introduction to interactive computer systems map-i...
TRANSCRIPT
Introduction to Interactive Computer Systems MAP-I -2014/15
Universidade de Aveiro
Departamento de Electrónica,
Telecomunicações e Informática
Beatriz Sousa Santos, 2014/2015
2
• “As popular computing has grown, the role of HCI (human-computer interaction) has increased. Most software today is interactive, and code related to the interface is more than half of all code.”
• “HCI also has a key role in application design. In a consumer market, a product’s success depends on each user’s experience with it. Unfortunately, great engineering on the back end will be undone by a poor interface, and a good UI can carry a product in spite of weaknesses inside.“
• “Innovation in the product is a nice virtue, but it’s an option in terms of marketability. Usability is not.”
Canny, J., The Future of ICH, ACM Queu, Jul.-Aug., 2006, pp. 25-32
3
• “Those of us who deal with user interfaces tend to think primarily in terms of computer programs. But user interface problems in the real world are often worse since the real world is not nearly as malleable as the computer world. An ideal solution, even if we know what it is, might not be practical to implement.”
• “The real world is just as much a nuisance to design for as the computer world, and maybe more.”
Blinn, J, “User Interface Stories from the Real World”, IEEE Computer Graphics and Applications, Jan./Feb, 2005, pp.92-93
“The interface between humans and computers is harder than ever to define, we can interact with computers just by walking through a public space.”
4
Sellen, A., Rogers, Y., Harper, R., & Rodden, T., “Human Values in the Digital Age”, Communications of the ACM, 52(3), March 2009, pp. 58–66
5
The Topic of this course:
Design, evaluation, and implementation of computing systems for human use:
• From a single user interacting with a single computer
• To spaces enhanced with ubiquitous technology to support the activities or needs of multiple users
Course Outline
• Introduction to HCI; usability principles and paradigms (> Prof. Rui José)
• Interaction styles, techniques and devices (> Prof. Paulo Dias)
• Evaluation: traditional and formal methods (> Prof. José Creissac)
• The user: relevant characteristics to the design of interactive systems
• Human Centered Design, interactive S/W life cycle
http://wiki.di.uminho.pt/twiki/bin/view/Education/MAPiICS/
• Students assignments/presentations:
- Usability evaluation of an application, system, device, ...
using “traditional” and formal methods (including report)
- Monographs (only presentation)
6
Icebraker
• Name
• Background (not only but also in HCI)
• University
• Prospective work subject
• Why ICS
• Mainling list
7
Usability Principles and Paradigms
Universidade de Aveiro Departamento de Electrónica, Telecomunicações e Informática
(Norman, 2013)
9
• Introduction
• History
• Usability and standards
• Principles
• Paradigms
10
• During the World War II the following disciplines emerged:
Ergonomy – physical aspects
Human factors – also cognitive aspects
Man-Machine Interaction Human-Computer Interaction (nighties)
11
• Interaction emerged as new independent field within Computing in the 80s, mainly due to: – Lower price of technology
– Technology migration (to be used by “non-computer expert users” )
– Need to increase users productivity
• It expanded rapidly
• It is currently an interdisciplinary field
12
Interactive systems design
• Interactive systems include a “module” which we don´t control:
The user, who:
- is very complex
- is not well known
- we cannot control
This makes design difficult
• To the user “the interface is the system”
• The user interface design involves a considerable effort
interface computer
user
Interactive System
Human factors in User Interfaces
• Limited short-term memory
• People make mistakes
• People have different capabilities
• People have different interaction preferences
We should study the user scientifically
13
The User (postponed to another session)
15
Interactive system design
• Involves knowing:
Usability principles (independent from technology)
Usability paradigms (more technology dependent) • We must know the success examples (usability paradigms)
• Understand why they work (usability principles)
• Use the adequate methods (user-centered approach)
• And evaluate, re-design, evaluate, redesign … until we attain the usability goals
+ Adequate methods + Evaluation
User-centered design main characteristics
• early and continual focus on users and their tasks
• empirical measurement of user behavior (mainly usability tests)
• iterative design
(many alternatives tested with low fidelity prototypes)
16
Low-fidelity prototypes
17
e-mail for elderly people
Shopping application
One-button paint
Shop managment application
e-Doctor
Car dashboard
Even in situations where it seems difficult to think of a paper prototype:
18
Pac-Man game controlled by gestures
User Centered Design Interactive Software Lifecycle (postponed to another session)
21
Usability
• Is directly related to the system capacity to allow users attaining their goals through its usage
• Three fundamental aspects (dimensions): - easy to learn and remember (learnability, memorability) - easy to use (fast and with few errors) (efficiency, efficacy) - satisfaction
Is defined in a context of use: is a system property of allowing specific users to perform specific tasks efficiently with efficacy and satisfaction
• Main usability benefits: • Higher user performance and satisfaction
• Lower development costs
• Lower support costs
– Higher profits for everyone!
22
23
Ian Sommerville, Software Engineering, 9th ed., Addison Wesley, 2010
Usability is a non- functional requirement
http://ifs.host.cs.st-andrews.ac.uk/Books/SE9/WebChapters/PDF/Ch_29%20Interaction_design.pdf
24
Usability standards
• ISO 13407 -> ISO 9241-210 (2010) Human-centred design processes for interactive systems • And others related ISO 13.180 Ergonomics
• ISO 9241-11 (1998) Ergonomic requirements for office work with visual display terminals Part 11 : Guidance on usability
Explains how to identify the information needed to specify or evaluate usability in terms of measures of: - performance - satisfaction
http://www.userfocus.co.uk/resources/iso9241/futureparts.html
To specify or measure usability, we need :
• a description of the intended goals
• a description of the context of use, including:
– users
– tasks
– equipment and environment
This may be a description of an existing context, or a specification of intended contexts
• target or actual values of effectiveness, efficiency, and satisfaction for the
intended contexts
25
ISO 13407
• The goal is to ensure that the development and use of interactive systems take account of the needs of the user and all stakeholders
• applies to software products, hardware/software systems, websites, and services
• describes best practice in user centered design
• provides guidance on design activities throughout the life cycle ICS
• describes an iterative development cycle
26
• ISO 13407 adresses:
• … Four Principles of Human-Centered Design:
- active involvement of users
- appropriate allocation of function to system and to user - iteration of design solutions - multi-disciplinary design
• ... and Four Human-Centered Design Activities:
- understand and specify the context of use
- specify user and organizational requirements - produce more than one candidate design solution - evaluate designs against requirements
27
Other related standards
...
...
29
Some usability paradigms (along the history)
Video Display Unities (VDUs) (1950s) Time sharing (1960s) WIMP (Windows, Icons, Menus, Pointers) (1980s) Direct manipulation (1980s) WWW (1990s) Ubiquitous computing (1990s)
Sketchpad (Ivan Sutherland, 1963)
30 http://www.youtube.com/watch?feature=endscreen&NR=1&v=USyoT_Ha_bA
Alto and Macintosh
31 < Xerox PARC, 1973
^ Apple Macintosh 512KB, 1984
UbiComp Prof. Rui José
32
33
Usability principles (a possible list)
User compatibility
Task compatibility
Work-flow compatibility
Product compatibility
Feedback
Coherence
Familiarity
Simplicity
Flexibility
Control
Technology invisibility
Robustness
Error protection
Usability goals: Easy to learn and memorise Easy to use (effective and efficiente) Satisfaction
34
Examples of the principles use in Windows / Office
35
And in other devices
36
Feedback (desktop) Visibility of the system status
37 console
Digital camera
ON
Feedback Visibility of the system status
39
Simplicity (defaults hide complexity)
Much of the functionality is not directly accessible
Familiarity (profit from the user experience)
40
41
Familiarity Windows Mobile
Android
42
Familiarity Familiar Icons
Familiar Icons
43
Flexibility (let the user choose)
44
Robustness and error prevention
Not accessible (in grey)
45
Robustness and error prevention
Nintendo DS lite
Play Station Portable
Digital camera
Some Usability problems
When the battery is charging the same LED chages from read to green
To daltonics:
The h/w button does not invoke the functionality that is shown on the screen
Usability problems “at home”
I always get confused;
Wouldn´t you?
47
Back light
Front light
Middle light
Up
Down
Usability / User experience (UX)
• The ease in which people interact with a system to achieve specific goals
• The experience a person has when he/she interacts with a product (encompasses all aspects)
Usability -> function
• Positive UX has a greater impact than just usability
48
• User experience (UX) involves a person's:
– behaviors,
– attitudes,
– and emotions about using a particular product, system or service
and also perceptions of system utility, ease of use and efficiency
• may be considered subjective; it is about individual perception and thought with respect to the system
• It is dynamic as it is constantly modified over time
50
Main bibliography
• Dix, A., J. Finley, G. Abowd, B. Russell, Human Computer Interaction, 3rd ed., Prentice Hall, 2003 (or previous editions)
• Shneiderman, B., Plaisant, C., Cohen, M., and Jacobs, S., Designing the User Interface: Strategies for Effective Human-Computer Interaction , 5th ed., Addison-Wesley, 2009 (or previous editions)
• The Encyclopedia of Human-Computer Interaction, 2nd Ed, Aarhus, Denmark: The Interaction Design Foundation. Interaction Design Foundation, 2013 http://www.interaction-design.org/books/hci.html
• Ian Sommerville, Software Engineering, 9th ed., Addison Wesley, 2010 ( chapter 29- Interaction Design) http://ifs.host.cs.st-andrews.ac.uk/Books/SE9/WebChapters/PDF/Ch_29%20Interaction_design.pdf
Other interesting books:
• Norman, D., Basic Books, The Design of Everyday Things: Revised and Expanded Edition , Basic Books, 2013
• Shneiderman, B., Leonardo's Laptop: Human Needs and the New Computing Technologies, The MIT Press , 2003
51
Interaction styles
Universidade de Aveiro
Departamento de Electrónica,
Telecomunicações e Informática
53
Interaction styles
A possible classification:
• Menus • Fill-in-forms • Direct manipulation • Function keys • Question and answer • Command languages • Natural languages • 3D interfaces Often two or more styles are used simultaneously ... In desktop interfaces and beyond
54
(Sneiderman, 2009)
Menus
Universidade de Aveiro
Departamento de Electrónica,
Telecomunicações e Informática
Menus
• The concept existed long before computers
• Selection of options
• There are several types
57
58
Menus: main advantages and disadvantages
Advantages (potential)
• Auto-explanatory
• Do not load memory (recall instead of remembering)
• Prevent syntactic errors
• Visible improvements
Disadvantages
• Not efficient
• Not flexible
• Not practical for many options
59
User profile to whom menus are adequate:
Knowledge and experience: • Low system and task experience • Frequent usage of other systems • Low computational literacy
Work and task: • Low frequency of use • No training • Optional usage • Highly structured tasks
60
Menu design: relevant aspects
• Menu structure
• Option ordering
• Option selection
• Menu invocation
• Navigation
61
Menu design: guidelines
• Adequate the menu structure to the task structure
• Minimize depth increasing breadth (within reasonable limits)
• Use an adequate ordering method
• Be coherent (design, option names, etc.)
• Give selection feedback to the user
• Indicate unavailable options
• Include tooltips if names or icons are not auto-explanatory
• Etc. etc.
Menu design: guidelines
• Find the adequate structure using card sorting:
a low-cost method that helps understanding how users expect to find content or functionality
62 http://www.interaction-design.org/encyclopedia/card_sorting.html
Card sorting example: think about how to sort the fruits and vegetables sold in a supermarket
64
http://www.interaction-design.
org/encyclopedia/card_sorting.html
Direct Manipulation
Universidade de Aveiro
Departamento de Electrónica,
Telecomunicações e Informática
66
Direct manipulation (name coined by Shneiderman, 1982)
• Can be traced down to Sketchpad
• Actions are performed directly on
visual representations of the objects
• It is characterized by:
1- Continuous representation of objects
2- Physical actions instead of command languages
3- Fast, incremental actions with visible results
67
Direct manipulationI does not necessarily imply icons; however, in most
situations they are involved
Example: When a section of a text is selected and draged elsewhere icons are not used, yet an action is performed on a visual representation of an object (text section)
68
Some applications are adequate to use direct manipulation:
One of the earliest commercially available UI using Direct Manipulation (MacPaint)
69
70
• It does not exist a “pure” direct manipulation User Interface (UI)
• Direct manipulation is a quality which may be present in different degrees
• According to Hutchins, Hollan e Norman (1986) a UI has the following aspects:
semantic Distance
articulatory
conversation (command languages) Engagement
model worlds (direct manipulation)
71
Semantic and articulatory distance
• Semantic Distance – subjective distance between the user’s goal and interface semantics
• Articulatory distance – distance between the meaning of the actions and their physical form
72
Semantic Distance
If the objects and actions do not support the users’ goals, semantic distance is high
If the user wants to draw rectangles the application on the left has a smaller semantic distance
,
73
Articulatory Distance
When the physical way actions are performed is more similar to their meaning, articulatory distance is smaller
Dragging an icon corresponds to smaller articulatory distance
Selecting an option corresponds to a greater articulatory distance
74
• According to Wolf e Rhyne (1987) there are two relevant aspects in any user interaction:
Object specification Action specification name generation name generation (write a name) visual correlation visual correlation (select) gesture generation (draw a symbol) analogous action coded selection (write a command)
75
• Specifying objects by visual correlation implies the presence of direct manipulation
• How actions are specified defines the degree of direct manipulation
- direct UI
Objects Actions
visual correlation + name generation
visual correlation
analogous action
+ direct UI
76
Main advantages and disadvantages of direct manipulation UIs
Advantages (potential)
• Easy to learn and remember
• Direct, WYSIWYG
• Flexible, easily reversible actions
• Immediate visual and context feedback
• Less prone to errors
Disadvantages
• Not auto-explanatory
• May be inefficient
• Difficult to draw recognizable icons (particularly for actions)
• Icons occupy more screen real estate then text
77
Knowledge and experience:
• Moderate system experience
• Moderate to high task experience
• Frequent usage of other systems
• Low computational literacy
Work and task
• Low frequency of use
• Moderate training
• Optional usage
• Low structured tasks
User profile to whom direct manipulation is adequate:
78
Direct Manipulation design: guidelines
Minimize articulatory and semantic distance
General guidelines to design a usable UI:
• Coherence
• Good conceptual model
• Feedback
• Adequate organization of functionality
• Adequate screen layout
• Adequate colour usage
• Adequate error handling
• Etc.
79
Coherent Icon scheme
better:
Visual selection feedback
80
Coherent Icon scheme
Melhor:
Adding names
(+ recognizable)
Visual selection feedback
better: same scheme Different schemes:
81
Express relation through icon similarity
Coherence in the icons production scheme
folders
Same type of file
82
Add names to icons to make them more recognizable
Allow name definition
83
Icons must be conceptually and visually distinctive
Conceptually similar icons
84
Icons should be specific/familiar not abstract/non-familiar
Familiar Icons?
85
Express objects attributes through icons
full empty
Better:
Natural language
Universidade de Aveiro
Departamento de Electrónica,
Telecomunicações e Informática
87
Natural language
• Communication between humans and computers
through natural language involves:
- recognition
- generation
• Natural languages as dialog style are not full blown natural languages, they are restricted natural languages
• Natural languages (as dialog style) differ in “habitability” (how easy and natural is it for users)
Note: It still is not possible to maintain
A conversation with a computer as in
2001 A Space Odyssey
Note:
natural language as a dialog style and voice interaction are different things!
88
• Habitability (mismatch between the users' expectations and the capabilities of a natural language) is related to the language domains:
– Conceptual - the set of objects and actions provided by the language
– Functional – what may be directly expressed by the language
– Syntactic – syntactic forms that may be understood
– Lexical - the variety of words that may be understood
• Conceptual model limitations are not very disturbing; however, limitations in any other domain make the language less habitable
89
Example:
• Imagine an information system of a University including a data base with information about employees that may be accessed using a natural language:
- Conceptual domain: information about employees
- The question “What is the salary of the University Restaurant manager?” may be out of the functional domain and imply two questions due to functional domain limitations:
• “Who is the University Restaurant manager?” (answer: Mr. XXX)
• “What is the salary of Mr. XXX?”
- “What is the salary of Mr. XXX?” may not be recognized (due to syntactic domain limitations) even if the information is stored in the DB
- “What are the wages of Mr. XXX?” may not be recognized due to lexical domain limitations if wages does not belong to the language
92
Main advantages and disadvantages of Natural Language dialog style
Advantages (potential) • Powerful • Flexible • Efficient
Disadvantages • Assume problem domain knowledge • Imply clarification dialogs • Imply typing skills (if written) • Improvements are not visible • May create unrealistic expectations, foster irresponsible behaviours and
generate negative reactions • Difficult and expensive to implement
93
User profile to whom Natural languages are adequate
Knowledge and experience High tasks experience Low application experience Low computer literacy High typing skill (if written) Task characteristics Low frequency of use No or little training Optional use
94
Current examples
IBM's CIMA (Conversational Interaction Management Architecture) Mobile phone intelligent personal assistants - Siri for Apple’s iOS - Google Now (2012)
Interesting Links
• http://domino.research.ibm.com/odis/odis.nsf/pages/focus.05.html
(retrieved April 14/2013)
• http://www.apple.com/ios/siri/
(retrieved April 14/2013)
• http://www.ibtimes.com/how-siri-works-iphones-brain-comes-natural-language-processing-stanford-professors-teach-free-online#
(retrieved April 14/2013)
• https://play.google.com/store/apps/details?id=com.google.android.googlequicksearchbox
(retrieved April 14/2013)
95
96
A few Design guidelines
• Provide a (restrict) natural language habitable in all domains
• Define a subset of a (real) natural language using the Wizard of Oz method
• Generate valid outputs concerning the four domains (e.g. always use words that the system recognizes
Wizard of Oz prototyping
• A prototype that only works by having someone behind-the-scenes “pulling the levers and flipping the switches” (named after the classical film)
• A user interacts with an interface without knowing that the responses are
97 http://www.usabilityfirst.com/glossary/
The “wizard” was a “man behind-the-scene”
98
Examples of using the Wizard of Oz method in other situations
- Definition of a set of gestures
to use in games
(Höysniemi et al., 2005)
(Human-Computer Interaction /UA., 2013)
100
3D User Interfaces and interaction devices Prof. Paulo Dias
102
Main Bibliography for interaction styles
• B. Shneiderman, C. Plaisant, M. Cohen, S. Jacobs, Designing the User Interface- Strategies for Effective Human–Computer Interaction, 5th ed., Addison Wesley, 2010
• Mayhew, D., Principles and Guidelines in Software User Interface Design, Prentice Hall, 1992
• Soegaard, Mads, Interaction Styles, 2010 (Retrieved 9 March 2013) http://www.interactiondesign.org/encyclopedia/interaction_styles.html • Hudson, William, Card Sorting. In: Soegaard, Mads and Dam, Rikke Friis (eds.).
"The Encyclopedia of Human-Computer Interaction, 2nd Ed.". Aarhus, Denmark: The Interaction Design Foundation, 2010 (Retrieved 9 March 2013)
http://www.interaction-design.org/encyclopedia/card_sorting.html
103
Interactive system design
• Involves knowing:
Usability principles (independent from technology)
Usability paradigms (more technology dependent) • We must know the success examples (usability paradigms)
• Understand why they work (usability principles)
• Use the adequate methods (user-centered approach)
• And test, re-design, test, redesign … until we attain the usability goals
+ Adequate methods + Evaluation
Evaluation
Universidade de Aveiro Departamento de Electrónica, Telecomunicações e Informática
(Nielsen, 1993)
105
• Usability is, according to ISO 9241-11:
“the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use”
How to measure it??
And evaluate interactive system?
108
“Traditional” Evaluation Methods
Heuristic Evaluation
• Analytical (without users) Cognitive Walkthrough inspection methods
Model based methods
Review methods
Observation usability tests
• Empirical (involving users) Query
Controlled Experiments
Formal methods Prof. José Creissac Campos
110
Assignment: product evaluation – presentation: Jan 5th, 2015
• Choose any (already deployed) application, system, device, …
• Define the potential users and context of use
• Define a list of tasks to be supported
• Prepare an evaluation protocol including:
- at least one traditional method (analytical or empirical)
- a formal method
• Submit a proposal until Dec 7th, to [email protected]
111
Heuristic Evaluation (Nielsen and Molich 1990)
• A “discount usability engineering method” for quick, cheap, and easy evaluation of a UI design
• The most popular of the usability inspection methods
• It is a systematic inspection of a design for usability
• Meant to find the usability problems in the design so that they can be attended to as part of an iterative design process.
• Involves a small set of analysts judging the UI against a list of usability principles ("heuristics")
112
Usability heuristics
• It is possible to use different heuristic sets
• Nielsen proposed 10 general principles for interaction design:
Visibility of system status
Match between system and the real world
User control and freedom
Consistency and standards
Error prevention
Recognition rather than recall
Flexibility and efficiency of use
Aesthetic and minimalist design
Help users recognize, diagnose, and recover from errors
Help and documentation http://www.nngroup.com/articles/ten-usability-heuristics/
How to conduct Heuristic Evaluation
1- Each evaluator inspects the interface alone (not communicating among them)
giving for each problem found:
- a description (usability heuristic not followed)
- a severity rate
2- The results can be recorded as written reports or by having the evaluators verbalize their comments to an observer as they go through the interface
3- Findings are aggregated only after all evaluations have been completed
4- A list of problems (description/severity) is given to the development team
114
• This ensures independent and unbiased evaluations from each evaluator
• Written reports are formal records of the evaluation, but require an additional effort by the evaluators
• An observer adds to the overhead of each evaluation session, but reduces the evaluators workload and may help
• The final report should help the development team to prioritize what to fix (according to severity and cost of fix)
115
Determining the number of evaluators
• Individual evaluators can perform a heuristic evaluation on their own
• Fairly poor results are achieved when relying on single evaluators
• Averaged over six of Nielsen’s projects:
• Substantially better performance
is possible by aggregating the
evaluations from several evaluators
• More evaluators should be used when usability is critical or large payoffs can be expected
116
Example:
• Heuristic evaluation of a banking system:
– 19 evaluators
– 16 usability problems
black square - problem found
white square – not found
117
Conclusion : in general 3 to 5 evaluators seems reasonable
http://www.nngroup.com/articles/how-to-conduct-a-heuristic-evaluation/#sthash.OmTrV7Og.dpuf
HE advantages and disadvantages
• Easy and fast to use • Easy to generate fixes • Can be used is early phases
• Subjective • Requires some knowledge and experience • Does not allow a way to assess the quality of redesigns • Loosely structured; runs the risk of finding low-priority problems • Usability experts may be hard and expensive to come by • Differences between development team and evaluators may arise
118
Cognitive Walkthrough (Wharton, et al., 1992)
• Usability inspection method not involving users
• Based on the fact that users usually prefer to learn a system by using it (e.g., instead of studying a manual)
• Focused on assessing learnability (i.e., how easy it is for new users to accomplish tasks with the system)
• May produce results quickly at a low cost
• Applicable at early phases, before any coding
119
How to perform a cognitive walkthrough
1- Task analysis: sequence of steps or actions required by a user to accomplish a task, and the system responses
2- Designers and developers walkthrough as a group, asking themselves a set of questions at each step
3- Data gathering during the walkthrough: answering the questions for each subtask usability problems are detected
4- Report of potential issues
5- UI redesign to address the issues identified
120
CW Four questions:
• Will the user try to achieve the effect that the subtask has?
(Does the user understand this subtask is needed to reach the goal?)
• Will the user notice that the correct action is available?
(E.g. is the button visible?)
• Will the user understand that the wanted subtask can be achieved by the action?
(E.g. the button is visible but the user doesn’t understand the text and
will not click on it)
• Does the user get feedback?
Will the user know that they have done the right thing?
121
Common issues
• The evaluator doesn't know how to perform the task; the method involves the optimal sequence of actions
• Involves an extensive analysis and documentation and often too many potential issues are detected, resulting very time consuming
Thus:
Lighter variants of Cognitive Walkthrough were proposed to make it more applicable in S/W development companies
122
Streamlined Cognitive Walkthrough (Spencer, 2000)
- Will the user know what to do at this step?
- If the user does the right thing, will they know that they did the
right thing, and are making progress towards their goal?
123
• Only two questions:
• And a set of rules to streamlining the walkthrough and trade-off granularity for coverage
comprises the 3 first questions of CW
According to Spencer the method can be applied successfully if the usability specialist:
• takes care to prepare the team for the walkthrough,
• avoids design discussions during the walkthrough,
• explicitly neutralizes defensiveness among team members,
• streamlines the procedure by collapsing the first three questions into one question,
• and captures data selectively
124
Example: Evaluation of a desktop photocopier UI • Machine UI:
- numeric keypad,
- "Copy" button,
- push button on the back to turn on the power
The machine automatically turns itself off after 5 min inactivity
• Task: copy a single page
• User: any office worker
• Actions needed: turn on the power,
put the original on the machine,
press the "Copy" button
125
http://hcibib.org/tcuid/chap-4.html#4-1
Practice the Streamlined Cognitive Walkthrough:
• Look for a phone number at the University of Aveiro Web site
user: any student from the University
127
Look for a person’s phone number at the University of Aveiro Web site user: any student from the University
Task analysis:
- look for the icon (directório);
- input part of the person’s name and search
- get the phone number
But the defined user profile (any student from the University) includes foreign students, thus a previous action is needed:
- select the English version
For each action we need to ask the two questions and put ourselves in the shoes of the user!
128
First action: find the icon Q1 - Will the user know what to do at this step? Even reading the tooltip (directório) possibly the correct icon is not recognizable! Q2 - If the user does the right thing (selects the icon), will they know that they did the right thing, and are making progress towards their goal? Probably yes; this looks a familiar search bar and it is adequatly labeled (lista telefónica; pesquisar)
129
Second action: input part of the person’s name
Q1 - Will the user know what to do at this step? Probably yes; the tooltip lets the user know he/she should input the person’s name and select “pesquisar” Q2 - If the user does the right thing (selects the icon), will they know that they did the right thing, and are making progress towards their goal?
Probably yes; however, some users might not recogize 24117 as a phone number (it only has 5 digits, as it is internal, and not 9 as possibly expected)
130
Limitations of Analytical Methods
– Are subjective
– Involve several usability experts
– Cannot find all usability problems
Thus, empirical methods (involving users) are needed
observation
query
controlled experiments (scientific approach)
132
Usability test (engineering approach)
Ethics in applying empirical methods
Involving users implies specific cautions:
– Asking for explicit consent
– Confidentiality
– Security (avoid any risk)
– Freedom (users may give up at any time)
– Limit stress
It’s the system that is under evaluation not the user!
133
Observation
Has many variants from very simple
to very complex and expensive:
• Direct: observer takes notes
• Undirect: through audio/ vídeo – more complex and time consuming
• Think Aloud: users are asked to explain what they are doing
• Logging: users activity is logged by the system
• Combinations of the previous, etc
134
Query
• Two main variants:
– Questionnaire (reach more people; less flexible)
– Interview
• Should be carefully prepared and tested
• Collected data should be carefully analyzed
135
Controlled experiments
• The work horse of science ...
• Have a complex logistics
• Should involve a statistics consultant
• Important issues to consider:
– Hypothesis
– Variables (input or independent; output or dependent)
– Secondary variables
– Experimental design (within groups; between groups)
– Participants (number, profile)
– Statistics 136
Usability tests
• Involve observation and query
• Main aspects:
– Participants
– Tasks
– Test facilities and systems
– Experimental design
– Usability measures
– Data analysis
• Have a complex logistics
• Standard: Common Industry Format (CIF) for usability test reports
137
Participants
138
• The total number of participants to be tested
(a valid statistical analysis implies a sufficient number of subjects)
• Segmentation of user groups tested, if more than one
• Key characteristics and capabilities of user group
(user profile: age, gender, computing experience, product experience, etc.)
• How to select participants
• Differences between the participant sample and the user population
(e.g. actual users might have training whereas test subjects were untrained)
Tasks
139
• The task scenarios for testing
• Why these tasks were selected
(e.g. the most frequent tasks, the most troublesome tasks)
• The source of these tasks
(e.g. observation of users using similar products, product specifications)
• Any task data given to the participants
• Completion or performance criteria established for each task
(e.g. n. of clicks < N, time limit)
Test Facilities and equipment
• The setting and type of space in which the evaluation will be done (e.g. usability lab, cubicle office, meeting room, home office, home family room, manufacturing floor, etc.)
• Any relevant features or circumstances that can affect the results (e.g. video and audio recording equipment, one-way mirrors, or automatic data collection equipment)
• Participant’s Computing Environment (e.g. computer configuration, including model, OS version, required libraries or settings, browser name and version; relevant plug-in, etc. )
• Display and input devices characteristics
• Any questionnaires to be used
140
Experimental design
• Procedure: the logical design of the test
• Participant general instructions and task instructions
• The independent variables and control variables
• The usability measures to be used: – a) for effectiveness (completeness rate, errors, assists)
– b) for efficiency (times)
– c) for satisfaction
141
Common Industry Format (CIF) for usability test reports ISO/IEC 25062:2006
• Specifies the format for reporting the results of a summative evaluation
• The most common type of usability evaluation is formative, (i.e. designed to identify problems that can be fixed)
• A summative evaluation produces usability metrics that describe how usable a product is when used in a particular context of use
• The CIF report format and metrics are consistent with the ISO 9241-11 http://www.usabilitynet.org/prue/cif.htm
http://zing.ncsl.nist.gov/iusr/
• Top ten things to know about the CIF (a list of do and don’t) http://zing.ncsl.nist.gov/iusr/top_ten.html
142
Heuristic Evaluation
• Perform the heuristic evaluation using the 10 Nielsen’s heuristics or any other set you deem more specific to your case
• Prepare a report including:
– Brief description of the product and its target users
– the usability problems description (preferably illustrated)
– their severity rates
• Ask a colleague to perform the evaluation
• Observe, take notes and prepare a report merging results of both analysis
143
Usability test
• Choose a set of usability measures (times, errors, complete rates…)
• Prepare all the documentation:
- list of tasks and perceived difficulty
- final questionnaire
- list of tasks for the observer to take notes
• Run a pilot test
144
To the user
To the observer
Controlled experiment
• Define an hypothesis
• Define the input, output and secondary variables
• Define experimental design
• Prepare all the documentation:
- list of tasks and perceived difficulty
- final questionnaire
- list of tasks for the observer to take notes
• Run a pilot test
145
To the user
To the observer
Example: comparing two selection techniques for a Virtual Environment
- Head Selection - Laser poiter
• Null Hypothesis: both are equally usable In the used conditions
• Input variable: selection technique • Output variables: performance and satisfaction
• Participants: 16 volunteers
• Experimental design: within groups
Danilo Souza, Paulo Dias, Beatriz Sousa Santos. “Choosing a Selection Technique for a Virtual Environment”. Virtual, Augmented and Mixed Reality. Designing and Developing Virtual and Augmented Environments, Lecture Notes in Computer Science, Volume 8525, 2014, pp 215-225, http://link.springer.com/chapter/10.1007%2F978-3-319-07458-0_21
• Protocol:
– Brief explanation
– Training cycle was performed right before each selection cycle
– Selections at different distances ( 5 to 70 meters)
– 56 selections; half for each method
– Test activation by the participants at every step
– Errors and times
• Main results
HS was better a
147
148
0
2
4
6
8
10
12
14
16
5 10 15 20 25 30 35 40 45 50 55 60 65 70
Erro
rs
Distance (m)
SOC
LP
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
5 10 15 20 25 30 35 40 45 50 55 60 65 70
Ave
rage
Tim
e E
lap
sed
(s)
Distance (m)
SHO
LP
Bibliography – Books and links
- Alan Dix, Janet Finlay, Gregory Abowd, Russell Beale, Human-Computer Interaction, 3rd
edition, Prentice Hall, 2004
- Ben Shneiderman, C. Plaisant, M. Cohen, and S. Jacobs, Designing the User Interface:
Strategies for Effective Human-Computer Interaction , 5th ed., Addison-Wesley, 2009
- Jenny Preece, Yvonne Rogers, Helen Sharp, D. Benyon, S. Holland, T. Carey, Human-
Computer Interaction, Addison Wesley, 1994
- The Encyclopedia of Human-Computer Interaction, 2nd Ed,, Aarhus, Denmark: The
Interaction Design Foundation. Interaction Design Foundation, 2013 http://www.interaction-
design.org/books/hci.html
- Jackob Nielsen, Usability Engineering, Morgan Kaufmann, 1993
- Peter Mitchell, A Step-by-step Guide to Usability Testing, iUniverse, 2007
- Ian Sommerville, Software Engineering, 9th ed., Addison Wesley, 2010 (chp. 29)
http://ifs.host.cs.st-
andrews.ac.uk/Books/SE9/WebChapters/PDF/Ch_29%20Interaction_design.pdf
Bibliography – links
- Gilbert Cockton, Usability Evaluation. In: Soegaard, Mads and Dam, Rikke Friis (eds.), The
Encyclopedia of Human-Computer Interaction, 2nd Ed, 2013, Aarhus, Denmark: The
Interaction Design Foundation. (retrieved March, 10, 2013)
http://www.interaction-design.org/encyclopedia/usability_evaluation.html
- Norman/ Nielsen Group - http://www.nngroup.com/articles/
- User Focus - http://www.userfocus.co.uk/resources/iso9241/futureparts.html
- Standard ISO 9241-11 - Ergonomic requirements for office work with visual display
terminals Part 11 : Guidance on usability
- Standard ISO/IEC 25062:2006 – CIF - Common Industry format for Usability Tests Report
- What is CIF? - http://www.usabilitynet.org/prue/cif.htm
- UXPA Code of Professional Conduct
http://www.usabilityprofessionals.org/about_upa/leadership/code_of_conduct.html
152