a supplementaryfreeware Flash 6 enhancement (not produced byMacromedia) to add to the functionality of theoriginal plug-in.

When using a mobile device for learning, thecontext of the content in some circumstancesdoes not support the objectives of the courseunless the learner uses the mobile device as anaid instead of a stand-alone tool. For example,there was one course module in which thescreenshots and content referred to a differentversion of the application from that available onthe mobile device (Microsoft Word and PocketWord). In this situation, as the screenshots weredesigned to give the learner an appreciation ofhow to carry out various tasks and this did notcorrelate to the version of the application on themobile device, the learner could either get veryconfused and frustrated, or could only use themobile device as a support tool while practisingthe application on a PC.

This idea of a support tool to some extentdefeats the objective of mobile learning, since tolearn from the course, a learner would have to beworking at a PC and therefore would not bemobile. However, this does suggest a differentuse for the mobile device – as a tool to supporte-learning, as part of a blended learningapproach. This would, however, be an expensivesupport tool and on this basis is not reallypractical, as Help programmes already exist inmost similar software.

Although the research carried out did not reallylook at ways that the mobile devices could beused to support learning, some of the contentreviewed was structured as small stand-aloneunits. It can be concluded that this type ofcontent could be used to support a blendedapproach, allowing a learner to take part oftheir course with them on a train or bus tobrowse at leisure.

The screen size of the mobile device could beconsidered a problem, especially for learners withvisual disabilities; however, as all the contentevaluated used audio files, some difficultiescould be overcome. This suggests that becauseof the limitations of the size of the mobile device,the use of multimedia, especially audio, is really anecessity to increase the usability of the courses.

4.2 Learning objects – a possible solutionFor this project, a definition of learning objectswas chosen (see section 1) which was based onan interpretation and understanding of the issuesinvolved. After the courseware evaluation hadbeen conducted, it became clear that this definitionwas not significantly better in describing alearning object than any of the definitions alreadydescribed in the literature review.

Nearly all the definitions reviewed attempted toidentify an all-encompassing concept of a singlelearning object. Learning objects are notnecessarily isolated pieces of learning material;they are building blocks for e-learning materialsand therefore, as with children’s Lego bricks, donot necessarily have to come in one shape or size.

During the testing and evaluation of learningmaterials, it was found that some of the contentevaluated was designed using various units thatcould potentially be considered as learningobjects according to the definition used for thepurposes of this project (see section 1). They allhad similar features, such as an interactive menu,information on the use of the course and use ofencapsulated software objects (eg video andaudio files). These features are separate from theindividual units, and do not represent the contentof the course: therefore they could be designedas reusable objects.

These particular objects support the functionalityof an e-learning course and consequently supportthe learning that may take place, so they can beconsidered as separate software objects thatsupport learning. This description suggests anexpansion of the current view of what a learningobject is. If a learning object is a building block forconstructing an e-learning course, a learningobject cannot only be the content within acourse, but must be a reusable software objectthat supports learning.

In conclusion, there is not one type of learningobject; instead, a learning object is a softwareobject that can be used to support learning.This also takes the concept of learning objectsfull circle and allows a course built using thisapproach to take advantage of all thetried-and-tested developments of object-orientedsoftware engineering, which is a much moreestablished IT field than e-learning and a keycontributor to the original development of thelearning object concept.

Some learning objects may not be usefulin isolation, but metadata can be used toidentify the potential use of an object.Using this approach, learning objectscan be mixed and matched appropriately.


For example, Netg could create: a single Netg‘menu’ object that would ensure that any coursedeveloped for Netg will function in the sameway; and a single Netg ‘help’ object that wouldoffer instruction on how to use every Netgcourse. The Netg developer could, however, alsouse ‘content’ objects designed by othere-learning companies and the finished coursewould not look any different to the learner orcustomer from any other Netg course. On thisbasis, the cost benefits are apparent and realinteroperability may be possible.

The idea of this learning object model may seemrelatively simple, but no definition or approachidentified in this research offers the samepotential for interoperability, while maintainingcontrol of the finished product for the companyor organisation that puts the resultant coursetogether. The various other approaches tolearning objects – other than Sloep’s (2003)suggestions – arise from the disparate definitionsthat currently exist. If one definition can permitany organisation to build their own version of acourse, and ensure that the objects they use arereused from another source or can be reused tosupport a different context without any redesignof the original object, then this definition has thepotential to become effectively a standard.

Extensible Markup Language (XML) has thepotential to support the development of a learningobject standard. The use of Scalable VectorGraphics (SVG) and the concepts behind XMLcould not only make the interoperability of learningobjects easier, but also open the door for contentto operate on any device, irrespective of the screensize or hardware. Since XML standards exist (albeitcurrently under review), a learning object standardthat standardises the model described above andincorporates the XML standards would makefeasible a complete learning object implementationthat is fully interoperable.

Sloep’s ideas (2003) on a modelling languagefor learning materials – Educational ModellingLanguage (EML) – also take a different approachto learning objects. His ideas for the use of EML(which takes advantage of the features of XML)and the separation of the software from theobjects is almost the same as that suggested inthis paper, in that EML has some objects thatrepresent the materials and some objects torepresent the other aspects of a course. The useof EML could enhance even more the modeldescribed above and so can be identified as anopportunity for further research.

Functional m-learning products are currently notwidely developed and used. The use of the termm-learning is perhaps one of the reasons thatinteroperability has not yet been fully achieved.

If mobile technologies and devices wereconsidered in an integrated approach toe-learning, then some of the interoperabilityissues might be resolved during the designphase of any e-learning development.

It is essential in software engineering to capturethe necessary requirements as early as possible.Current e-learning courseware does not appearto take account of the use of the content onmobile devices; it is therefore likely that such ause was not included in the specifications of thecontent when it was being designed. As the useof mobile devices – particularly smart mobilephones – is increasing at a phenomenal rate,when a content designer is specifying a newdevelopment of e-learning materials, he or sheneeds to include use on mobile devices withinthe specifications. In this way, it ensures that theactual use of the materials is irrelevant, whetherit be at a desk in a learning centre, at home on alaptop, or on the bus using a mobile device.

ReferencesBrown J and Hoyle D (2003). Learning objectsand materials design. Unpublished document,Learning Quality and Standards Committee, Ufi.

Keegan D (2001). The future of learning:from eLearning to mLearning. London: Taylorand Francis.

Mangiavacchi F (2003). Sfera metadata.Unpublished paper written for MOBIlearnWork Package 3.

MOBIlearn (2005). Website.At www.mobilearn.org, accessed10 September 2005.

Rehak D and Mason R (2003). Keeping thelearning in learning objects. In A Littlejohn (ed.)Reusing online resources: a sustainable approachto eLearning. London: Kogan Page.

Rodin L (2004). E-learning’s little brother: aninvestigation into current technologies form-learning, their availability and content designconsiderations for mobile devices. MSc thesis:Middlesex University, UK.

Sloep PB (2003). The language of flexible reuse.Reuse, portability and interoperability of learningcontent. Open University of the Netherlands.At www.ou.nl/open/psl/Publicaties/LanguageOfFlexibleReuse2003.pdf, accessed20 May 2003.

Welsch E (2002). SCORM: clarity or calamity?Online Learning Magazine, 9 November issue.At www.onlinelearningmag.com/onlinelearning/magazine/article_display.jsp?vnu_content_id=1526769, accessed 28 May 2003.

Rodin 175

AbstractThe objective of this paper is to document thework of Intel IT Innovation Centre as related tothe skoool* educational initiative, now available ingeographies throughout Europe and beyond. Thispaper combines information from internalsources and respected academics andpractitioners in m-learning and pedagogyworldwide. The initial skoool offering ispresented, identified limitations are documented,as is the redesign work and innovations thatwere implemented to address these limitations.Current analysis of the work is documented andrecommendations and strategy for going forwardare outlined. The paper concludes with a generaloverview of the future of wireless technologywithin education and how Intel IT InnovationCentre intends to progress towards the ultimategoal of ‘educational content anytime, anywhereon any device’.

Keywordsskoool, peer-to-peer, PDA, Smartphone

1 BackgroundIntel Ireland, and indeed Intel worldwide, isinterested in education and technology ineducation for a number of reasons.

It is essential to the microprocessor market thatnew usage models are continually emerging andevolving, and education is among the maindrivers in this area. Funk (2004) tells us that asthese new usage models mature, and networkeffects and positive feedback develop, the aim isto reach a critical mass of users. To this end,there is an opportunity to demonstrate thepotential of new technologies.

From another perspective, Intel is in a uniqueposition in Ireland to leverage a historicallyacademically strong nation, innovation-friendlygovernment and a talented, educated workforce.

Educational content anytime, anywhere on any deviceGerard [email protected] IT Innovation CentreIntel IrelandLeixlipCounty KildareRepublic of Ireland

In recent years, there has been a significantdecline in students opting for and excelling in thesciences and mathematics in Ireland and the UK.According to Lord May (President of the RoyalSociety), ‘The flow of talent now looksthreatened right at the start of the pipeline, inschools. This is particularly true in the physicaland mathematical sciences’. It is crucial for acorporation such as Intel that we continuallyaddress the requirements for high-levelgraduates in these areas.

2 Case study – skooolskoool is the generic project name for ane-learning initiative under way at Intel IT InnovationCentre. The key strategy of this initiative focuseson bringing learning content, interaction,visualisation and real-world application of thisknowledge to the learner. It strives to engage thelearner and make the learning content accessiblethrough the use of appropriate technologies. Theinstructional approach taken focuses on wheretechnology can support learners to understandand explore difficult and abstract topics.

Working in public/private partnerships withgovernments and leading private sectorcompanies, skoool is free to the user and itstarget demographic is students (12–18 yrs),teachers and parents. It is well established inIreland and is currently being introduced in theUK [London Grid for Learning (LGfL)], Swedenand other geographies.

2.1 skoool.ie*skoool.ie was the first-of-its-kind e-learningresource in Ireland when it was launched in2002. It provides students withcurriculum-relevant resources to complementtheir traditional studies in examination subjects atJunior and Senior Cycle levels, while it providesteachers with high-quality content and tools.

Smyth 177

Currently, approximately 80% of allsecond-level examination students haveexposure to the portal.

2.2 skoool.co.uk*skoool.co.uk was launched in the London Grid forLearning (LGfL) in March 2004. This offering istailored to the UK curriculum, specifically Keystages 3 and 4 of the science and mathematicssubjects. The redesign and innovationsincorporated into this offering reflect learningsfrom our earlier work on skoool.ie. There arechannels for teachers and parents which providesupport content.

2.3 skoool.se*skoool.se is a pilot project that has been underway for some months in Karlstad, Sweden. Itspurpose is to evaluate the use of rich mediae-learning with a view to developing a model thatis scalable and extensible across Sweden.

3 skoool: initial challengesskoool was designed to help demonstrate anddrive the adoption of broadband for learning andteaching. However, the roll-out of broadband hasbeen slower than anticipated and this led tolimitations being identified in the accessibilityand usability of content in the initial product.High-quality courseware was found to be slow onnon-broadband connections. In the classroom, itwas found inefficient that the portal wasaccessed by each individual; while in the home, itwas considered expensive to stay online for longperiods. Significantly, it was also noted that manyusers were also mobile phone – and to a lesserextent – PDA users. This paper will demonstratesolutions developed to address these problemsincluding peer-to-peer caching technologies andmobile device learning models.

Furthermore, according to Pham, Schneider andGoose (2000), the phenomenal growth of mobiledevices and wireless technologies, as well as theneed for ‘anytime, anywhere’ learning, hasresulted in a major focus towards mobile andubiquitous computing. Major technologicaladvancements, combined with the wide range ofmobile and handheld devices that are nowbecoming available at price points feasible for thestudent/educational arena, make it imperativethat we harness these resources andcontinuously rework our educational approachin delivering pedagogically sound content toskoool users.

4 skoool: initial innovationsTo resolve issues identified in feedback and inorder to leverage recent technologyadvancements, a significant amount of redesignand development work has been implementedon selected geographies, while other work is atproof-of-concept stage. Feedback thus far haslargely been positive, with some areas identifiedfor further development.

4.1 Introducing myskoool*:high-quality coursewareavailable offlinemyskoool was primarily developed to address theissues of low bandwidth, intermittentlyconnected users and multiple users on a localarea network (LAN); for example, a schoolnetwork. Using peer-to-peer (P2P) technology,the user can download the myskoool application,and subsequently courseware, from either theskoool portal or the nearest hosting peer. Thecourseware may then be viewed offline withoutthe latency associated with a low-bandwidthconnection. Also, for a school network, an itemof courseware need only be downloaded once –it will then be propagated through the LAN toother users. At home, the user may downloada number of courseware items during a periodof low internet activity and then use theseitems offline.

4.2 Content for handheld devicesWork is ongoing in the implementation andprototyping of online and downloadable learningmaterial for handheld devices such asSmartphone and PDA.

Downloadable content

Selected courseware is currently available atwww.skoool.co.uk for download to PDA.Proof-of-concept work has been successful fordownload of content to Smartphone (seeFigure 1 opposite).


Figure 1 Screenshots of Smartphone content

Content is initially downloaded to PCand transferred to the handheld deviceusing integrated infrared or Bluetoothwireless technology.

Online content

Customised online content is available forviewing via Wireless Application Protocol (WAP)and Smartphone.

Types of content for PDA and Smartphone

The learning content for PDA is similar in formatand structure to the content on the learningportal. It has, of course, been modified inconsideration of the constraints of thehandheld device.

The mobile learning content currentlyavailable comprises:

p LearnSteps* – instructional learning objectsuseful for learning and revision

p LearnSims* – simulations allowing the user tointeract and apply learning

p Study Notes – reminder or revision tool usingprimarily text-based content.

SMS Messaging

An SMS messaging facility has been in operationat www.skoool.ie over the last 2 years and hasproved very popular with students. The usersubmits their mobile phone number to the facilityon the skoool site and selects their areas ofinterest; for example, study tips, competitions.Over the course of the school year, a number ofSMS campaigns are organised for participatingusers. To date, examination tips have been themost popular, with 45,000 messages being sentout in May 2004 to Irish pre-examinationstudents. Notification of new content is anotherway in which this technology is being applied.

5 skoool: ongoing critical analysisFor this project to be a success, Intel lTInnovation Centre acknowledges that it isimportant to employ an iterative approach to thedevelopment of the skoool portal. Design,development and implementation should befollowed by feedback and evaluation, leading tofurther requirements being identified, and thusbeginning the cycle all over again. With this inmind, educational specialists within Intel lTInnovation Centre have conducted user surveysand appraisals on work to date. In-depth usagepattern and demographic data is also closelymonitored by business and marketing specialists.

5.1 Evaluation findingsA recent evaluation of the skoool.co.uk portal,conducted in May 2004, made the followingconclusions with respect to handhelddevice usage.

Content

p Very clear benefit to learners that there is aportion of the portal devoted to m-learning.

p The content provides the learner with a lot ofuseful information that is well structured, clearand pedagogically sound.

p The content provided leverages mobility andinteractivity through the use of simulations.

p Progress can still be made in leveraging other keybenefits of the mobile device. Not enoughopportunity currently exists for the learner toengage in the creation of ideas and knowledge.

p The integration strategy allows the user tohave a similar learning experience to onlinewhile mobile.

p Examination Centre content may need to bemade more concise – presently too wordy.

Devices

p The handheld device is currently being usedprimarily as an administration and organisationaltool, rather than a tool to support learning.

p Navigation and readability on devices wererated as OK.

p Survey respondents suggested that the mobiledevice lends itself to reflection, exploration andinstruction in that order.

Smyth 179

5.2 Usage statisticsAs shown in Figure 2, our experience withskoool clearly indicates that there is a strongdemand for quality technology-based learning24 hours a day, 7 days a week. Widespread useof mobile devices will extend this potentialsignificantly further.

The proliferation of mobile devices will surelyincrease usage throughout the day, but especiallyat peak times when students are ‘on the move’on their way to or from school or on field trips orresearch work (Areas A and B in Figure 2).

6 skoool: strategy forgoing forwardBased on our experience of developing theaward-winning learning service www.skoool.ieand more recently www.skoool.co.uk, Intel ITInnovation Centre has researched, experimentedwith and developed a wide range of learningapproaches for advanced handheld computingdevices. The question we must ask is, ‘Whichhorse to back?’ Technology adoption has alwaysbeen a highly unpredictable phenomenon withtwo highly unpredictable vectors: timing anddirection. Based on this uncertainty, we havechosen an options-based approach, developinglearning technologies for a range of devices andconnection models.

6.1 Which learning and technologymodels will prevail?The notebook computer remains the mostpowerful and flexible computing and learningdevice. The PC is the workhorse platform formyriad tasks including report writing, research,data analysis and the use of advancedmultimedia learning resources. Mobility is key,with the proportion of notebook computerpurchases continuing to increase and Wi-Fitechnology capabilities now included as standard.

With this as a given, we see a model whereusers utilise a range of low-cost handhelddevices that integrate with and work alongsidethe laptop, which sits at the centre of thesedevices and acts as a hub, collecting anddistributing data from these devices andthe internet.

The devices will include:

p Tablet PCs

p handheld PDAs

p multimedia-enabled Smartphones

p digital still and video cameras

p MP3 music players

p digital gaming devices.

Connection models will include fixed andwireless internet, wireless mobile phoneconnection, synchronisation facilities andBluetooth and infrared connections. Occasionally,connected models will continue to be employedwhere users take advantage of high-speedconnection at hotspots and transfer data asappropriate to other devices for offline usage.


Figure 2 Activity level by hour of day (averaged over week)

6.2 A design recommendation forfuture mobile device learningThere are two aspects to such a designrecommendation; how best to use the device asa learning tool based on its capabilities; and whatsupport context would best support this design.

Based on the evaluation work outlined earlier,our educational specialists propose an outlinedesign focusing on the constructivist principlesof reflection and collaboration. The corecapabilities of the mobile device identified tosupport this design are mobility and capability toshare information.

The mobile device as a reflective andcollaborative tool

Much of the curriculum-driven learning in class isbased on instruction and to varying degrees,encourages a behaviourist approach to learningand teaching within the classroom. With theproposed design, the mobile device would beused to support reflection and collaboration,allowing learners to build their ownunderstanding of a topic, thus encouragingconstructivist thinking.

The reflective component of the design wouldinclude worksheets based on semanticorganisation tools such as a concept-mappingtool designed to help learners analyse andorganise information. Nesta Futurelab (2004) hasshown that the effectiveness of computer toolsin such scenarios appears to be enhanced whenused by learners in pairs or in groups. In line withthis, the collaborative aspect of the design isbased around learners sharing worksheets andconducting peer reviews of each other’s work.

Furthermore, the design proposes that learnersmay upload their reflective and learningcontributions to a collaborative area on an onlinelearning portal for use and development byothers. This would leverage the capabilities ofboth the PC and mobile device and, as Bransford,Brown and Cocking (1999) outline, this couldresult in an ‘intellectual partnership with thecomputer in that the whole of the learningbecomes greater than the sum of the parts’.

7 Looking to the future

7.1 The digital home/school/officeThe arrival of the pervasive digital mediaenvironment in the school, home and officeseems imminent. The key factors contributing tothis include:

p wider availability of broadband

p increased interest in home networks

p increased sophistication of personal digital devices

p ever increasing libraries of digital content.

According to Otellini (2004), this will herald a newera of consumer electronics that will be ‘equallyexciting to the rise of the PC’. The success ofnew consumer devices will hinge on some keyattributes, including:

p Wi-Fi (wireless connectivity)

p Veri-Fi (security capabilities to safeguardconsumers’ data)

p Hi-Fi (high-quality)

p Ampli-Fi (the ability to move data across networks)

p Simpli-Fi (intuitive user interfaces).

Though his comments were made in relation tothe digital home specifically, it is worthconsidering them in the context of the schoolenvironment and projecting how they will applyin that setting.

7.2 The mobile internetMobile computing has taken major stepsforward with the wide adoption of Wi-Fi, Intel®CentrinoTM mobile technology and the upcomingWiMax technologies. However, mobile internetservices over next-generation networks havebeen relatively slow to gain ground in Europe andthe US when compared to Japan and Korea.

Among the reasons for this, arguesFunk (2004), are:

p The mobile Internet is more disruptive forWestern firms than for Japanese and Korean firms.

p Western firms focus more on business usersthan their Japanese and Korean counterparts,while the market for the mobile internet seemsto be young users.

p The creation of a micro-payment system by NTTDoCoMo* for its i–mode* service, which made iteasier for content providers to collect revenues,thus increasing the amount of content available.

p Higher subsidies are being paid by Japanesefirms to retail outlets to sign up new subscribers.

Smyth 181

Whatever the reasons, bearing in mind thatsuccess in Japan and Korea is clearly due toyoung users, it would be wise to learn from theirsuccess and plan on how this technology can beleveraged in education when it gathersmomentum in Europe.

Some of the findings thus far include:

p Young people are the initial users.

p PC internet penetration is not related.

p Ringtones and screensavers are the initial market.

p A micro-payment system is needed for content.

p We should not over-hype the mobile internet.

p The mobile internet, whether based on WAPor cHTML, is a superior technology to SMS.

p The mobile internet is not just a service – it is anew network industry requiring service, content,software providers and device manufacturers.

8 ConclusionIn this paper, we have looked at the skooolproject and the iterations that it has been throughto date in the pursuit of delivering media-richcontent tailored appropriately to the users for itsintended geography. We have examined thedifferent models of delivery available currentlyand have anticipated what these may be in thefuture and how best we can use them.

Successes and failures identified thus far havebeen presented in an effort to share knowledgeand thus increase overall learning. This, webelieve, is the best strategy for moving forward.

ReferencesBransford JD, Brown AL and Cocking RR (eds)(1999). How people learn: brain, mind,experience and school. Washington DC:National Academy Press.

Funk JL (2004). Mobile disruption: thetechnologies and applications driving the mobileinternet. Hoboken, NJ: John Wiley and Sons, Inc.

Hamilton P (2004). Personal communication.Intel IT Innovation Centre.

Kynäslahti H and Seppälä P (eds) (2003).Mobile learning. Helsinki: IT–Press.

Nesta Futurelab (2004). Literature reviewin thinking skills, technology and learning.At www.nestafuturelab.org/research/reviews/ts01.htm, accessed 21 June 2004.

O Donovan A (2004). Personal communication.Intel IT Innovation Centre.

Otellini P (2004). Intel enters a new eraof consumer electronics. In Proceedings ofConsumer Electronics Show, Las Vegas,January 2004.

Pham TL, Schneider G and Goose S (2000).A situated computing framework for mobileand ubiquitous multimedia access using smallscreen and composite devices. In Proceedingsof the 8th ACM International Conference onMultimedia. New York: ACM Press.


AbstractThis paper describes how Short MessagingSystem (SMS) was identified and used as a formof ‘mobile scaffolding’ in an experiment tosupport first-year university students, addressingissues related to supporting student retention.The research and issues underpinning therationale for the experiment are summarised,including new data related to types of messagingtechnology used by students. Results from theexperiment are presented. The paper concludeswith a discussion of issues raised, including thesurprisingly low rate of student participation inthe evaluation process.

Keywordsblended learning, student retention, mobilelearning, e-learning

1 IntroductionUK universities are increasingly concernedwith issues of student retention (EESC 2001;May and Bousted 2003) and in the light of the UKgovernment’s policy aims of having 50% ofschool leavers attending university by 2010(Sutherland 2002), the different routes thatstudents can take to achieve this will increase.The corresponding increase in diversity ofbackgrounds and experiences will have an impacton the demands for retention support (Cox andBidgood 2002). At the same time, there is acontinuing increase in the deployment ofe-learning tools and strategies in theuniversity sector.

2 BackgroundGrounded Theory research was undertaken atKingston University into student perceptions ofthe benefits of e-learning (Alsop and Tompsett2002). Key themes identified included somewhich were less to do with technology or tasks,

Blended learning, mobility and retention:supporting first-year university students with appropriate technologyAndy [email protected] of Computing and Information SystemsKingston UniversityKingston upon Thames KT1 2EEUnited Kingdom

and more to do with students’ notions of timeand space. The same technique was proposed assuitable for identifying the means by whichstudents use communication technologies tocollaborate (Stone, Alsop anf Tompsett 2003); thiswas tested by asking a group of 20 second-yearstudents to undertake an exercise whichdescribed a good and bad experience ‘…ofusing communications to collaborate and workin groups’.

SMS and Instant Messenger (IM) applicationspolled the same amount of ‘good’ responses(50%), with e-mail and face-to-facecommunication mentioned equally as secondarymodes. However, the most frequently mentionedtechnology in relation to ‘bad’ experiences wase-mail (40%), followed by face-to-facecommunication (25%) and voice calls (20%).Core concepts have been identified as based onissues of time, location and effectivecommunication. Given the (current) reliance onfixed connections for IM communicationbetween students (it could be argued that thedistinction between SMS and IM will blur asmobile handset capabilities evolve), it was feltthat this exercise highlighted the potential forusing SMS to support students as acomplementary technology to those in use bylearning providers at this time.

To further test this hypothesis, a survey group offirst-year students was asked if they would beinterested in SMS services to support them.Without exception, all students said they wouldbe interested in using such a service (Stone2004). Furthermore, the types of service theyexpressed an interest in were consistent withother studies (Divitini, Haugalokken and Norevik2002; Lehner and Nosekabel 2002), as well asfindings from work done in parallel on studentretention across Kingston University as a whole(May and Bousted 2003).

Stone 183

3 The experimentHaving identified candidate services andtechnologies, an experimental SMS service wasoffered to first-year students to support them onone of the core modules in the first semester. Itwas publicised in many ways during the start ofthe degree programme: during induction-weeklectures; at the start of the module; and byannouncements made via Blackboard, thelearning management system (LMS) currentlyused at Kingston University. Although it was anentirely voluntary (opt-in) process, 115 studentsregistered within the first 6 weeks. The nature ofthe messages changed over the course of thesemester: from timely, relevant informationprovision (‘where and when’; see Figure 1) toSMS ‘tips’ on where to find information withinthe e-learning system and elsewhere.

Figure 1 SMS alert received by student

Figure 2 Student takeup of SMS over time

It was noticed that student registration with theSMS service took place over a longer time periodthan expected (see Figure 2). There is evidencethat supports two hypotheses as to why thiswas so:

p the role of peer recommendation through socialinteraction and usage

p shared experience of a technical issue(a prolonged network outage on campus), whichmade further benefits of the SMS systemapparent to students, as the two main formalconduits for information dissemination(ie Blackboard and the university e-mail system)became inoperable and inaccessible forseveral days.

4 Feedback and discussionFeedback was requested through e-mailquestionnaires. Students were requested inlectures, via announcements on Blackboard andvia SMS to complete these. The number ofresponses was disappointingly low (only sevenwere received), but this has been experiencedelsewhere (Baty 2003). Nevertheless, there weresome questionnaire responses that areconsistent with informal responses given bystudents in face-to-face encounters.

Students believed that they came to be aware ofthe service more via Blackboard than throughlectures, with reminders for deadlines(particularly on assessments) being the mainreason. This type of SMS message was alsofound to be the most useful to students, closelyfollowed by notification of changes. Given theproblems experienced with Blackboard, it isreasonable to speculate that this acted as a driverfor students to use the trial service in theabsence of any formal online informationprovision system (be it e-mail or Blackboard).

When asked what services they would have likedto have seen, notification of updates onBlackboard was the most popular, closelyfollowed by a roll-out of the pilot service to allfirst-year modular courses. There was also arequest made for notification as to how to collectmarked coursework from the departmentaloffices for the first time. Since some of the earlycoursework for the module in the experiment fedinto subsequent assessments, it was felt thatthis should have been done so that studentswere given start-to-finish support for their firstpiece of assessed work.


There is some evidence to suggest that studentsdo share the information that was relayed tothem at short notice via SMS, but in a variety ofways. They are just as likely to inform their peersface-to-face as they are by a voice call or SMS.Due to the low number of respondents, it is notfelt that this result can be used in isolation.However, from this and informal face-to-faceconversations with students, it would seemuseful to bear in mind that there are a variety ofmodes of communication that can be (and are)used by students to communicate with theirpeers. This should be remembered whenconsidering novel support mechanisms,particularly where technologies may be thoughtof as able to replace, rather than complement,existing modes of communication. As indicatedabove in section 2, the need for ‘appropriatetechnology’ for the individual student, rather than‘information technology’ for an assumed,homogeneous set of users, is somethingwhich ought to inform designers of blendedlearning strategies.

Finally, although students expressed aunanimous interest in using a full implementationof the pilot service, those against any form ofpayment for the service are in a 2:1 majority, withmany comments alluding to the expectation thatuniversities should offer such support free ofcharge. In addition to this expectation, concernsalso exist in the minds of students about issuesof security and trust (Carroll and Hartnell-Young2004; Sørensen 2004). These are issues whichwould have to be resolved if such a service wereto develop into a fully operational support systemfor students.

ReferencesAlsop G and Tompsett C (2002). Grounded Theoryas an approach to studying students’ uses oflearning management systems. Associationfor Learning Technology Journal, 10(2), 63–76.

Baty P (2003). Student survey plans falls apart.Times Higher Education Supplement, 21 March.

Carroll J and Hartnell-Young E (2004). Availablejust enough: the mobile phone as a safety netfor parents and their teenage children.Paper presented at The Life Of Mobile Data:Technology, Mobility and Data SubjectivityConference, University Of Surrey, 15–16 April.

Cox B and Bidgood P (2002). Wideningparticipation in MSOR [Mathematics, Statisticsand Operational Research]. MSOR Connections,22(1). At http://ltsn.mathstore.ac.uk/newsletter/feb2002/pdf/widening.pdf

Divitini M, Haugalokken OK and Norevik P (2002).Improving communication through mobiletechnologies: which possibilities? In Proceedingsof the IEEE International Workshop on Wirelessand Mobile Technologies in Education.Ltechnology in relation to ‘bad’ experienceswas e-Alamitos, CA: IEEE Computer SocietyPress, 86–90.

Lehner F and Nosekabel H (2002). The role ofmobile devices in e-learning – first experienceswith a wireless e-learning environment. InProceedings of the IEEE International Workshopon Wireless and Mobile Technologies inEducation. Los Alamitos, CA: IEEE ComputerSociety Press, 103–106.

May S and Bousted M (2003). Retention projectfinal report. Unpublished internal paper, KingstonUniversity, UK.

EESC (2001). Higher education: studentretention. Sixth Report of the Education andEmployment Select Committee, UKGovernment. HMSO.

Sørensen C (2004). Research issues in mobileinformatics: classical concerns, pragmatic issuesand emerging discourses. Paper presented atThe Life Of Mobile Data: Technology, Mobilityand Data Subjectivity Conference, University OfSurrey, 15–16 April.

Stone A (2004). Mobile scaffolding: anexperiment in using SMS text messaging tosupport first year university students. In Kinshuk,C-K Looi, E Sutinen, D Sampson, I Adeo, L Udenand E Kähkönen (eds) Proceedings of the 4thInternational Conference on Advanced LearningTechnologies. Los Alamitos, CA: IEEE ComputerSociety Press, 405–409.

Stone A, Alsop G and Tompsett C (2002).Grounded Theory as a means of user-centreddesign and evaluation of mobile learningsystems. In Philosophy, Psychology, Education.Vienna: Passagen Verlag.

Sutherland J (2002). University challenged.The Guardian, Monday 22 July.

Wolcott LL and Betts KS (1999). What’s in it for me?Incentives for faculty participation in distanceeducation. Journal of Distance Education,14, 34–49.

Stone 185

AbstractBased on preceding work on a mobile learningscenario, a Java 2 Platform, Micro Edition(J2ME)-based game-like application, calledHistoBrick, is being designed. It serves as aplatform for exercises, assignments andself-study in combination with the highereducation (HE) courseware on descriptivestatistics. This paper focuses on the mobileapplication and ignores its embedding into asuitable learning management system (LMS).HistoBrick aims to provide a ubiquitous tool forlearning about distributions and their mostimportant characteristic numbers. The didacticalbackground is inspired by recent ideas about howlearning works in the so-called constructivistsetting, in general; and by the findings ofgame-based learning, in particular.

Keywordsmobile learning, edutainment, J2ME,game-based learning

1 How can edutainment promotemobile learning?The evaluation of a mobile course on descriptivestatistics (Ströhlein 2003; Ströhlein and Fritsch2003) revealed that students appreciate theubiquitous access to remote communicationprovided by mobile phones and their relativelylong battery duration period. Recent mobilephone models can display a variety of documentformats and have considerable computingcapabilities as well: thus, they are an interestingtool for course authors, learners and tutors.

Students considering the use of a mobile phonefor learning purposes are typically thinkingof learning while riding on a bus or train orwhile spending their spare time in publicplaces; thus they will easily be distracted.

HistoBrick: utilising J2ME to enable students to use mobile phones for game-based learning on descriptive statisticsGeorg Strö[email protected]ät in HagenLG DVTD–58084 HagenGermany

But a game-like mobile learning application cancreate the necessary ‘grip’ to attract a student’sattention completely: this is a main preconditionfor effective learning, and hence one reason todevelop an edutainment application. The finalevaluation phase of the HistoBrick application(eg Meier and Seufert 2003) in 2005 will includea check on whether the other effects ofgame-based learning predicted by its advocates(eg Prensky 2000, 2001, 2003, 2004) areactually felt.

Another reason why a game-like scenario hasbeen chosen to encourage students to reachtheir learning objectives – even in a subject thathas a reputation for dealing with ‘dry numbers’and being ‘boring’ – is a change in attitudetowards learning during the last few decades.If these young people are to be enticed to spendmore time on learning instead of listening tomusic or talking with friends, the coursewarehas to capture them with some fun, at least.The same psychological principle inducespharmacists to add a sweet taste tocough mixtures.

Of course, this game-like application has to beable, for example, to handle the small screen sizeof a mobile device and the possibly bumpyconditions in a train or bus. This is why devicesthat can be operated using the fingers of onlyone hand seem more suitable for playingHistoBrick than those requiring a stylus. Whenstudents are travelling on a train or theunderground/metro or spending their spare timein modern buildings made of ferroconcrete, theyhave unreliable access to a mobile network oreven none. Consequently, HistoBrick is designedmainly for offline usage; but when a sufficientlystrong mobile network power signal is available,students may exchange information with theirpeers or tutors; for example, to submit results orreceive new exercises.

Ströhlein 187

2 Brief description of HistoBrickThe main purpose of HistoBrick is to enablestudents to develop and deepen their knowledgeof the relation between (the shape of) adistribution and its most important characteristicnumbers; that is, mean and standard deviation aswell as all quartiles. By creating suitable sets ofexercises to be solved within HistoBrick, tutorscan guide students to experience the flaws andfallacies that occur if the characteristic numbersare dealt with without having a look into the data.This means that the application is based on aso-called constructivist didactic (eg Thissen 1997).

2.1 Technical backgroundIn order to keep the effort required to create andmaintain HistoBrick to a reasonable level, across-platform and cross-device language have tobe used for coding; in this project, Sun’s J2ME(Java 2 Platform, Micro Edition) has been chosen.J2ME provides ways to establish HyperTextTransfer Protocol (HTTP) connections in order tocontact other students or the tutor. Theseconnections also enable the use of media –pictures, music or video – for communicationor when the tutor wants to reward a studentfor a successfully solved exercise. It isrecommended that the mobile device has acolour screen of at least 4k depth and 128x128(WxH) pixels resolution.

The ‘number crunching’ capabilities of themanufacturers’ implementation of Sun’s J2MEplatform differ considerably from phone tophone, according to the versions of both theConnected Limited Device Configuration (CLDC)and the Mobile Information Device Profile(MIDP). The latest phones come with CLDC 1.0and MIDP 2.0 and a few have begun toimplement CLDC 1.1 and MIDP 2.0. HistoBrickis going to depend on MIDP and CLDC Javaclasses only, so that it is highly portable. Even amoderately complicated mathematical formulafor calculating standard deviations requires afloating point number format, which fortunatelyhas been introduced in CLDC 1.1. Because allvariable values are integers in HistoBrick and thuscan be processed by CLDC 1.0-enabled devices,a version that deals only with all quartiles is beingdeveloped and tried out first. The improvementsthat MIDP 2.0 adds in terms of security,networking and gaming features stronglyrecommend its usage for the realisation ofHistoBrick. One of the first phones in the shopsin Germany that is shipped with the combinationCLDC 1.0/MIDP 2.0 and which most studentscan afford is the Sagem myX–7, from which thescreenshots in this paper are taken.

2.2 Features and design of the gameThe game aspect of HistoBrick is created by thefact that students have to direct an automatically‘falling brick’ into one of eight boxes (seeFigure 1 opposite). The aim of the game is tocreate a distribution of ‘bricks’, the characteristicnumbers of which meet predefined target valuesas closely as possible. These target values can beself-defined, chosen randomly, or provided by thetutor or other students via a suitable channel ofcommunication, the design of which is subject toongoing research. At present, the preferredsolution is to use HTTP connections betweenphones and the application web server togetherwith server-side scripting, as this allows studentsto use relatively cheap General Packet RadioService (GPRS) volume tariffs. An alternativewould be to use the relatively expensiveShort Messaging System (SMS) or MultimediaMessaging System (MMS) for directcommunication between phones.

The text language of HistoBrick can be chosenfrom a screen which automatically appears anddisplays all available languages; though at themoment, only German and English areimplemented. Selecting a language automaticallybrings up the main menu, which provides accessto all the important features of HistoBrick – starta game with randomly chosen conditions; load,set, save or send game conditions; display theinstructions or additional information; and finally,exit the application. The receipt of values fromother students or the tutor is handled in the ‘loadgame’ sub-menu.

The top area of the game screen (see Figure 1opposite) is used for initially placing the ‘bricks’and moving them into the boxes drawn below bypressing the corresponding buttons. The boxesarea shows a diagram indicating how many‘bricks’ have already been directed into all theeight boxes; and the feedback area at the bottomof the screen shows two indicators: the upperone corresponds to the characteristic numbers ofthe distribution that has actually been created;and the lower one corresponds to their pre-settarget values. In the top left corner, the overallnumber of ‘bricks’ and that of the one that is justfalling are indicated numerically in order to notifya student how carefully a ‘brick’ has to be placed.

During a game, approximately three times persecond, a ‘brick’, which is depicted as a smallsquare and initially appears at the very top of thescreen, is automatically moved a small waytowards the boxes. In each step, by pressing theappropriate button or pushing the mini-joystick ofthe mobile device, a gamer is able to choose todirect the ‘brick’ to the left or right, to move itdirectly into the box below its actual position,


or to interrupt the game for about 2 seconds toadjourn for further thought. If a gamer is requiredto stop the game due to environmental influence(eg background noise, unexpected movement ofthe device, incoming call), pressing theappropriate menu key of the device stops thegame and lets the main menu appear on thescreen to save the actual boxes’ scores andpre-set values into the persistent memory of themobile device, if wanted. The game conditionsare automatically saved when a call comes in.

The colours of the game screen are chosencarefully. In order to indicate the closerelationship between (a) the ‘bricks’, (b) the barsshowing how many ‘bricks’ are found in theboxes and (c) the indicator of the characteristicnumbers of the bricks’ distribution just below theboxes, only one colour (blue), is used for all ofthem. The indicator of the pre-set characteristicnumbers uses a second colour (red). Bothindicators use two levels of colour saturation: therange corresponding to the measure of spread isindicated using half saturation, and the positionof the measure of location is indicated using fullsaturation. The background, the box walls and allscales as well as additional information arepainted in different saturation levels of grey andblack. The saturation level of the box walls ischosen so that they do not visually interrupt theshape of the bricks’ distribution bars, but areclearly visible above them (see Figure 1). This isimportant because on the one hand, the shapeand the indicator form the very part of HistoBrickthat is most important for the learning process;and on the other hand, the upper ends of the boxwalls have to be clearly perceptible in order topermit a ‘brick’ to be easily moved into thechosen box.

Figure 1 Screenshots of Histobrick as displayed on theSagem myX–7

3 Future prospectsAfter coding and thoroughly testing all offlinefeatures, the online features like sending andreceiving game settings will be created. The nextstep is to identify an open source LMS that caneasily be adapted for use with mobile handhelddevices. After the necessary changes have beenapplied to the LMS, the mobile courseware willbe evaluated with first-year students at severaldifferent universities. Then didactic issues can beaddressed too; for example, whether thehorizontal starting position of the ‘bricks’ shouldbe chosen randomly instead of keeping it fixed; ifthe Central Processing Unit (CPU) of the mobiledevice is powerful enough; whether a demomode could be added to create modeldistributions. Another idea is to use ‘bricks’with a different ‘weight’ in order to let studentsget a broader sense of how distributions evolvein reality.

ReferencesMeier C and Seufert S (2003) (in German).Game-based learning: Erfahrung mit undPerspektiven für digitale Lernbeispiele in derbetrieblichen Bildung. [Game-based learning:experience with and prospects of examplesof digital learning in workplace education.]In A Hohenstein and K Wilbers (eds) HandbuchE-Learning (Ergänzungslieferung). [Handbook ofe-learning (additional material).] Köln: FachverlagDeutscher Wirtschaftsdienst.

Prensky M (2000). Digital game-based learning.McGraw–Hill.

Prensky M (2001). Interview with Prenskyon game-based learning. Athttp://elearningpost.com/elthemes/prensky.asp,accessed June 2004.

Prensky M (2003). But the screen is too small…At www.marcprensky.com/writing/Prensky%20-%20But%20the%20screen%20is%20too%20small.pdf, accessed June 2004.

Prensky M (2004). Use their tools! Speaktheir language!

At www.marcprensky.com/writing/Prensky-Use_Their_Tools_Speak_Their_Language.pdf,accessed June 2004.

Ströhlein G (2003) (in German). Bericht überEntwicklung und Erprobung eines Kurses fürmobiles Lernen. [Report on development andtry-out of a course designed for mobile learning.]In K David and L Wegner (eds) Mobiles Lernenund Forschen. [Mobile learning and research.]Kassel: Kassel University Press.

Ströhlein 189

Ströhlein G and Fritsch H (2003). Test andevaluation of a course designed for mobilelearning. ZIFF Papiere No 120. Hagen:FernUniversität.

Thissen F (1997). Inventing a new way oflearning? Constructive fundamentals ofa multimedia teaching methodology.At www.frank-thissen.de/Inventing%20a%20New%20Way%20of%20Learning.pdf,accessed June 2004.


AbstractThe major goal of the multidisciplinary DigitalLearning 2 project is to develop an online tool toevaluate the usability of digital learning materials.This paper concentrates on describing thetechnical and pedagogical usability criteria andhow they are specified in order to take accountof the special features of mobile learningmaterials and environments. This was done byincorporating a ‘components of mobile learning’(CML) evaluation model, which had beenpreviously developed and empirically tested, intothe criteria, after reviewing the other earliercategorisations for evaluation of mobile learning.

Keywordsevaluation, mobile usability, pedagogical usability,technical usability

1 IntroductionThe multidisciplinary Digital Learning 2 (DL 2)research project, which started in 2001, hasmembers from the research fields of education,computer science and hypermedia. The majorgoal of the project is to develop a tool to evaluatethe usability of digital learning materials andenvironments. In order to reach that goal,numerous concepts and models have beendefined; for example, the criteria for technical andpedagogical usability and the Components ofMobile Learning (CML) model.

Evaluation of the technical and pedagogicalmobile usability

Syvänen Nokelainen 191

Antti Syvänenantti.syvä[email protected] LaboratoryUniversity of TampereFIN–33014 Tampereen yliopistoFinland

Petri [email protected] Centre for Vocational EducationUniversity of TampereFIN–13101 HämeenlinnaFinland

The following components of the technicalusability criteria were specified (Nokelainen 2004):

1 accessibility

2 ‘learnability’ and memorability

3 user control

4 help

5 graphical layout

6 reliability

7 consistency

8 efficiency

9 memory load

10 errors.

In addition, the following pedagogical usabilitycomponents were specified (Nokelainen 2004):

1 learner control

2 learner activity

3 cooperative learning

4 goal orientation

5 applicability

6 effectiveness

7 motivation

8 valuation of previous knowledge

9 flexibility

10 feedback.

The Components of Mobile Learning (CML)model was developed for evaluation of mobilelearning. The model (Ahonen et al. 2002)consists of:

1 continuity and adaptability

2 learning as a personal process

3 contextuality in learning

4 accessibility

5 support for time and learning management

6 flexible interaction.

From these criteria, we derive the questions thatare presented to the user when he or she isevaluating mobile learning material orenvironments. The purpose of the evaluation toolis not to classify evaluation targets as ‘good’ or‘bad’ ones, but instead to show in anunderstandable manner some of the designprinciples and goals behind the evaluation target.There are at least two explicit ways to use thetool: first, to evaluate any given system orlearning material by means of a self-reportmultiple-choice questionnaire; and second, toseek for other users’ evaluations of learningsystems or materials.

The criteria for evaluation of the technical andpedagogical mobile usability have beendeveloped in the following stages:

p investigation of the previous research reportsdealing with the characteristics of mobility andmobile learning (2001–2002)

p development of the first version of the CMLmodel (2002–2003; Ahonen et al. 2002)

p development and empirical testing of the criteria(2003; Syvänen et al. 2003, 2004)

p review of earlier categorisations for theevaluation of mobile learning and incorporationinto the technical and pedagogical usabilitycriteria (2004).

2 Evaluation criteriaThe incorporation of the criteria for evaluatingmobile learning into the original technical andpedagogical usability criteria was done after thereview of earlier categorisations for evaluation ofmobile learning (see Table 1 opposite).

The degree of sensitivity in the different criteriacomponents for an evaluation of mobile learningis presented in Table 2 (overleaf).


Syvänen Nokelaine 193

Table 1 Summary of the earlier categorisations for evaluation of mobile learning

Technical

Pedagogical

Öquist, Goldstein and Chincolle (2004)Factors of mobile usability

Portability

Attentiveness

Manageability

Learnability

Kynäslahti (2003)Elements of mobility

Convenience; rationality

Expediency

Immediacy

Quality of life

Luff and Heath (1998)Types of mobile collaboration

Micro-mobility

Local mobility

Remote mobility

Perry, O’Hara, Sellen,Brown and Harper (2001)Properties oftechnology supportingmobile workers

Lightweight andhighly flexible

Allows more effectiveplanning of activitiesand flexible allocationof resources

Supports effective useof ‘dead time’ to planfor upcoming mobileactivities and catch upwith non-mobile work

Allows the use oflocation and accessto locally availableresources

Allows monitoringof remote activitiesmore easily

Sharples (2000);Sharples, Corlett andWest-mancott (2002)Tools for lifelong learning

Highly portable

Individual

Unobtrusive

Available

Adaptable

Persistent

Useful

Easy to use

Savill-Smith and Kent(2003)Why use palmtopcomputers for learning?

Relatively inexpensive

Possibility ofubiquitous computing

Access to informationand information literacy

Possibility ofcollaborative learning

Possibility ofindependent learning

Vavoula and Sharples(2002)Requirements for lifelonglearning organisers

Easily transferablebetween places

Available and functionalat any time

Smooth transitionbetween learning topicsand construction of meaningful, integrated knowledge

Gay, Rieger andBennington (2002)Levels of objectives:mobile computersin education

Productivity

Flexible physical access

Capturing andintegrating data

Communicationand collaboration

3 ConclusionsIn this paper we have specified the criteria forevaluation of mobile learning material in detailand shown how mobile learning materials andenvironments are evaluated with an onlineevaluation tool. The real-life empirical testing ofthe evaluation tool was difficult to conduct as atthe time, there were only a few learningmaterials or environments defined solely formobile use. However, the progress of developingsuch materials all over the world is rapid, andthus we see the mobile evaluation criteria as ameaningful part of the evaluation tool. In thefuture, we intend to develop the evaluation toolthrough empirical testing of real-life mobilelearning materials and environments.

AcknowledgementsThe Digital Learning 2 research project isfinanced by TEKES (the National TechnologyAgency of Finland; www.tekes.fi/eng). Theproject is being carried out in collaboration withtwo Finnish research groups in Häme Polytechnicand at the Hypermedia Laboratory at theUniversity of Tampere. For further details,visit: http://dll.hamk.fi/dl2 andhttp://evaluator.hamk.fi:8181/evaluator_english/

ReferencesAhonen M, Koponen H, Syvänen A and Turunen H(2002) (in Finnish). New mobile innovationssupporting learning. Digital Learning 1 project.Saarijärvi: Häme Polytechnic and UniversityofTampere.

Gay G, Rieger R and Bennington T (2002).Using mobile computing to enhance field study.In T Koschmann, R Hall and N Miyake (eds)CSCL 2: carrying forward the conversation(computers. cognition and work). Mahwah, NJ:Lawrence Erlbaum Associates Inc, 507–528.

Kynäslahti H (2003). In search of elementsof mobility in the context of education. InH Kynäslahti and P Seppälä (eds) Mobile learning.Helsinki: IT–Press, 41–48.

Luff P and Heath C (1998). Mobility incollaboration. In Proceedings of the ACMConference on Computer SupportedCollaborative Work, Seattle, 14–18 November.New York: ACM Press, 305–314.

Nokelainen P (2004). Conceptual definition ofthe technical and pedagogical usability criteriafor digital learning material. In Proceedings ofED–MEDIA 2004: World Conference onEducational Multimedia, Hypermedia andTelecommunications, Lugano, Switzerland,2004(1), 4249–4254.

Öquist G, Goldstein M and Chincholle D (2004).Assessing usability across multiple userinterfaces. In A Seffah and H Javahery (eds)Multiple user interfaces. Chichester: John Wileyand Sons, 327–372.

Perry M, O’Hara K, Sellen A, Brown B andHarper R (2001). Dealing with mobility:


Table 2 Evaluation of mobile learning materials and environmentsNumbers in this table relate to the numbered lists of criteria in section 1

Technical Pedagogical

Component

1 Accessibility

2 ‘Learnability’ and memorability

3 User control

6 Reliability

7 Consistency

8 Efficiency of use

9 Memory load

Questions for the userrelating to mobile material

Manageability

Cross-platform use

Attentiveness

Offline use

Cross-platform use

Manageability

Attentiveness

Component

2 Learner activity

3 Cooperativelearning

5 Applicability

6 Effectiveness

8 Valuation of previousknowledge

Questions for the userrelating to mobile material

How personal arethe learning products?

Flexibility in collaboration

Support for contextual useAttentiveness

Added value of mobility

Continuity of learning process

understanding access anytime, anywhere.ACMTransactions on Computer–HumanInteraction, 8(4), 323–347.

Savill-Smith C and Kent P (2003). The use ofpalmtop computers for learning. A reviewof the literature. London: Learning and SkillsDevelopment Agency 2003.At www.m-learning.org/docs/the_use_of_palmtop_computers_for_learning_sept03.pdf, accessed 20 January 2004.

Sharples M (2000). The design of personalmobile technologies for lifelong learning.Computers and Education, 34, 177–193.At www.eee.bham.ac.uk/sharplem/papers/handler%20comped.pdf, accessed20 January 2004.

Sharples M, Corlett D and Westmancott O(2002). The design and implementation ofa mobile learning resource. Personal andUbiquitous Computing, 7(6), 220–234.At www.eee.bham.ac.uk/sharplem/Papers/mobile%20learning%20puc.pdf, accessed20 January 2004.

Syvänen A, Nokelainen P, Ahonen M andTurunen H (2003). Approaches to assessingmobile learning components. In L Mason,S Andreuzza, B Arfé and L Del Favero (eds)European Association for Research on Learningand Instruction. Abstracts of the10th EuropeanConference for Research on Learning andInstruction, Padova, 581.

Syvänen A, Nokelainen P, Pehkonen M andTurunen H (2004). Mobile learning future views.In Proceedings of ED–MEDIA 2004: WorldConference on Educational Multimedia,Hypermedia and Telecommunications, Lugano,Switzerland, 2004(1), 2112–2118.

Vavoula G and Sharples M (2002). Requirementsfor the design of lifelong learning organisers. InS Anastopoulou, M Sharples and G Vavoula (eds)MLearn 2002: Proceedings of the EuropeanWorkshop on Mobile and Contextual Learning,Birmingham, 20–21 June, 23–26.

Syvänen Nokelaine 195

AbstractThis paper describes the way that handhelddevices are used as multimedia guides in theMuseo Diocesano di Catania (Italy). Particularimportance was given to designing these devicesand their contents so that they would allow for anextended visit, going beyond the Museo’scontents and covering more general topics thatare related to the exhibits.

Keywordschurch museums, PDA, beni culturali ecclesiastici

1 Do we really need amultimedia guide?Visiting a museum is an enriching experience,which involves different levels of perception,from the aesthetic and emotional to the rationaland cognitive. Indeed, one can enjoy amasterpiece at an aesthetic level only, beingemotionally involved in what the artist shows andsuggests. It is expected, however, that amuseum visit could draw more widely onvisitors’ knowledge and promote a deeperinterest in the items they look at, leading them toread and study further on the topics shown in theexhibit or related to them.

Often ‘lesser’ museums prove to havestimulating and exciting contents. The MuseoDiocesano di Catania is a place where visitorscan find out about the many different ways ofexpressing religious feeling across the centuries,and learn about the history of Catania and thedevotion attributed to Sant’Agata, the patron saintof Catania.

The extensive printed explanations thatcomplement exhibits develop cross-cultural themesand give a large amount of information on theworks on display. However, for practical reasonsthese explanations are outside the glass cases ofthe exhibits, far away from the pieces they refer to.

A mobile interactive guide to the Museo Diocesano di Catania: how to ‘open’ the Museo using PDAsDr Carlo [email protected] SrlViale Caterina da Forlì 3220146 MilanoItaly

References to single works of art are sometimesnot so easy to understand, and are usually madeby describing or referring to the object using anumber. Users have to make a considerableeffort if they want to interpret and understandthe items on display.

Audio guides have become increasinglywidespread recently. These guides let you listento a recorded voice that explains the objects. Torefer to a specific section of an exhibit, you haveto key in a number or you follow a predefinedpath along which there are ‘listening points’,usually identified by an icon. These vocal guidesare not too flexible and make it difficult to definea customised path during a visit: the risk of losingthe correlation between the recorded voice andwhat you are looking at is very high.

2 The deviceTo address these problems we have designed amultimedia guide that instead of simplysupplying a recorded description allows fordeeper interaction and the display of text andpictures to supply further information on theworks in the exhibit. The devices most suited forthis purpose are the new generation PDAs,equipped with an operating system Pocket PC2002, with a 64k colour depth display and goodresolution (320 x 240 pixels).

These tools allow visitors to widen their visitbeyond the museum walls. They supplyinformation about exhibits in a way that is easierto understand and customised to individual userswhen they want it. These two features are thekey difference between multimedia guides andconventional printed explanations about exhibits.Now it is possible to make it easier to understandany work of art on display, by combining text,pictures and audio.

Tognoni 197

All the information about the Museo is loadedinto the device memory, which is large enough tocontain a wide set of multimedia contents,suitable for exhibits larger than those held by theMuseo di Catania. The continuing reduction in thecost of memory and the increasing size ofmemory available makes this possible.

The PDAs are equipped with a web browser thatdisplays HTML pages. A player is installed inorder to allow for the reproduction ofMacromedia Flash movies. This solution enablesexisting technologies to be used. Texts aredisplayed as HTML files that embed compressedimages (GIF or JPEG). Audio is controlled andplayed using the Macromedia Flash player, whichallows for a very high compression in MP3format, without compromising quality. The audiois played through a small loudspeaker embeddedin the device for small groups, or could bereproduced using headsets from the standardconnector provided.

The guides are rented at the museum entrancewhere tickets are sold. They are optional and asmall additional fee is charged.

3 The man–machine interfaceWe have adopted a very simple and conventionalform of interaction, placing value on the usabilityof the multimedia guide, not on fancy butuseless features, such as animation or jingles.The hierarchical menu on the guides mirrors theMuseo’s structure and lets visitors access theexplanations of the various rooms during thevisit, either in the natural sequence in which theyare encountered or according to an order definedby users themselves, according to their interestsor preferences.

Visitors use the tactile screen and stylus tooperate the guides, which allows for a very highpointing precision. During the usability tests weassessed that this style is a much more naturalway of interacting with devices than the mouse.Even users who are not accustomed to newtechnologies have found it intuitive and easyto use.


4 ContentsThe multimedia guide in this very first versiondescribes only a small number of the objectsdisplayed in the Museo, but gives usersinformation about all the Museo’s rooms. There isa floor plan of the Museo’s rooms, with zoom-infunctionality and easy identification of the worksof art displayed; this lets visitors identify at aglance what is most relevant for their interests.

The page structure is kept as simple as possible.At the top of the page there are only two icons.One can be used to navigate back to previouslyvisited pages; the other one turns the audiovoice-over on or off. The text contains a numberof hyperlinks to navigate among the topicssupplied in a non-sequential way. For instance,from the S. Cataldo reliquary you can accesspages on reliquaries in general and those withhuman body features, and can get moreinformation on the works of Paolo Guarna, theauthor of S. Cataldo reliquary. In a similar way,the painting of the miracle of S. Isidoro Agricola islinked to other works by different painters on thesame topic, which can be found at the PradoMuseum in Madrid, Spain.

The multimedia guide particularly addresses theworship of S. Agata. The museum containsseveral items related in some way to the worshipof the saint, but they cannot easily be understoodby visitors who haven’t been at the Februarycelebrations to honour the saint. The bare fercolo,on display in Room IX, for instance, is enlived bythe description of its use during the celebrationsof S. Agata as supplied by the multimedia guide.

Some themes that are outside the scope of themuseum but strongly related to some of theitems on display are developed in further detail.For instance, the guide contains a detaileddescription of the symbolic meaning of chalicesused during the Mass (see overleaf), supplied byProfessor Pfeiffer of the Pontificia UniversitàGregoriana in Rome.

Tognoni 199

AcknowledgmentsWe wish to thank Pirelli Real Estate and Co. forsupporting this project financially; Don SantinoSalamone, director of Museo Diocesano ofCatania, for his material and spiritual support;Professor Pfeiffer of the Pontificia UniversitàGregoriana in Rome for his enlightening andinteresting explanations on the spiritual meaningof sacred works of arts; Archictect GiovannaCannata for the rich and important informationshe supplied; and all the collaborators of MuseoDiocesano di Catania for their support and theenergy they provide during their daily work atthe Museo.

ReferencesFantoni, S F. Museums with a personal touch.At www.ecdc.nl/publications/papers/Museums%20with%20a%20personal%20touch.pdf

Marucci, L and Paternó, F (2001). Designand evaluation of an adaptive virtual guide forweb applications. Pisa: CNUCE–C.N.R.

Sakamura, K (2003). Personalized digital museumassistant. At www.um.utokyo.ac.jp/dm2k-umdb/publish_db/books/dm2000/english/01/01-17.html,accessed 5 May 2003.


AbstractWith the approach of endless bandwidth andincreasingly powerful CPUs, we are overcominglimitations to producing ‘the perfect content’.This is a period wherein computation,communications and interfaces are becoming‘transparent’ elements of our mobile lifestyle. Itis also a period when applications will take on anincreasingly ‘humane’ identity – reflective both ofthe need to address directly mounting societalproblems such as unemployment, educationalfailures and mental illness; and of Englebart’s1965 vision for justifying the PC as ‘anaugmentative’ filter in service to evolvinghumanity. In evaluating the capacity formobile/wireless devices to augment moreimportant parts of our lives, I have chosen tofocus on developing career-centric evaluationtools for phones/PCs, and share 10 designepiphanies I have evolved for developing moresocially valuable applications acrossmobile/wireless devices.

Keywordsaugmentation, lifelong learning, social andemotional Learning

1 The approach of pervasiveinteractive communications

1.1 Unlimited bandwidth/MhzWith the approach of endless, affordablebandwidth and low-power, high-yield CPUs weare overcoming the limitations to producingpervasive interactive communications, devicesthat are perceived as intimate and ‘trusted’, and‘the perfect content’ for fulfilling any task.

The result is a fundamental shift in our abilityto produce software and services that notonly support lifestyle and quality of life,

The shift to seamless augmentation and ‘humane’ applications via mobile/wireless devices: a view to a future for lifelong learningDavid C [email protected] UniversityEpiphany Partners50 Woodside Plaza, # 601Redwood CityCA 94061USA

but also are optimised for addressing realproblems in our lives and society such asunemployment, educational failures and mentalillness. Why not consider a greater focus ondeveloping applications that are more ‘humane’,given both the immeasurable harm thatunemployment plays on attitude and economicdevelopment, and given the fact that 1–5% ofmost populations suffer from mental illness –much of which is addressable via timelycommunications with an expert?1

Of course, choosing to develop software thatuses popular mobile devices to address issues aspersonal as our survival and existence is fraughtwith critical challenge. Applications developed toassist with our mental health would be moreproblematic than, say, delivering languagelearning, music, sports and the news. Moralmandates, societal values and legal liabilitiescloud one’s ability to develop applications thatmight assist individuals across diverse cultures.

Yet the notion of evolving truly ‘augmentative’software that increasingly and effectivelyaddresses our higher order needs is not onlyexciting, but inevitable. It is a vision that nowseem not only unstoppable, but in somethoughtful ways worthy of acceleration –particularly within the more broad content scopeof ‘lifelong learning’.

2 At least one future for lifelonglearning: career counsellingWhen one thinks of education – whetherclassically or virtually – one may typically think ofthe classics (mathematics, science). But what ofthe less than classical types of ‘attitudinal’learning referenced above such as curriculumspecific to counselling and coaching?Considering current challenges such as:

Traub 201

p the high number of unemployed

p the social, emotional and economic impactof this status on individuals and their familiesand society

p the availability of electronic and human resourcesto assist in the process of re-careering

p the unique and trusted capacity of mobiledevices to provide access to their resources.

I have chosen to commit my energy in service todeveloping career counselling and augmentationsoftware. I have begun this campaign with thedevelopment of a ‘mythological evaluation tool’that enables the unemployed make a betterselection of an ideal career path based onan overarching ‘narrative’ analysis of theiraggregate careers.

2.2 Ten design epiphaniesIn developing my emerging process I haveevolved 10 design epiphanies reflecting my20 years of experience as a digital producer ofeducational, entertainment and enterpriseapplications which I share in this document.Particularly germane to this notion of more‘humane’ applications, these 10 considerationsshould enable designers to create mobile andwireless learning applications of particularmeaning and value to their targeted user base.They are as follows:

1 Commence all designs with a ‘top 10’ list ofinnovations you hope to bring to market at thetime of launch; this strategy will assist in thepositioning, marketing and promotion of yoursolutions to consumers, distribution partners andthe press.

2 Remember that despite their appeal and obviousvalue, ‘m’ (‘e’ and ‘t’) learning benefit from thevalue of relationship-based ‘scaffolding’ germaneto conventional learning environments (seeprolearners.com); reintroduce peer-orientedgroup interaction into the mix.

3 Consider that ‘attitudinal learning’, the fifth ofGagne’s learning taxonomy (motor skills, verbalskills, problem solving, higher order thinking andattitudinal learning), remains the most neglectedin the classroom, workplace and home – and isperhaps the most required in service tosolving many of today’s greatest economic andsocial problems.

4 Increase your focus on aggressivelycross-disciplinary research (eg technology plussocial sciences), out of which entirely newdomains of socially valuable applications andservices innovation will emerge – particularly ascommunications infrastructure evolves towardstoward maturity and ubiquity.

5 Reflect that the primitive, one-off user profilingnow conducted sporadically by retailers will giveway to ever-aggregating, lifelong client ‘profiling’,providing ever richer, seamless and pervasivemeans by which to identify, offer anddeliver educational, cultural, economic andpsychological value.

6 Reflect that we are ‘just in time’ centric – drivenprimarily by the immediate relevance of anychallenge or opportunity to our core needs, thisinsight suggesting that ‘profile-driven’ solutionswill emerge as mainstream.

7 Believe that mobile phones and their ever-moreintelligent offspring are likely to emerge as theprimary ‘client’ for all humans within five toseven years – our primary interface to the world’saggregate data-based wisdom.

8 Remember that the inevitable algorithmic growthof product and services capacity (Moore’s Law)continues at a consistent pace, suggestingever-receding technological limitations – and evermore unlimited capacity to implement organicprocesses that mirror human intelligence andemotions in service to ever more compellinginterfaces and engines.

9 Remember the foundational vision of DouglasEnglebart, father of the mouse, and co-father ofthe modern PC. In 1965 he envisioned a daywhen personal computing technology wouldevolve as the basis for assuming ‘augmenting’our quality of life by taking responsibility for‘lower level needs’ such as computation, storageand so on, so that our quality of experience ashumans could evolve.

10 Please remember that the things theconsumers – all humans – really care about willalways remain the same: health, knowledge,self-esteem, friends, love and home.

I share the above design epiphanies with theintent to encourage more of us to develop evermore humane and compassionate services andapplications that are truly relevant to all of ourneeds as happy, healthy and effective humans.

Endnote1 International Herald Tribune, 3 June 2004, 5.


AbstractMobile learning using handheld computers,smartphones and dedicated gadgets, oftenwith the added functionality of text andpicture messaging, built-in cameras andlocation-awareness, presents a variety oftechnical and pedagogic opportunities andchallenges to educators and developers. Theseare becoming understood and mobile learning isslowly making the transition from small-scaleshort-term trials to sustainable institutionalembedding. This process implies a continued needto explore appropriate methods for evaluation.

This paper looks at the ethical foundations ofsuch evaluation and at the ways these newtechnologies impact on how we can monitor andevaluate in ways that are legally, socially andculturally acceptable.

The paper sets out some current trends in ethicsin the transition from codes to principles andframeworks, and explores how this affects thedelivery and evaluation of mobile learning.It also looks at the relationship betweenphilosophical perspective and practical ethics asthey affect critical issues in evaluation such asinformed consent.

Keywordsethics, evaluation, monitoring

1 IntroductionThe issue of ethics influences mobile learning attwo different stages. The first is that evaluatingmobile learning trials requires ethicalconsideration. The second is that any ongoingonline activity, including mobile learning, has anethical dimension and this needs to be identifiedand explored as mobile learning evolves. In thefirst instance the responsibility is of the evaluatoror researcher for the research participant. Inthe second, it is of the teacher for the learner.

Mobile learning – the ethical and legal challenges

In both, the idea of ethics encompasses aspectrum from statutory issues to culturalissues, from what is defined as legallyacceptable to what is defined as sociallyacceptable. Ethics is a significant issue becauseevaluation and provision that do not conform toexplicit ethical guidelines may be:

p seen as improper or immoral

p breaking laws or regulations

p unacceptable to the research community.

There are however a number of potentialproblems in attempting to develop ethicalguidance for mobile learning.

p The ethics of computer-based and remote onlineresearch, including evaluation, with participantsare still immature and untested.

p Online learning, including mobile learning, couldtake place in several different countries andconsequently across different legal jurisdictions.

p Online learning, including mobile learning, mightbe working with participants whose ages arenear the legal age of majority (and this agemay vary from country to country).

2 Ethical principlesMany professions and bodies attempt to giveethical guidance to their members and quite afew of these would be relevant to thepractitioners of mobile learning. There are bodiesfor learning technologists, educationalresearchers and software developers, and foracademics within universities. Each offers someethical guidance to practitioners. In some fieldsthere has been a change of emphasis from codesto principles (eg BACP 2002). This allows ethicalpractice to be proactive and responsive tocontext or situation. It may then match progressin pedagogy and technology, rather than beingmerely reactive and always pursuing events.

Traxler Bridges 203

John [email protected] for Learning and TeachingUniversity of WolverhamptonWulfruna StreetWolverhampton WV1 1SBUnited Kingdom

Nell [email protected] School of EducationUniversity of Bristol35 Berkeley SquareBristol BS8 1JAUnited Kingdom

Significant ethical issues and ethical differencesmay not be raised, discussed or resolvedbecause they are ‘taken-for-granted’ orconversely considered ‘not worth mentioning’.Ethical frameworks are often assumed andtherefore obscure (Warner 2004).

At a rather more theoretical level, ethics can bebased on divergent philosophical foundations thattake, for example, different positions on ‘reality’,knowing and the self. A simple division can bemade between modernism, with its confidencein the ability of scientific method to give accessto a pre-existing reality, and postmodernism,which is far more sceptical about such ‘truthclaims’, believing instead that reality iscontinuously and collaboratively constructed.

Within a modernist framework the researcherstrives for objectivity. Engaging in postmodernresearch involves researchers reflexively askingwhy they are doing this research in this way,what is it silent about, what gives it authority,and who is privileged by it?

These apparently recondite issues can make aprofound difference to practical ethics, as weshall see when we look at the topic of ‘informedconsent’ in a later section.

3 Ethics: the mobile learningdimensionIn order to consider the impact of mobile learningon the ethics of evaluating educational provision,it is easiest to start with models of ‘pure’ mobilelearning. We can then move on to the more likelyforms of provision, those involving mobilelearning blended with other, more conventionalor traditional formats and media. In this sense,‘pure’ mobile learning means learning deliveredsolely by mobile technologies.

The major ethical issues in relation to theevaluation of ‘pure’ mobile learning includeinformed consent (and participant withdrawal),anonymity and confidentiality, participant risk,payment to participants and cultural differences(Anderson and Kanuka 2003, Hewson et al. 2003).

3.1 Informed consentInformed consent is a core ethical issue forresearch and evaluation. It refers to participants’understanding of the nature, extent, duration andsignificance of their involvement with the research.

Returning to issues of underlying philosophy,where truth is problematised informedconsent cannot be straightforward, sofor postmodernists it can never be total.

The research process is complex and dynamicand as such cannot be simply represented toparticipants. Data can take us in unexpecteddirections, which the participants have not givenprior consent to (Weatherall, Gavey and Potts2002). The binary division between researcherand researched is seen as problematic especiallywith regard to privilege and power. Consequently,there is more likely to be an ongoing process ofnegotiating issues of informed consent. From arealist position, giving information required forinformed consent may influence the responsesgiven and so ‘contaminate’ the data.

In conventional learning, informed consent mightbe achieved by a face-to-face briefing or writtenoutline with the opportunity for follow-upquestions. This briefing or outline might cover theaims and objectives of the research and thenissues such as anonymity and confidentiality,parental permission, participant risk, withdrawal,remuneration and so on. Participants then havethe chance to express their consent by signing aconsent form.

For ‘pure’ mobile learning, this is potentiallyproblematic for the following reasons.

p It may be difficult to explain fully the scope ofmobile learning in a succinct and appropriatefashion in a way that is consistent with mobilelearning itself.

p Participants may only engage in mobile learningvia SMS or VoiceXML, where there are veryfew precedents for gaining and signifyinginformed consent.

p A mobile learning system may not preservepersistent learner identities across sessionsor across devices, thereby possibly confusingthe source of consent and the data to whichit relates.

p In any remote research, it is easier todeceive researchers and evaluators. Anyinternet-based or online researchers haveno reliable face-to-face cues.

p There may be insufficient means for would-beparticipants to check their understanding of theresearch and their part in it.

p Only participants who understand the natureof their participation can legally grant consent.In mobile learning, it may be difficult toestablish whether this is the case withoutface-to-face contact.

p Expressing consent, especially in a legallysignificant fashion, usually requires aconventional written signature.


p Informed consent assumes that any informationor explanation on which the consent rests isexpressed in a fashion that is inclusive of variedliteracy, linguistic and physical abilities (andaccords with the provisions and spirit of SENDAand comparable national legislation).

p Accidental or deliberate personation isalways possible.

Existing practice

Existing practice in relation to obtaininginformed consent can come from a variety ofinternet-based evaluation and research formats,as follows.

p Obtaining consent for e-mail interviewing:the explanatory briefing is mailed or e-mailed toparticipants, or they are directed to the projectwebsite; their consent is e-mailed back orprinted, signed and faxed or posted back.

p Obtaining consent for conferencing: theexplanatory statement is posted in theconference; participants then post or e-mailqueries and then signal consent as above.

p Obtaining consent for chat: is real-time and soinformed consent is problematic. Anasynchronous area may be used in parallel toengage in the consenting process.

p Obtaining consent for participant observation:mail lists, news-groups and so on presentpossibilities for undisclosed but problematicparticipant observation – so-called ‘virtualethnography’ (Hine 2000), there is aconsent issue.

p Collecting personal electronic documents: thereis a consent issue, one also associated withharvesting system data on users’ behaviour.

However, the issue of informed consent isapparently less legally problematic than it mightappear since mobile learning trials only run therisk of participant withdrawal (rather thanlitigation) and could easily put itself in a positionwhere participation implied consent. Theinvolvement of minors in research or evaluationdoes not, in fact or in practice, require parentalconsent as long as the participants are sufficientlymature to understand the nature of theirparticipation. So-called minors give their de factoconsent to all sorts of transactions in everydaylife without needing to refer to their parents. Ineach case, the issue is whether their majority issufficient to comprehend the implications of thespecific transaction in question.

3.2 Participant riskMobile learning trials pose risks to participantsinsofar as they may inadvertently exposeparticipants to unsafe or unsavoury behaviour ormaterial via various media, from externalweb-based sources and from other participants.Participants should be aware of the nature andextent of this risk, and so especially should theparents or guardians of any under-age participants.Risks might include access or exposure toharassment, abuse, ‘spam’, hate mail or unsafeinternet sites. It is clearly the duty of researchersand evaluators to shield participants from harmand to minimise its consequences.

Some forms of mobile learning encourage orallow peer-to-peer collaboration. This raises thepossibility that participants may be exposed torisks, for example obscene picture messaging,over which the organisers cannot possibilityhave any control but for which they mightnevertheless still be legally responsible.

3.3 Participant withdrawalInternet-based procedures should makewithdrawal easy at any point. Researchers andevaluators are clearly interested in keepingparticipants live within the trials and there issome tension between these two considerations.

3.4 Payment or compensationIn much research or evaluation work it isunderstandable that participants arerecompensed for the time or effort of theirinvolvement. This establishes some kind of‘contract’ between them and the researchersand thus raises two kinds of issue. The first issueis methodological and in essence is the questionof whether payment or compensation hasskewed or contaminated any results orconclusions. The second issue is ethical andconcerned with whether the compensation canbe viewed as fair or equitable for the time oreffort involved. Issues related to differentials inculture, status and power may obscure easyaccess to answers about whether the ‘contract’is actually perceived as fair by participants,especially the learners themselves and possiblyalso host organisations in the field. Wherebenefits are tangible, the relationship is legallycontractual but if participants are merely beingrecompensed or the benefits are intangible, thisis probably not the case in law.

Traxler Bridges 205

3.5 Confidentiality and anonymityResearchers and evaluators are expected toguard the anonymity of research participants inany published accounts. There are routinepractices for this in conventional research. Inpurely virtual or online learning, this may not bean ethical problem since participants’ realidentities and attributes could be wholly fictional.Such an eventuality would however be aconsiderable methodological problem.Researchers must also guarantee theconfidentiality of any data from or aboutparticipants. In large and complex mobile learningconsortium projects with many people, partners,intermediaries, components and devices this ispotentially problematic legally.

Some researchers (Buchanan 2004) have pointedout the inherent contradiction, when quotingrespondents verbatim, with acknowledging theircopyright while preserving their anonymity.

A different issue involving privacy is the extent towhich some mobile learning technologies revealdata about individuals that they might not knowwas accessible. Some devices are location-awareand disclose the participant’s movements.

3.6 Private and public distinctionsThere is an increased need in online interactionto distinguish whether the participants regardonline exchanges and interactions as public orprivate. This is relevant to e-mail, mailbases,newsgroups, chat, forums and so on since thedistinction can become blurred or forgottenor the interaction can quickly move from thepublic to the private. This is possible inmobile learning sessions.

3.7 Roles: researcher, teacher,organiser, managerSimilarly, there is a need to maintain distinctionsabout roles. A mobile learning project workermay have more than one role in relation toparticipants, most likely developer,teacher/lecturer (with pastoral and academicresponsibilities), evaluator and researcher. Theseroles have different and possibly conflictingethical requirements and responsibilities andhence it is important for participants tounderstand the context for any given interaction.

3.8 Status and powerResearchers and research organisations, in thiscontext universities and colleges, must recognisethat any interactions with participants, includinggaining and maintaining informed consent, takeplace across myriad differentials in status andempowerment, for example between learner andteacher, or learner and researcher, and so on.These differentials are related to positions withinorganisations, within social structures, withinsociety itself and within the education system.They pose ethical and methodological problems,which may be at odds with each other. Feministresearch in particular has been at pains to teaseout the consequences of these issues forparticipant research and to develop appropriatemethodologies. Some of this thinking hasinformed research with online communities.

3.9 Cultural differencesMany evaluations take place across cultural (orsub-cultural) divides where various parties maybe dissimilar in terms of language, expectationsand values.

This will be true in mobile learning where theparticipants are culturally and ethnically diverse,usually young, and often economically exposedand possibly close to ‘street’ culture, while themobile learning researchers, usually universitylecturers, will be party to rather different types ofdiscourse and expectations.

3.10 Developing effective debriefingResearchers and evaluators are obliged to ensurethat participants are unharmed at the conclusionof their involvement with research and, to thisend, debriefing procedures are often established.These procedures might provide debriefing textwith contact details and invite comments andqueries. This may be relatively lightweight butmobile learning evaluators should neverthelessconsider some mechanism to concludeparticipants’ involvement and give ‘closure’.Many researchers make their interpretationsavailable to participants (often called ‘memberchecks’); this may or may not be part of theinformed consent process. For modernists it islikely to be seen as increasing accuracy, whereasfor postmodernists it may be to allow a morecollaborative or co-constructed process, to givegreater credence to previously excluded ‘voices’and to respond to challenges regardingresponsibility in representing the ‘other’.


3.11 NetiquetteEach new network communications technologyevolves its own specific cultural expectationsgoverning interactions and exchanges broadlytermed ‘netiquette’. Currently the most obviousof these govern e-mail, chat and conferencing:

p E-mail netiquette includes emoticons and‘smileys’, a relaxed attitude to grammar andspelling and the expectation of brevity.

p Chat netiquette includes expectations aboutinappropriate ‘ringing’.

Conferencing (and Usenet, mailbase and so on)netiquette often has published codes ofbehaviour precluding, for example, ‘spam’ andflaming. Mobile learning will have a variety ofchannels, some equivalent to existing onesidentified above and some completely novel,such as SMS texting. For each one, teachersand learners will need some initial guidancewhile intrinsic or indigenous ethics and etiquetteare developing.

A rather different perspective on the culturalexpectations governing interactions andexchanges comes from the growing literature(Katz and Aakhus 2002) that looks at the ways inwhich mobile devices, usually mobile phones, areredefining judgements about aspects ofinteraction in public spaces. In some respects,mobile phones enable users to define privatespaces and private interactions in hitherto publicspaces. Mobile learning will inherit thesechanged attitudes and perhaps define spaces forlearning that had not previously existed.

4 Conclusion: the practicalitiesIn practice of course learning will almost alwaysbe delivered by a blend of mobile andconventional technologies and consequentlysome or all of the difficulties associated with‘pure’ mobile learning will not arise.

The use of mobile learning to supplement orenrich mainstream campus-based learning incolleges and universities is probably ethicallyunproblematic.

Distance learning is perhaps the most ethicallyproblematic of the standard formats used inuniversities and colleges, especially if studentsare dispersed and geographically remote. Similarissues may arise with part-time students at adistance, particularly if mobile learning is onlyblending with online learning.

5 ReferencesAnderson T and Kanuka H (2003). E-research –methods, strategies and issues. Boston, MA:Allyn and Bacon.

BACP (2002). Ethical framework for good practicein counselling and psychotherapy. Rugby: BritishAssociation for Counselling and Psychotherapy.

Buchanan, E (ed). (2004). Readings in virtualresearch ethics – issues and controversies.London: Information Science Publishing.

Hewson C, Yule P, Laurent D and Vogel C (2003).Internet research methods. London:SAGE Publications.

Hine, C (2000). Virtual ethnography. London:SAGE Publications.

Katz JE and Aakhus M (eds). (2002). Perpetualcontact – mobile communications, private talk,public performance. Cambridge: CambridgeUniversity Press.

Warner S (2004). Contingent morality andpsychotherapy research: developing applicableframeworks for ethical processes of enquiry.Journal of Critical Psychology, Counselling andPsychotherapy, 4(2), 106–114.

Weatherall A, Gavey N and Potts A (2002).So whose words are they anyway? Feminismand Psychology, 12(4), 531–539.

Traxler Bridges 207

AbstractIn the light of the increasing need for highereducation, recent technological developments,and the vast and logarithmic developments in alldisciplines, this research reports on theoutcomes of an implemented mobile technologyproject and provides some guidelines for futuredevelopment and implementation of mobiletechnology at the Tshwane University ofTechnology (TUT). Emphasis is placed on thelecturer as primary educator, and the student aslearner, as well as on the responsibility they bothhave in the educational process. Acceptance ofthis responsibility demands thoroughintrospection and a re-engineering of theteaching, learning and technology processes atthe TUT. This results in a new teaching, learningand technology strategy wherein the roles of thelecturer and student change. In the changedroles, the lecturer and the student are managers.

The lecturer accepts responsibility for teachingand the student becomes his or her own learningmanager. The WWW, WebCT as our learningmanagement system (LMS) and a PDA (ourmobile device) were the main instruments instimulating student self-activity.

Implementing mobile technology as part of the multimode (blended) teaching and learning academic strategy @ the Tshwane University of TechnologyProf H J van der [email protected]: Telematic EducationTshwane University of TechnologyPretoriaSouth Africa

The paper addresses:

p the changed role of lecturer and student in a neweducational model

p teacher centred instruction vs studentcentred learning

p the value of mobile technology

p the outcomes of the nature conservationm-learning project

p some criteria for m-learning to be successful andto create a lifelong learner (and lecturer!)

p training, motivation and support for facultyand students

p the instructional design process.

Keywordsmobile learning (m-learning), blended learning,mobile technology, lifelong learning (LLL),instructional design, web-based instruction,learner centred learning, learner managedlearning, multimedia

van der Merwe 209

AbstractMobile learning as a concept has come aboutthrough the availability of mobile technologies.A number of questions about the phenomenonof mobile learning are raised: how does mobilelearning integrate with other learning activities?When in the day does it happen and where?What does it relate to? What portion of everydaylearning does it constitute? This paper outlinesthe first stages of an investigation aimed ataddressing these questions through a study ofeveryday learning episodes.

Keywordsmobile learning, everyday learning, diary-basedstudy, learning episodes

IntroductionMobile learning as a concept has come about byvirtue of the availability of mobile technologiessuch as mobile phones and handheld computers.New learning practices are emerging, such asbitesize revision and just-in-time learning. Unlikemore traditional forms of learning, mobilelearning has not so far been studied as aphenomenon in the world; rather, it has beenintroduced as a new technology-led practice thatwill potentially lead to new learning phenomena.The research reported in this paper aims to fill inthis gap by studying the conduct of mobilelearning in relation to everyday learning. Thestudy is currently being undertaken as part ofthe MOBIlearn project at the Universityof Birmingham.

A study of mobile learning as part of everyday learning

Study methodsThe aim of the study is to uncover how peoplelearn on the move or outside their normallearning environment, with the technologies thatare currently available, such as mobile phonesand PDAs. A diary-based method of recordingdata has been adopted for the purposes of thestudy and participants are asked to keep a diaryof their everyday learning episodes. The diary isavailable as a paper template, as a desktopapplication, or as a set of web-based forms. Allthree formats have the same structure. Theparticipants are required to fill in, for eachlearning episode, information about:

p temporal context: the date; the time spanduring the day when the learning took place;and its duration

p social context: the other people whowere involved in the episode and the rolesthey assumed

p situational context: the location and event duringwhich the learning episode took place

p educational context: any learning method thatwas employed; the forms of assessment applied;the purpose, if there was an explicit purpose;what was learned in relation to what it wasoriginally intended should be learned; and thearea in life to which this episode relates (work,hobbies, community work)

p activity context: the learning topic; the kind ofsupport that was available in terms of help fromother people, printed or online manuals and otherresources; the different activities that wereperformed; the different resources that wereused; the problems that arose before, during orafter the episode; and any greater learningproject that this particular episode related to

p historical context: other activities, not directlyrelated to the learning, that were performed justbefore, during and immediately after the learningepisode, to capture how learning interleaves withother, everyday activities.

Vavoula Sharples O’Malley Taylor 211

Claire O’[email protected] of PsychologyUniversity of NottinghamUniversity ParkNottingham NG7 2RDUnited Kingdom

Josie [email protected] UserlabOpen UniversityWalton HallMilton Keynes MK7 6AAUnited Kingdom

Giasemi N VavoulaMike [email protected]@bham.ac.ukUniversity of BirminghamEdgbaston B15 2TTUnited Kingdom

The instructions encourage the participants toreport all their learning episodes, irrespective ofwhether they are mobile or not, so that we caninfer what portion of everyday learning is mobile.To characterise a learning episode as mobile, wewill use the situational and activity context data;episodes that involve the use of mobilecomputing or communication technologies, orwhich take place away from a fixed environmentsuch as a person’s workplace or home, whetheror not mobile technologies are involved, will becharacterised as mobile.

The study has begun and will continue until theend of the MOBIlearn project in December 2004.Initial data has been collected and is beinganalysed, the results of which will be presentedat the conference.


AbstractStudents use mobile devices to document theheritage. Once documented they use theirphones to upload the information to the e-guideusing an application based on a phone webbrowser – Nokia 3650, Nokia 6600 and EricssonP900 were used in the exercise. Documentingthe heritage was a combination of using GPS,vector maps, satellite photos, phones supportinga web browser, GPRS network and anapplication that connect the existing e-guide(www.chimer.org) with the new information.

Keywordsmobile, children, CHIMER, 3G

CHIMER phone web browserplatformThe Chimer Project (www.chimer.org) developedand implemented applications for mobiletechnologies in a cultural heritage environment.Children from different educational areas – fromrural schools of Spain where mobile networks arethe only network available, Chanovice in theCzech Republic, Bad Doberan in Germany, VilniusLithuania and Den Helder in Holland – are usingthis platform for content creation on the fly in realtime using mobile devices such as Nokia 3650,Nokia 6600, Ericsson P900 and pocket PCs.Device portability is a key point in theCHIMER Project.

The information can not only be created but beretrieved through any pocket PC, tablet PC andthe new generation of Ericsson, Samsung andNokia 60 series of Java-based mobile phones.Video, audio, photos, text, vector maps, routesand satellite images can be retrieved.

The information is connected to a geographicinformation service (GIS) based on Flashtechnology that enables users to interact in realtime with vector and satellite images, andassociate information with any point of the map.

Using mobile phones and pocket PCs for creating, delivering and retrieving content on the fly, based on mobile web browsersDaniel [email protected] Coordinator of Chimer ProjectCiberespacio SLRua Atenas 515009 A CoruñaSpain

Besides the map and the information is an openplatform enhancing the competence of childrenin the use of leading technology andcommunication systems, in order to create,search and modify networked informationresources. They can share, explore and exploittheir local cultural heritage with children fromother European countries. Educational andmuseum experts have provided the informationto the children while teachers worked to coachchildren to research the problem of contentcreation using mobile tools.

The project is seen as a task force that is beingused to solve a problem. In one way CHIMER is amultinational company operating in differentcountries in its mission to document the heritagethrough mobile technologies. It creates a criticalmass of information to help identify the problemit has in documenting the heritage.

The project is fully operational on GPRS, WIFIand UMTS platforms. It has been successfullytested in different environments and is workingat 100% capacity.

This paper will show how to create content usingmobile devices, and how to upload and retrievethe information using an application based on aphone browser using a GPRS connection inroaming. A live demo will be conducted duringthe conference, which will allow participants tocreate their content on the fly, upload it andretrieve it with any phone that supports aweb browser.

ReferencesWeiss D (2003). The summer school book.

Weiss D (2004). Chimer goes mobile.At www.cultivate-int.org/issue8/chimer/

Weiss 213

215

International Workshopon Mobile Learningfor Emergency Management(MLEM) papers

AbstractIn major emergencies, the medical team is movedto the site of the emergency instead of waiting forthe casualties at the casualty ward. Ensuring fasttreatment in administering first aid and stabilisingthe casualties before they are transported to thehospital is of major importance for saving lives. A system for training doctors for such situationshas been developed and implemented on a PC.The system will allow the doctor to make use ofequipment that is accessible from a handbag thathe or she brings to the site, and he or she will betrained in taking decisions related to this specificstand-alone situation without having thecasualty-ward team directly available.

Keywordsmobile training, medical first aid,emergency management

1 IntroductionIn order to maintain the skills acquired duringprevious education, it is important to carry outtraining sessions, especially ones related tosituations that might be unusual compared to thedaily routine. This happens, for example, in criticalsituations experienced by operators whonormally work in controlled environments, butoccasionally and rarely, find themselves in themiddle of an emergency.

The most efficient way of learning how to handle a critical situation and of maintaining the existingskills is by frequent management of real-lifesituations. The next best way to learn efficiently andto maintain that learning is by full-scale exercises,in which the technical and human resources are thesame as in a real situation, but the role of victimsis taken by stand-ins. However, real-life emergencysituations are, fortunately, not frequent andfull-scale exercises are extremely costly in termsof time, and human and financial resources.

A mobile first aid training system for training doctors in stabilising casualties on siteVerner [email protected] Analysis DepartmentRisø National LaboratoryDK 4000 RoskildeDenmark

Therefore, easier and more cost-efficientprocedures for such training are needed; and acomputerised training system for trainingmedical staff for on-site emergency operations –partly funded by EUREKA (a network formarket-oriented R&D) – has been developed byRisø National Laboratory and IFAD (providingsoftware design and development expertise), andimplemented and put on the market by IFAD. Thetraining system has been developed formulti-user training (Andersen 2001); however, thesystem may also be used as a single-user mobiletraining system.

From a mobile learning point of view, two issuesare highlighted: the ‘training anywhere andanytime’ aspect and the ‘mobility of people’aspect. As regards the first, the system –installed on a laptop – may be seen as a mobiletraining set-up for individuals, to be used fortraining anywhere and anytime. In terms of thesecond issue, having the system installed in anetwork has solved the mobility problem. Thisallows the trainees either to be placed in thesame training location in order to support thesocial aspects of the training session; or to beplaced anywhere, geographically, in relation tothe network and still address the team trainingaspects of the same scenario.

2 The training systemThe realism of the system lies in a sharedinteractive scenario, in which actions taken bythe acting participant will influence the scenarioand be reflected in the presentation of thescenario, as in a real-life situation.

The training system must be adaptable tovarious situations in accordance with theactual type of emergency. When a medicaltrainee is faced, for example, with a minoremergency or a situation where only onemedical doctor has arrived on the scene,

Andersen 217

he or she will have to give first medical care tothe victims in the order in which he or she findsthem or they are presented to him/her, andstabilise them until transport to a nearby hospitalis available. As a multi-user training system, thesystem will – in addition – support the coordinatedtraining also for the doctor in charge of themedical team, the doctors responsible for triage,and the group of doctors administering first aid.

3 The patient modelIn the simulated world, the medical personnelmust have some (pseudo)-realistic impressionsof the synthetic casualty patient in order to reactproperly. He or she must be able to ‘see’ thepatient, ‘talk’ with and ‘listen’ to him or her andbe able to perform ‘examinations’. At the sametime, other types of information concerningevents in the surrounding scenario must bepresented to help or to distract the medical mind.The aim of the system is to teach the traineehow to work in unusual surroundings, identifypatient reactions, make fast decisions, and sendand receive informative signals to and from otherlevels in the system, all at the same time.

The casualty patient is presented to the traineeon the computer screen as a combination of textand a simplified figure (see Figure 1 opposite).

If more than one casualty has arrived at the sametime, it is possible to see them all in one view,but the figures are proportionally smaller. Thedrawing shows the visible signs and injuries; forexample burns, bleeding, and pale skin. Only oneside is visible at a time. The patient is fullydressed, but may be undressed if needed.Text explains the condition; for example, vomitingor shivering.

If the patient is conscious and able to talk,a ‘spontaneous dialogue’ mimics thecommunication with the medical person. Inputfrom the scenario appears continuously. This isthe platform for the medical reactions.

From menus on the screen, the trainee can askthe casualty patient about his/her condition andget answers (dialogue).

A wide range of medical examinations can bechosen from another menu. The resources (egmaterials and skilled personnel) might be limited,and therefore make it impossible to perform thedesired procedures.

The results of the different medical tests taketime, and during the process, the system can donothing else in order to mimic the constraints ofthe real world. When the result appears, it isshown on the screen for a few moments andthen disappears. No curves or tendencies areshown. It is the trainee’s job to keep track of theimportant information. Only parameters that areasked for are shown.

Medical treatment can be chosen from themenu. Time, available instruments and availablepersonnel can be used in the treatment ifnecessary. The effects of the therapy are relatedto the patient model, and have an influence onthe condition of the patient. Some types oftreatment have a positive influence; others canbe wrong or have no effect.

Other reactions – for example, orders toevacuate the casualty patient to the next level –can be selected from the menu. As in the otherforms of menu work, the computer will notallow you to do something that is not possiblein terms of the scenario (eg there might be a lackof ambulances).

The trainees themselves choose if they want tohave an overall view of the environment –revealing the complete number of casualties –or if they want to approach a specific casualty forcloser inspection. Figure 1 (opposite) shows agraphical presentation of the latter, indicating anumber of possible actions and highlightingspecific features concerning measurements andobservations. Likewise, Figure 1 (opposite)indicates the human and technical resourcesavailable to the ‘doctor’.

Examples of potential actions to be taken by thedoctor are:

p communication with the casualties, making useof standardised questions, in case the casualty isable to communicate despite the injury

p possibility to zoom in on the casualty for betterview of the damages

p possibility of undressing the casualty for betterview of the damages

p turning the casualty for a complete inspection

p checking the colour and humidity of the skin

p measuring the pulse

p checking the eye reactions

p classifying the casualty for further treatment

p measuring the respiration by quality or frequency

p measuring of blood pressure

p checking the capillary response.


They may also carry out treatments such as:

p giving morphine

p attaching a drip

p putting on a bandage

p adding support

p stabilising the casualty for transport to ahospital selected according to expertise,capacity, and distance.

Figure 1 Layout of the presentation

4 Scenario executionA number of scenarios are pre-planned, and maybe chosen in order to start the session for thetrainee. During the session, the trainee mayexperience the full sequence of events as in a realemergency situation. If so planned in the scenario,this goes from the very moment when the alarmgoes off to organising the transport by car orambulance. During the virtual tour on the way tothe emergency site, the trainee may get furtherinstructions about the emergency, and she or hemay mentally prepare her/himself for the situation.

5 Results and evaluationAfter the session, a detailed log can be producedof the medical reactions; for example, time spenton each patient, use of different types ofresources, and the outcome for each patient.

The medical effects in the simulated system canbe compared with the clinically widely acceptedAbbreviated Injury Scale (AIS 1990) and the InjurySeverity Score (ISS; Baker et al. 1974). Bothtechniques are based on an anatomical approachas in this simulation system.

Likewise, graphs may indicate the patient’s likelydevelopment without any treatment and with thetreatment performed, and both options can beassessed against the likelihood of a fataloutcome to the situation.

6 ConclusionThe system has been tried out by doctors in thenine somatic hospitals in Greater Copenhagen,partly replacing and partly ensuring betterpreparation for real on-site exercises. Theoutcome has been very promising, especiallyin terms of the increased number of traineesput through training sessions as comparedwith the 20–30 trainees per year, which wereall that constraints of time and resourcespreviously allowed.

ReferencesAIS (1990). The Abbreviated Injury Scale.1990 revision. Des Plaines, IL: Association forthe Advancement of Automotive Medicine.

Andersen V (2001). Training of medical teamson-site for individual and co-ordinated responsein emergency management. International Journalof Emergency Management, 1(1), 3–12.

Baker SP, O’Neill B, Haddon W and Long WB(1974). The Injury Severity Score: a method fordescribing patients with multiple injuries andevaluating emergency care. Journal of Trauma,14, 187–196.

IFAD (2005). Website. At www.ifad.dk

Andersen 219

AbstractThe paper illustrates how systems developed inthe past as platforms for emergency training andpreparedness plans could, if redesigned forutilisation via the newly available mobile computingresources, offer more opportunities to deal withnovel and unusual emergency situations.

A schematic view is also proposed as to how theroles and duties of the emergency organisationscould change, and become more efficient tohandle emergency situations.

Keywordsmanagement of novel situations, emergencyorganisations, emergency coordination,mobile training

1 IntroductionNowadays, emergency managers must handlesituations that are often new, unusual anddifferent from the traditional types of crisis forwhich they are trained. The increased complexityof countries’ infrastructures and the excessiveexploitation of environmental resourcesproduces novel and emergent types of crisis.

Consider, for example, the wide range ofimprovised activities that would have beennecessary after the catastrophic collapse of theWorld Trade Centre following the terrorist attackon 11 September 2001 (Kendra and Wachtendorf2002). Nobody could have imagined such ananomalous and unexpected situation.

Apart from this terrible event, in the last fewyears, many regions have also been involved inmany small crises caused by the loss offunctionality of network information systems onwhich people and industries are dependent foreveryday life.

Management of novel situations throughmobile learning resourcesClaudio [email protected] (Italian National Agency for New Technologies, Energy and the Environment)Via Anguillarese 30100060 RomeItaly

Using the internet, we increase our connectivitywith many other people around the world. On apositive note, the organisation of this conferencetoday has probably been realised at a lower costby exploiting the great connectivity capabilityavailable through the Web. The downside is thathackers and malicious people use the samee-mail channels to make their attacks worldwide,creating viruses that replicate themselves insidethe mailboxes of e-mail users. These types ofevent and emergency are also novel andunexpected. Traditional techniques for detectinganomalies are unsatisfactory because theyidentify individual data points that are rare due toparticular combinations of features.

In this context, creativity seems to be today animportant element of successful disasterresponse. The literature of creativity is vast,spread among the arts, psychology, business andmanagement, and philosophy. Clemen (1996)describes creativity as:

…new alternatives with elements that achievefundamental objectives in ways previouslyunseen. Thus, a creative alternative has bothelements of novelty and effectiveness, whereeffectiveness is thought of in terms of satisfying[the] objectives of a decision maker, a group ofindividuals, or even the diverse objectives held bydifferent stakeholders in a negotiation.

Given these factors, it becomes difficult to trainemergency managers and other personnel tohandle the new real emergencies usingtraditional methods such as the ‘on-site’exercises that are normally practised in advance,or best practice manuals including norms andprocedures designed for well-known types ofcrisis and disaster.

Balducelli 221

2 Management ofemergency activityThe management of emergency activities isdifferent from the management of activities, forexample, inside a factory. In the emergency,novel situations and contingencies frequentlyarise that have to be handled by low-levelpersonnel without the help and the support ofhigher-level managers.

If we represent in a single schema all themanagerial activities inside a factory, we have thepicture in Figure 1 – a pyramid, in which thestrategic decisions are taken at the top; in themiddle, we have some kind of directional controlof the low-level activities; and at the bottom, thereal operative activities are decided on.

The strategic activities are developed during amedium-length or long time period; themiddle-management activities define themedium- or short-term objectives; and theoperational activities realise these objectives inthe real activity domain.

Figure 1 Schema of managerial activities inside a factory

In such a type of framework, the operationalactivities of low-level management are realisedby a certain set of rules or procedures whoseselection is determined by control and strategicobjectives defined at the higher level. In this way,the low-level personnel have only theresponsibility for the execution of a certaindecision and not for its control and selection.

During emergency management, the previousschema may no longer be applicable, especiallywhen new and unexpected phenomena occur. Insuch cases, a predefined strategy to be appliedduring new contingencies may be not available,and the low-level emergency personnel have tochoose a different strategy; due to time andgeographical constraints, it may be difficult to askmiddle and top management for a new strategy.


3 Emergency training exercisesTraining exercises, within a real or simulatedemergency domain, could help the low-levelemergency personnel to be prepared to handlethe emergency situation. They could learn moreabout the availability of emergency resources,the possibility of resource sharing, the need forcoordination with other operators, etc.

On-site exercises are carried out inside criticalplants or infrastructures, by a simulation of themethods and resources planned for use in thecase of emergency. All low- and high-levelmanagers who participate in the exercises areasked to contribute to the simulation of thewhole emergency scenario.

The main advantages of this type of exercise areto check the availability and working condition ofall the needed resources, but the greatdisadvantage is that every person involved isnormally advised beforehand, so that no type ofnovel situation can be introduced and lack ofcoordination is not verified. In addition, the costof such exercises is very high and for this reason,they cannot be repeated more than once.

Emergency manuals, containing the norms andthe procedures that must be applied during thecrisis are often distributed to emergencyorganisations and also to the people that couldbe involved in a particular disaster due to thelocation of their home or to their type of work.The disadvantage here is that novel situationscannot be ruled out and this type of manual maybe not available for a rapid consultation duringthe emergency itself, because people do notalways take such manuals with them.

In the past, many systems have been developedto support a sort of ‘virtual’ training of emergencymanagers and operators (Balducelli, Bologna andVicoli 1996). Computer platforms based oncomputerised emergency procedures, modelssimulating accidents involving fire, smoke andthe release of toxic substances, andvisualisations of emergencies using virtual realityhave all been developed and tested.

Such types of system may give support not onlyduring the training phases, but also during a realemergency, as, compared to a simple manual,they include simulation codes and virtualscenarios that can be configured using theparameters and the variables of an incomingemergency event. The problem is that normallyduring emergency contingencies, constraints ontime would not allow the setting up andconfiguring of a system that is used rarely and fortraining purposes only. To be ready to use suchsystems during accident conditions, theemergency managers and operators would needto use it normally also during everyday work, butthis may be impossible because they arenormally static systems, not mobile.

The recent availability, however, of ‘mobile’techniques supporting all types of computeriseddevice, such as portable computers and mobilephones has greatly increased the possibility ofusing emergency training systems morefrequently. Compared to the first technologies,such as GSM (Global System for MobileCommunications), new technologies like GPRS(General Packet Radio System) based on TCP/IPhave greatly increased the data transmissionspeed of mobile devices (from 9.6 Kbps to 124Kbps). Many systems could be transformed fromstatic to portable and it becomes possible to usethem also during real emergency situations.

This new situation will be useful for emergencyorganisations that must be present on theground at operative level.

4 Mobile emergency managementcontainersThe new working methods to addressemergency situations are supported by theincreasing availability of emergency knowledgecontainers that can be queried and exploitedduring real operation. As pictured in Figure 2(overleaf), two main classes of container maybe available: the tools container and theknowledge container.

Balducelli 223

Figure 2 Interconnectivity between emergency managers and emergency trainees

4.1 Tools containersMany tools are useful duringemergency management.

p A GIS (Geographic Information System) toolconnected with a GPS (Global PositioningSystem) communication service is useful in orderto have a ‘real time’ topological view of theterritory involved in the emergency.

p Tools simulating the spread of fire or smoke,radiation levels, predicted traffic congestion, etccould also be useful resources running on top ofthe GIS system.

p Tools to communicate alarms or to check thenumber of people present inside the areainvolved may be also available: the majority ofpeople today have a personal mobile phone thatcan be detected and contacted inside theemergency location.

p Tools including advanced algorithms able toanalyse the incoming emergency data couldbe also developed and included insidetools containers.

4.2 Emergency knowledge containersSeveral types of knowledge are useful duringemergency management:

p databases containing useful data for makinganalysis and evaluations of the different types ofaccident scenario

p procedures to be applied by the emergencyorganisations during the crisis management andbest practice norms that must be followed by thepopulation during the emergency phases

p photographs, movies and every other type ofdigital document and map that can be useful inhandling the emergency contingencies.

As pictured in Figure 2, the new mobiletechnologies permit a new type of mobileconnectivity layer that allows for transportation ofall relevant data and tools from the serviceinformation layer towards the applicative layerthat is, in this case, represented by the on-sitepersonnel, emergency authorities,instructors/experts and trainees.


4.3 On-site personnelOn-site personnel are generally emergencycoordinators who must operate near theemergency location (eg the Fire Brigade). Theyhave to ask off-site emergency authorities aboutthe resources and the methods they think are (orwill be) necessary to handle the emergency inthe short or medium term. They decide on-sitethe best operative strategy to adopt in themanagement of the emergency resources.

Information and data collected by on-sitepersonnel regarding the actual emergency stateis very important and must be transmitted toemergency coordinating authorities.

It is obvious that for this type of personnel, radiotransmitter–receiver devices needed tocoordinate the work of other personnel involvedin the emergency are very important. At thesame time, they need a continuous connectionlink with the emergency authorities.

Radio transmitter devices are better than mobilephones as they use dedicated communicationfrequencies that are generally not congestedeven during an emergency. However, thedisadvantage of these radio transmitters is thatthey do not permit communication exchangesbetween different public safety agencies: in fact,in many cases, they are based on incompatibletechnologies from different vendors.

Also, unlike mobile phones, they do not allowinternet connection capacity which could beimportant in many cases.

In the last few years, new portablecommunication technologies have beendeveloped with more standard communicationprotocols; for example, TETRA systems(Vilppunen 2003) which provide users withstate-of-the-art features such as packet dataapplications, internet access, full duplextelephone interconnect and sophisticatedend-to-end encryption. New-generation handsetswill be ergonomically designed, compact,lightweight and have added functionalitiessuch as built-in GPS, a Smart Text Editor andlarge ideographic displays that enablecustomised languages.

4.4 Emergency authoritiesEmergency authorities are emergencycoordinators that must manage/prevent thepotential consequences that an emergency posesto the population at large, if emergency conditionshave spread outside the location where it initiallyarose. In some cases, they must inform thegeneral public using all types of availablecommunication device. To carry out this task, it isvery important for them to have the informationcoming from the personnel in the field regardingthe emergency situation. For emergenciescovering a wide area, they have to coordinate thesharing of resources between a number ofpersonnel in the field. They have also, in somecases, to call the emergency experts who maybe located far from the emergency location andquickly ask them the same questions.

In addition to communication devices, it is crucialfor this type of emergency coordinator to makeuse also of computing devices furnished withemergency support systems.

4.5 Experts and instructorsThe availability of the new mobile technology isanalysed in an interesting work (Hansewinkel andBerggren 2003) where the authors discuss adual-use possibility – how consumer technology,bought for recreation and entertainment, couldalso be used as tools for experts working fromhome. Many people today use entertainmentand recreation technology at home, likewide-screen TV, surround sound and videointegrated into a sort of home theatre concept;they have also internet connection, e-mailservices, web digital camera and so on. A fewyears ago this technology was considered to bestate of the art, reserved for governments,research centres and high-tech corporations.

Although this new technology is normallyavailable at home, when it comes to vital socialservices such as the education of children inremote areas, long-term medical treatment,rehabilitation and telemedicine, it is more oftenused in public and semi-public buildings.

Hansewinkel and Berggren (2003) make the pointthat ‘surprisingly little effort has been made to usethese technologies for society [social] services,in the context it has been developed for –the homes, by persons interested in thisnew technology, who often buy it anyway fortheir recreation’.

There is, in particular, one area where thistechnology has the potential to be used for vitalsocial services: the potential for emergencyexperts to ‘work from home’.

Balducelli 225

This technology may be of special interest toemergency management experts who couldparticipate in a virtual team to support societywhen it is under severe strain from emergencysituations. There is the possibility of arrangingweb conferences, making this entertainment andrecreation technology useful for the experts whomust be involved during an emergency,especially if they reside a long way from theemergency location.

These technologies could also be useful foremergency instructors who could arrangecourses for personnel in the field andemergency authorities.

4.6 Emergency traineesUsing these new technologies, the gap betweenthe emergency simulation for training and thereal emergency conditions will be reduced. Thesimulation of emergency training scenarios,created by the same entertainment andcommunication/computing technology that isused at home and in offices during the normalworking day, will be more easily accepted byemergency managers, experts and authorities.

Computer-simulated training exercises that couldalso be carried out at home would be less costlyand could be undertaken more frequently by thetrainees. The exercises can be logged andexecuted more times in different locations fordifferent types of trainee. Availability of moresophisticated mobile devices will allow us in thefuture to carry out part of these computer-basedexercises at real emergency sites, so that thedifferences between on-site exercises andsimulated ones will be reduced. This has positiveimplications for a more extensive use ofemergency support systems and alarm systemswhich are able to reach people living near the siteof an emergency. In fact, for an effectiveutilisation, the trainees need to be very familiarwith what the support systems can do, and thistype of knowledge can be acquired only by afrequent use of systems of this type.

5 ConclusionsThe availability of new mobile computingtechnologies will in the future affect the roles,duties and powers of all emergencymanagement personnel. The evolution of mobiletechnologies inside the emergency managementorganisations will proceed in the same way as itdoes inside other types of organisation orworking group. One of the main difficulties willprobably reside in the fact that knowledge andpossession of the new mobile computingresources are unevenly divided between youngand older people, but this is more a commongenerational problem that exists in every type ofactivity in our society.

ReferencesBalducelli C, Bologna S and Vicoli G (1996).Evaluating the efficacy of virtual scenariosgeneration to support the training of emergencyorganisations. In Proceedings of the 8thEuropean Simulation Symposium: Simulation inIndustry, 24–25 October, Genoa. SCS, 108–112.

Clemen RT (1996). Making hard decisions:an introduction to decision analysis, 2nd edn.Pacific Grove: Duxbury Press.

Kendra J and Wachtendorf T (2002). Creativityin emergency response after the World TradeCentre attack. In Proceedings of the 9th AnnualConference of the International EmergencyManagement Society, 14–17 May, Waterloo,Ontario. TIEMS, 69–81.

Hansewinkel H and Berggren P (2003).Experts@home. In Proceedings of the10th Annual Conference of the InternationalEmergency Management Society, 3–6 June,Sophia-Antipolis, France. TIEMS, 315–325

Vilppunen H (2003). The communicationpreparedness of Finland for emergencies anddisasters. In Proceedings of the 10th AnnualConference of the International EmergencyManagement Society, 3–6 June,Sophia-Antipolis, France. TIEMS, 419.


AbstractTelemedicine now has many applications in thefield of hospital activities, but few have beenrealised in the context of isolated sites, or mobileor disaster situations, when the basicinfrastructures are no longer available. This lackhas now been addressed with the I–DISCAREsystem based on satellite technologies.

1 BackgroundIn 2001/02, thanks to grants from the EuropeanSpace Agency’s (ESA) ARTES 5 Programme, inthe context of the DELTASS (Disaster EmergencyLogistic Telemedicine Advanced Satellite System)project, engineers and medical doctors havesucceeded in developing, integrating anddemonstrating – via full-scale simulations – theutility of satellite systems and telemedicine incatastrophic situations. The industrial team of theproject received recommendations from a‘medical group’ involving European specialists indisaster and emergency medicine. After theDELTASS pilot phase, the I–DISCARE system isnow entering its deployment and utilisationphase, which is funded by ESA.

Logistics and medical support for isolated, mobileand disaster situations: the I–DISCARE concept

2 I–DISCARE capabilitiesIn the context of emergencies like earthquake,flood, war or individual medical emergencycases, it is necessary to take the right and rapiddecision concerning the triage and evacuation ofcasualties. For this to happen, the peopleresponsible for this job need to have a clear viewof the casualties’ medical status and theactivities of the teams operating in the field. TheI–DISCARE system offers this clarity of vision atthe three levels of the operation: search andrescue (SAR), evaluation of the casualty at theadvanced meeting point (the place where thecasualties are brought together beforeevacuation); evacuation of the casualty to thecare centres. The I–DISCARE system allowspersonnel to follow (with timing and maplocalisation) the casualty hand-over, and amedical field card is generated both in apaper format and in parallel in a tele-transmitteddigital format.

The general I–DISCARE system architecture (seeFigure 1 overleaf) demonstrates this capability.

Larsen Bove Comet 227

Eli [email protected] Centre for Telemedicine (NST)University Hospital of North NorwayPO Box 35N–9038 TromsøNorway

Antonio [email protected] Giulio Vincenzo Bona 8700156 RomeItaly

Bernard [email protected] de Medecine et Physiologie Spatiale (MEDES)Clinique SpatialeBP 7440431405 ToulouseCedex 4/France

2.1 The PDA terminalsThe PDA terminals (shown in Figure 2) equip theSAR teams and allow them to forwardinformation to hospital personnel on the localeand identity of the casualty (via a predefined5-letter code read on the paper medical fieldcard), the level of gravity of their injury and apreliminary diagnosis. This equipment alsopermits phone communication between thedifferent personnel via the Globalstar system.

Figure 2 The PDA terminal

2.2 The portable telemedicineworkstationsThe portable telemedicine workstations(see Figure 3 opposite) equip the medical staffat the advanced meeting point and allow thetransmission of objective medical data – 10-leadelectrocardiogram (ECG), non-invasive bloodpressure (NIBP), blood oxygen saturation (SaO2),heart rate (HR), body temperature (T°) – and thelocale of the casualty and medical staff [by GlobalPositioning System (GPS) service], alwayskeeping this data attached to the unique casualtyidentification. The Globalstar satellite mobilephone can also be used for voice communicationbetween the different personnel.


Figure 1 I–DISCARE system architecture

Figure 3 The portable telemedicine workstation

2.3 The ambulance terminalThe ambulance terminal (shown in Figure 4)equips the ambulance that ensures theevacuation of casualties to care centres. Itpermits the transmission of objective medicaldata – 3-lead ECG, NIBP, SaO2, HR, T° – andinformation on the locale of the ambulance,always keeping this data attached to the uniquecasualty identification. Again, the Globalstarsatellite mobile phone can also be used for voicecommunication between the different personnel.

Figure 4 The ambulance terminal

2.4 Coordinating operational centreAt the level of the coordinating operational centre(which can be a permanent centre or a temporarydeployed centre), all the data coming from theSAR team, the team at the advanced meetingpoint and the ambulance teams can be displayed.At each tele-transmission, the position of theteam is attached, which allows a mapping of thedifferent personnel in the field.

The operational management team has apermanent view of the rescue and the ongoingevacuation operations; information on thenumber of victims with their medical status isaccessible. The traceability and history of therescue operations is automatically provided. Themanagement of the rescue operation becomeseasier thanks to the I–DISCARE system.

3 Who are the potentialI–DISCARE users?The complete I–DISCARE system should interestthe medical emergency organisations working indisaster situations, like the Fire Brigade, civilianprotection organisations, humanitarianorganisations and military institutions. Somesubsystem of I–DISCARE can be used in othersituations, as follows.

p The PDA terminals are well adapted forsearch and rescue operations by mountainemergency organisations.

p The portable telemedicine workstations are welladapted for medical emergency situations onboard long-haul flights; in a maritime context onboard boats; for off-shore platforms; for remoteworkplaces; for expedition/traveller assistance inisolated areas; for sanitary stations in remoteareas; for doctors working as generalpractitioners (GPs) in remote regions.

p The ambulance terminals are well adapted forambulance companies working in remote areas.

4 ConclusionsThe DELTASS project has made available acomplete system that will now be deployedwithin the users’ community via the I–DISCAREpilot operations. The proposed system is sizeableand adaptable to users’ specific requirements.

The I_DISCARE project will be piloted on a freebasis, offering the following options:

Larsen Bove Comet 229

p adaptation of the man–machine interfacesoftware of the terminals and consoles tosatisfy the user’s criteria: (language to be used,format and content of the field medical cards,necessary maps)

p a loan of the chosen equipment for a period of2–4 months to try it out: equipment includes:three PDA terminal(s), one portable telemedicineworkstation(s), one ambulance terminal(s), and ifnecessary, one server and one console on theuser’s premises

p training for personnel in how to use it

p installation of the system on their premisesand in ambulances. We ensure the repair andmaintenance of the equipment during the periodof the loan.

The following will be proposed to customers ona commercial basis:

p either periodical or temporary leasing ofequipment to customers (including satellitetelecommunication services subscription, thecommunications cost being covered by thecustomer) with a purchase option. TheI–DISCARE system will be made available anddeployed in the customer premises within48 hours in the European area.

p or a simple purchase of the equipment withsatellite telecommunication services subscription.

For both options, the following services willbe included:

p installation and configuration of the system

p annual maintenance and upgrading of the system

p regular user training.

Further informationFor further information on the project andthe organisations involved, please visit thefollowing websites.

Elsacom: www.elsacom.com

Globalstar: www.globalstar.com

MEDES: www.medes.fr/idiscare

NST: www2.telemed.no/english/nct/index.html

http://telecom.esa.int/telecom/www/object/index.cf m?fobjectid=750

http://telecom.esa.int/telecom/www/object/index.cf m?fobjectid=7564

www.esa.int/export/esaCP/ESACZSTHN6D_Benefits_0.html

www.esa.int/export/esaMI/Telemedicine_Alliance/ESAPX27708D_0.html


Author index

Ahonen, Mikko 109

Ally, Mohamed 5

Almer, Alexander 77

Andersen, Verner 217

Arrigo, Marco 9 11

Arvanitis, Theodoros N 43 115

Attewell, Jill 15

Baber, Chris 43 115

Balatti, Enrico 123

Balducelli, Claudio 221

Barth, Artur 21

Beale, Russell 23 25 43 115

Benta, Kuderna-Iulian 27

Berthet, Jean-Pierre 129

Bertolotto, Michela 135

Bove, Antonio 227

Brasher, Andrew 29 33

Brennan, Attracta 149

Bridgefoot, Larry 39

Bridges, Nell 203

Brundell, Pat 43 115

Bull, Susan 39

Byrne, Will 43 115

Cacace, Filippo 47

Capuano, Nicola 53

Carboni, Davide 163

Chalon, René 129

Chiappone, Giorgio 11

Chionsini, Valentina 89

Christ, Andreas 133

Chua, Hui-Na 85

Cinque, Maria 47

Colley, Jo 57

Comet, Bernard 227

Corlett, Daniel 39 59

Cremene, Marcel 27

Crudele, Michele 47

Da Bormida, Giorgio 143

David, Bertrand 129

Evans, Diane 63

Feisst, Markus 133

Ferscha, Alois 67

Freysen, Johan B 73

Fritz, Gerald 77

Frohberg, Dirk 81

Gaeta, Matteo 53

Gentile, Manuel 9 11

Gardner, Michael 85

Giordano, Stefano 89

Göth, Christoph 95

Grande, Sergio 139

Harrison, Bob 101

Häss, Urs-Peter 95

Hitz, Martin 105

Holzmann, Clemens 67

Iannello, Giulio 47

Insolvibile, Gianluca 89

Jäppinen, Anu 109

Jeffery, Ann 113

Jones, Matthew 25

Kiddie, Paul 39

Kilbride, John 135

Kreis, Gottfried 21

Larsen, Eli 227

Lonsdale, Peter 43 115

Luley, Patrick 77

231

Malliou, Eleni 119

Mangina, Eleni 135

Marianczak, Tom 39

Mazza, Riccardo 123

McArdle, Gavin 135

McGreal, Rory 127

Mercier, Fabrice 129

Miliarakis, Antonis 119

Miranda, Sergio 53

Mistry, Chet 39

Mitic, Jelena 133

Monahan, Teresa 135

Moreno, Anna 139

Murelli, Elena 143

Murto (née Turunen), Hanne 153

Nokelainen, Petri 191

O’Malley, Claire 211

O’Nuallain, Caoimhin 149

Oppl, Stefan 67

Paddeu, Gavino 163

Padurean, Razvan 27

Paletta, Lucas 77

Panza, Gianmarco 123

Pappacena, Laura 53

Pehkonen, Marika 153

Pincas, Anita 157

Pintus, Antonio 163

Piras, Andrea 163

Plattner, Stefan 105

Polese, Pietro 89

Popat, Kris 167

Roddolo, Enrico 123

Rodin, Leon 171

Sandle, Neil 39

Sanna, Stefano 163

Savvas, Stavros 119

Schwabe, Gerhard 95

Seifert, Christin 77

Shahi, Anuroop 85

Sharples, Mike 39 43 59 115 211

Smyth, Gerard 177

Sotiriou, Sofoklis 119

Stead, Geoff 57 167

Stone, Andy 183

Stratakis, Manolis 119

Ströhlein, Georg 187

Syvänen, Antti 191

Taibi, Davide 9 11

Tanhua-Piiroinen, Erika 109

Taylor, Josie 33 63 211

Tegolo, Domenico 11

Tognoni, Dr Carlo 197

Traub, David C 201

Traxler, John 203

Vainio, Teija 109

van der Merwe, Prof H J 209

Vavoula, Giasemi N 211

Venditti, Marco 47

Webb, Keith 113

Webster, Tamatha 15

Weiss, Daniel 213

Williams, Daniel 39


The third annual international mobile learningconference MLEARN 2004 was organised by the MOBIlearn and m-learning projects. The main purpose of the conference was to bring together people interested in developingopportunities, systems and content for learningwith mobile and wireless devices and networksas well as researchers and educationalists with an interest in mobile learning. Speakers and delegates included education and trainingpractitioners, designers of mobile learningmaterials and services, hardware and softwaretechnology developers and researchers frommobile learning projects around the world. The interesting and varied papers in this edited book are based on presentations given at MLEARN 2004 and they provide a flavour of current achievements and work-in-progress in the field of mobile learning.

Conference sponsors

The m-learning and MOBIlearn projects were supported

by the European Commission Directorate-General Information Society

IST–2000–25270 (m-learning) IST–2001–37187 (MOBIlearn)

The m-learning project and this publication were supported by the Learning and Skills Council as part of a grant to the Learning and Skills

Development Agency for a programme of research and development