integrity and authenticity of digital cultural heritage

Integrity and Authenticity of Digital

Cultural Heritage Objects

Thematic Issue 1 August 2002

Integrity and Authenticity of DigitalCultural Heritage Objects

Thematic Issue 1

CONTENT

5

6

9

1011141618

20

212123242425262828

30

32

34

35

Photo: Beeld en Geluid

Guntram GeserIntroduction & Overview

Seamus RossPosition Paper on integrity and authenticity of digital cultural heritage objects

Interview with Hartmut WeberAnalogue versus digital storage of our history

Michael SteamsonDigiCULT experts search for e-archive permanence‘Define Authenticity, Integrity’

Related projects and standards:A (small) selection‘Identify Methods,Technologies’‘Make Conclusions, Recommendations’

Interview with Ulrich KampffmeyerOnly co-operation can prevent us from drowning

Friso Visser / Pieter KopArchiving in the audiovisual production context – A case study on the Netherlands Institute for Sound and VisionInterview with Annemieke de JongAbout Sound and VisionServicesInternationalIntegrity and Authenticity

The digital broadcaster example: BBCLessons LearnedFuture Scenario

Literature

The Barcelona Forum participants

DigiCULT: Project information

Imprint

DigiCULT 5

INTRODUCTION AND OVERVIEWBy Guntram Geser

FUNCTION AND FOCUS


DigiCULT, as a support measure within theInformation Society TechnologiesProgramme (IST), will for a period of thir-

ty months provide a technology watch mechanismfor the cultural and scientific heritage sector. Backedby a pool of over fifty experts, the project monitors,discusses and analyses the impact of new technologydevelopments on the sector.

To promote the results and encourage early takeup of relevant technologies, DigiCULT will publishits results through a series of seven Thematic Issues,three in-depth Technology Watch Reports (with thefirst issue available in early 2003), as well as push outthe e-journal DigiCULT.Info from a growing data-base of interested persons and organisations (sub-scription at www.digicult.info), and manage the pro-ject website

The Thematic Issues focus on topics examined atthe expert round tables organised by the DigiCULTsecretariat.They present and interpret the results ofthese forums, as well as provide information and opinions in the form of interviews, case studies, shortdescriptions of related projects, and are rounded offwith a selection of relevant literature.

In contrast to the Technology Watch Reports, theThematic Issues will address more the organisational,policy, and economic aspects of the technologiesunder the spotlight. Placing special attention on waysof enhancing the adoption of new perspectives andapproaches, such as new business models, as well asfostering co-operation between cultural heritageorganisations, industry players, researchers and otherstakeholders.

Following the first Forum on ’Integrity andAuthenticity of Digital Cultural Heritage Objects‘(Barcelona, May 6th, 2002), will be the round tables:’Digital Asset Management Systems‘ (Essen,September 3rd, 2002, in the context of the AIIMConference @ DMS EXPO), ’XMLTechnologies/Resource Discovery ‘ (Darmstadt,January 2003), and ’Learning Objects‘ (Amsterdam,June 2003). For updates, please consult www.digi-cult.info.

6 DigiCULT

TOPIC AND CHALLENGE

OVERVIEW

levels.This situation reinforces the need for imme-diate action in the development and implementationof policies, guidelines, procedures and technical solu-tions. Mr Weber is convinced that archives ‘cannot sitand wait until decisions have been made’.

Michael Steemson from Caldeson Consultancy,who assists DigiCULT as a scientific consultant,describes in detail the many facets, view points, argu-ments and references of the discussion on the inte-grity and authenticity of digital objects.The sum-mary reflects the liveliness of the discussion, whereconsensus was and was not reached, and how muchremains to be done.

In particular, Ulrich Kampffmeyer, ProjectConsult, Germany, in his interview draws attentionto the need for highly effective solutions to managethe life cycle of e-records and other digital objects.Highlighting the reality that public administrations andarchives do not always receive from ICT supplierswhat they need. On the other hand, he continues,archives have also been shown wanting in definingthese needs.

Friso Fisser and his colleague Pieter Kop fromPricewaterhouseCoopers Consulting round off thisThematic Issue with a case study on the NetherlandsInstitute for Sound and Vision.The Institute servesthe Dutch broadcasters, as well as preserve nationalaudio-visual heritage.The interview with Annemiekede Jong, Head of their Information Policy SectorICT, provides insights into the tremendous changethe digitisation has brought to the Institute.Theunprecedented growth in size and complexity ofinformation calls for the implementation of anenhanced infrastructure of technologies, protocolsand metadata to ensure that information can be pre-served and reused in the future. De Jong thinks thatsuch a goal is only achievable by turning every pro-gram maker into a part time archivist. She also high-lights that one of the most important aspects here isto motivate all stakeholders to use the same standardsand protocols.

Supporting information including short descriptionsof related projects and standards, and a selection ofliterature can be found throughout this issue.

We hope that this issue will inform and stimulatefurther discussion on how to assure the integrity andauthenticity of our digital cultural heritage.

This first Thematic Issue concentrates on aquestion that is critical to all organisationsthat archive and provide access to cultural

heritage objects: How to preserve and prove theintegrity and authenticity of digital objects? Theimplementation of new technologies have presentedorganisations with unparalleled opportunities to sup-port administrative, scholarly, educational, as well ascommercial uses of ’born-digital‘ and digitisedobjects.These opportunities, however, bring withthem critical issues relating to the heterogeneity, lifecycle, and in particular integrity and authenticity ofdigital objects.

The challenge is most acute for e-archives thathave a highly structured working relationship withpublic administrations, institutions or businesses.They need to be involved strategically in the mana-gement of the life cycle of the digital objects, toimplement appropriate policies and working proce-dures necessary for the preservation and re-use ofrecords, cultural objects, research results and otherassets.

For public records archives, preserving authenticrecords in the digital environment demands the set-up of a system of controls that extends over the enti-re life cycle of records, from creation to permanentpreservation.Which for archives servicing the broad-cast industry has also become mission critical as theypartner in the production context. Both examplesfigured prominently in the expert round table heldin Barcelona on May 6th, 2002, the basis of thisThematic Issue.

Seamus Ross, director of HATII (University ofGlasgow) summarises the main issues in theuse of procedures and technologies for assuring

the integrity and authenticity of digital objects. Hecontends that the situation of creation, disseminationand preservation is ‘uncontrolled’, and calls for moreconsistency of approach, and points to research que-stions that need to be addressed.

In the interview with the journalist Joost vanKasteren, Hartmut Weber, the President of theGerman Bundesarchiv, tackles the challenges facingthe sector in the shift from paper to e-archives. Heargues that this shift has placed enormous pressureon administrations and archival organisations at all

When we work with digital objects wewant to know they are what they purport to be and that they are complete

and have not been altered or corrupted.These twoconcepts are encapsulated in the terms Authenticityand Integrity. If the authenticity and integrity of adigital object cannot be established questions arise asto its genuineness and utility. As digital objects aremore easily altered and corrupted than say paperdocuments and records, creators and preservers oftenfind it challenging to demonstrate their authenticity.As digital objects that lack authenticity and integrityhave limited value as evidence or as an informationresource. If this is the case what are the require-ments of authenticity and integrity functionality andwhat can be done to ensure that they are present indigital objects or in the systems that maintain them?

The Barcelona DigiCULT Forum began itsexamination of the organisational and technologicalissues related to authenticity and integrity withconsideration of the diverse perceptions of authenti-city held by creators, preservers and users of digitalinformation and objects. Underpinning authenticityand integrity and their preservation over time are theconcepts of fixity, stabilisation, trust, and the require-ments of custodians and users. The discussants werenot alone in finding it difficult to determine whatneeds to be protected in order to maintain authenti-city and integrity over time, as other examinations ofthe issue have reached the same conclusion. As an

authentic digital object is one whose genuinenesscan be assumed on the basis of one or more of thefollowing: mode, form, state of transmission, andmanner of preservation and custody. The questionwas raised of whether authenticity can be consideredfrom a number of different perspectives such as thoseof the different categories of user needs. The needsand requirements of different types of users vary andmay even be dependent upon types of digital objectsthey encounter and how they encounter them.Thedifferent types of digital objects, including records,online journals, databases, audio-visual materials eachappear on the surface to have their own require-ments in relation to authenticity.Although there hasbeen much research into the issue of authenticitythere remains disagreement as to whether all digitalobjects and their users could be treated in similarways or whether object-sensitive and user-contextua-lised solutions were required.

In this regard problems that required further inve-stigation included:° Could general characteristics of authenticity be

identified that would apply to all digital objects? ° Or do different types of digital objects, record

keeping procedures and digital object creation practices, alongside the variety of institutional requirements mean that digital object preservation would require a range of mechanisms for enabling user and preservers to ascertain the authenticity of material.

DigiCULT 7

POSITION PAPER ON INTEGRITY AND AUTHENTICITY OF DIGITAL CULTURAL HERITAGE OBJECTS

By Seamus Ross

Pho

to:B

eeld

en

Gel

uid

Attention to questions of authenticity and integri-ty tend to focus on whether objects have these pro-perties. More research needs to be centred on thecreation of digital objects to establish how theymight acquire these properties.A central lies in thecurrent idiosyncratic climate of records and objectcreation.The challenges that remain to be addressedinclude:° How in practice can the creators be influenced to

produce and create digital objects in ways that can guarantee and provide evidence of authenticity and integrity?

° How can vendors be encouraged to produce products that will allow creators to create records and objects with characteristics that enable authenticdigital objects? It is currently impossible to purchase a ‘preservation solution’ off-the-shelf.

Of course we might tackle this problem by agre-eing two sets of requirements for authenticity: onefor the creator and the other for the preserver.Authenticity does not vary across the processes butthe method for ensuring it, such as provenance andcustodial history. The role that appraisal plays in theprocess has received too little attention in the past.

The communities that create and preserve digitalobjects forming the backbone of this discussion werepublic administrations, broadcasters, publishers andlibraries. In all four sectors, as in most others, fewcontrols existed on the creation, preservation and dis-semination of digital objects. Each sector argued thatits requirements, technology needs and cultural envi-ronment were just different enough to require spe-cialised approaches to authenticity. On the otherhand it is evident that the best solution would beone that could be adopted across all sectors. Ofcourse, even if mechanisms could be found the reali-ties of organisational structures have an impact onapproaches to the integrity and authenticity ofobjects (see the BBC example below).The practicali-ties of making changes in a large, complex organisa-tion indicate that any major shift in thinking andpractice, whether at individual or departmental level,would depend upon internal power struggles as well

as increased resource allocations within organisations.Can there then be one solution that all types of organi-sations, regardless of size and institutional culture,could adopt that would support the creation andpreservation of authentic digital objects? In practicalterms it is essential that such a solution be developed.

It is evident from reading the report of theBarcelona Forum that more consistency of approachacross the heritage sector would be essential:° to develop mechanisms to ensure integrity and

authenticity of digital objects.° to improve communication among heritage

organisations about the challenges and approaches to integrity and authenticity.This is crucial becausethere appears to be a lack of motivation to understand other perspectives and approaches.

° to conduct more case studies.They would provide an essential window to improve our understandingof why digital objects are created and how they aremanaged.

° to investigate the role trust plays in authenticity and integrity of digital objects. In many instances users and preservers establish authenticity on the grounds of trust in the organisation involved or technology used in the preservation of the digital object. The current understanding of the major factors that drive trust decisions in the digital world, as well as the risks involved with having andimplementing this sort of trust is limited.

° that in view of the number of independent projectsconducting work in the area of digital preservationand especially in the area of integrity and authen-ticity of digital objects enhanced collaboration between these projects has become essential.A good starting point would be a survey of current research to identify complementarily, overlaps, and gaps in research.

° that emerging guidelines on practices to support the authenticity and integrity of digital objects need to be framed in ways that make them accessibleand usable by a variety of communities and encourage suppliers to provide adequate levels of functionality in their products (see Ulrich Kampffmeyer interview page 20).

8 DigiCULT

Photos: Beeld en Geluid

DigiCULT 9

Afew years ago, FrancisFukuyama declared, thatwe had reached the end

of history in Hegelian terms; thatis, history as a confrontation bet-ween ideologies. Prof. Dr. Hartmut Weber, presidentof the Bundesarchiv, the German federal archives, ismore afraid that we might loose our history. If weare not careful in archiving electronic records.Amajor problem in this respect, is the documentarytrail.Weber states: ‘A good archive contains not onlythe end document, for instance the text of an agree-ment, but also documents which allow you to recon-struct the process which led to the agreement.Authenticity demands that this archiving process betransparent, in the sense, that you are able to identifyand verify all the elements of the process in reachinga decision.’

The reconstruction of this documentary trail isstill mostly based on paper documents, such as lettersor notes that are kept in archives, but as the migrationof this information from paper to electronic formincreases, so does its volatility. ‘That, in itself, isnothing new’, comments Weber. ‘Previously, you hadinformal meetings of which no minutes were taken.The same goes for telephone conversations.The diff-erence is that the results of informal meetings andtelephone conversations were formalised somewherein written form. Nowadays they tend to remain inthe virtual domain without a physical representation.’

The problem is that the media containing theinformation are not very ‘tenable’.To be able toaccess the information, the records have to be con-verted ever so often, because either the hardware it-self or the software used for access changes.Throughthese repeated migrations the intrinsic value of therecords might get lost, including the context of therecord.According to Weber, that is a temporary pro-blem. ‘At the moment, the Bundesarchiv is executingsome projects with the well known method of datamigration. But, there is a need to think about emula-tion of systems, a process whereby not only thedata are transferred from one medium to theother, but also their context and functionality. Inthis way, we hope to be able to guarantee theiridentity and integrity.’ Until that day, a lot of digitaldocuments will have to be converted to their analogue – paper – form.

More importantthan the technologyitself is the lack ofcomprehensive pro-cedures for archi-

ving electronic documents. For instance, letters sentby email or text adjustments in electronic docu-ments.The Bundesarchiv is quickly developing suchprocedures, for instance in archiving emails, becauseloosing history is not only a loss for historians whowant to explore the archives, but also for the govern-ment, the organisation itself and the wider commu-nity.Weber: ‘For its present policies, a governmentoften has to fall back on archived records. If they do,they have to be sure that the information on recordis correct and complete in the sense I mentionedbefore, that is that you can reconstruct the process.The same goes for citizens who have a right toknow under our constitution.That means, that younot only have to guarantee the authenticity of docu-ments, but you also have to stabilise the context.’

Hence, it is, according to Weber, of the utmostimportance that government organisations developprocedures for archiving electronic documents toprevent a possible loss of information. ‘The use ofnew technologies puts a lot of pressure not only onthe national government, but also on regional andlocal governments and of course on the EU. But,before you can start developing procedures, you haveto raise the awareness among government officialsand civil servants about the importance of completeand transparent archives.As professional archivists, wecannot sit and wait until decisions have been made.I think we have to make it a co-operative effort tostore records in a transparent way, making themaccessible by electronic means.’

Weber foresees an archive which stores most ofthe data in electronic form. Decisive documents,though, will – also – be stored on paper.The task athand is to design an archiving system that combinesthe advantages of a traditional paper archive (tenabi-lity, identity and integrity) with those of an electro-nic archive (accessibility, ease of use). ‘It will takesome time, possibly some decades, before a stablesymbiosis has been realised’, reinforces Weber. ‘But,eventually, it will happen. I am an optimist.Archiveshave survived 1000 or more years, so I trust we canuse digitalisation to our advantage as well.’

DIGITALHISTORY

ANALOGUE VERSUSSTORAGE OF OUR

An Interview with

Hartmut Weber, President

Bundesarchiv, Germany

by Joost van Kasteren

Photo: Bundesarchiv

What's at Risk?‘Virtually every private and

public organization which usesinformation technology to facilitateits recordkeeping functions hasexperienced the undesirable effectsof adopting new technologies without forecasting and planningfor the consequences of the proprietary nature of software applications, media and digitalobsolescence, and hybridpaper/digital environments.’

InterPARES 1,http://www.interpares.org/background.htm

10 DigiCULT

DIGICULT EXPERTS SEARCH


SUMMARY OF THE FORUM IN BARCELONA, MAY 6, 2002

By Michael Steemson

What size is the risk inherent in the newtechnologies for cultural heritage preser-vation? The institutions agree it’s huge.

Technologies come and go with such shattering rapi-dity.What can be done about it? That’s the task thenine experts – archivists, librarians, technologists andacademics – were set at the first DigiCULT Forum,held in May 2002.

They met in round table discussion in sunnyBarcelona, Spain, to debate the Integrity andAuthenticity of Digital Objects.They differed over whatthese terms actually meant but identified this confu-sion as one of the problems: Many different kinds ofdigital objects, countless usages and values, and innu-merable users each bringing their own evaluations ofthe objects, whether they be single documents, booksor video recordings.

There was, as yet, little adequate technology to dothe job, they decided. Solutions lay in the hands ofthe object creators and preservers, who were some-times one and the same but who needed criteria towork to. But, what should those criteria be? Theymust find out, the Forum experts realised.

But they did discover a future for recordkeepers.One of the Nine told the group: ‘We don’t call them archivists any more. If we called them archivists,nobody would let them near the place.’

(Nationaal Archief), told the experts: ‘What the agendadoesn’t say is who are really involved in dealing withauthenticity and what do we understand authenticityis.We have people from different backgrounds andthey might have different perceptions of what authenticity is.’

Years ago, that would not have been difficult forSir Hilary Jenkinson, the grand old man of Britishrecordkeeping who took from a dusty academic shelfthe Archivists’Art and returned it a Science. In his1960’s Manual of Archive Administration, he definedauthentic archives as those ‘preserved in officialcustody . . . and free from suspicion of having beentampered with’.1

But this was about the time the Father of theInternet, New Yorker Leonard Kleinrock, was at theMassachusetts Institute of Technology preparing hisPhD thesis, Information Flow in Large CommunicationNets,2 destined to be the first little twinkling light atthe far end of the line of communication that trans-formed into the roaring information superhighwayof the 21st Century.

Now, in the new millennium, the problem is morecomplex. One of the Forum experts, LucianaDuranti, professor at the School of Library,Archivaland Information Studies at the University of BritishColumbia in Vancouver, summed up the problem tothe Forum e-journal, DigiCULT.Info: ‘The fast pacewith which technology for creating and recordinginformation is developing threatens the authenticityof records.Archivists, governments and other institu-tions that rely on these records are losing control.I would not hesitate to call the situation disastrous.’3

As the Barcelona discussion opened, the Forum’ssearch for ‘authenticity and integrity’ was questionedby another Netherlands expert,Annemieke de Jong,from the Netherlands Audiovisual Archive(Nederlands Audiovisueel Archief). She argued that thetheme implied that Forum members already knew‘what authenticity is in the digital domain’. Sheasked: ‘Can these traditional concepts of authenticityand integrity still be applied on digital objects in thefirst place? Can we still think of authenticity in thedomain of digital objects?’

Archivschule Marburg archives science lecturer,Nils Brübach, approached it from another angle. Hepreferred to see the concepts as functional ratherthan technical. He questioned current opinion thatsaw authenticity and integrity as absolutes and he

DigiCULT 11

FOR E-ARCHIVE PERMANENCE

Provenance Corrupt or Not‘Authenticity in recorded information connotesprecise, yet disparate, things in different contextsand communities. It can mean being original butalso being faithful to an original; it can meanuncorrupted but also of clear and known provenance,“corrupt” or not.’

Abby Smith, Authenticity in a Digital Environment,Council on Library and Information Resources(CLIR).Washington, D.C., May 2000. http://www.clir.org/pubs/reports/pub92/contents.html

‘DEFINE AUTHENTICITY, INTEGRITY’

The agenda called for investigation of thequestion ‘How to implement methods forassuring authenticity and integrity in the

long-term’. Forum Moderator, Hans Hofman, archi-vist with the National Archives of the Netherlands

1 Jenkinson, Hilary:

A Manual of Archive

Administration. London:

Percy Lund, Humphries &

Co. Ltd., 1965.2 Kleinrock, Leonard:

Information Flow in

Large Communication

Nets. Cambridge:

Massachusetts Institute of

Technology, 1961.

http://www.lk.cs.

ucla.edu/LK/Bib/

REPORT/PhD/.3 DigiCULT.Info,

Issue 1, July 2002.

http://www.digicult.info/

pages/newsletter.html

proposed that, as functional concepts, they applied tothree levels, the content, context and structure ofdigital objects.

He said: ‘The question is about interrelationshipand … how to guarantee authenticity and integrity atall three levels.The question is also do we need toestablish a concept of authenticity and integrity on allthree levels?’ He added: ‘I would say the first thinkingshould be on fixity and stabilising digital objects andthen we could think about integrity and authenticity.’

very detailed methods of control in place all along,so that you can say that you have a trusted system, itdoesn’t work.’

InterPARES had decided on two levels of require-ments, she said. One was a need to presume the authenticity of the records based on how they weregenerated and maintained.The other was the require-ment to create authentic copies of the records topreserve them over time.

Paul Fiander, Head of Information & Archives,BBC, had an example.The Corporation had almostone and a half million commercial recordings ran-ging in medium from wax cylinders to CDs.Theproblem lay with the 78 r.p.m. and LP vinyl recordsthat were too fragile and low quality to issue for use– ‘too many clicks, so we clean them up’.They werecopied onto CDs and, as a result, were no longerauthentic versions.

University of Antwerp (Bibliotheek UniversiteitAntwerpen) librarian, Julien van Borm, didn’t particu-larly mind that process, so long as he knew what hadbeen done, particularly if the original no longer exi-sted. He considered that: ‘In the future, I think weneed not only the document in itself but also thehistory, lets call it a C.V. of the document that has tobe documented in the document itself.’

Dr. Brübach agreed completely.An archival objecthad to include both the digital object and its proces-sing history … ‘what has been changed and maybewhat has been lost and both together this could givethe user a hint of authenticity, not authenticity itself ’.

And he went on: ‘When we turn to born-digitalobjects, does an original really exist in the digitalworld? Do we have an original there which can beidentified independently as an original? I would sayno.Any original in the digital world can be definedas an original by somebody using some means,maybe metadata, which would be the instrument tosolve the problems you have just outlined.’

Talk around the Forum table began to turntowards the responsibility of creators to contributetowards digital authenticity and integrity. HansHofman developed a diagram showing the ‘digitalobject’ hemmed in by three entities, the creator, thepreserver and the user.

He explained: ‘What we are talking about is digitalobjects in different domains created by different creators and used by all kinds of different users.Whatauthenticity means is, in my perception, what thecreator has an intention to convey to a user with theobject. So, we are talking about the relationship be-tween the creator, the digital objects he his creatingand the user. But the user has to know what theintention was of the creator.’

12 DigiCULT

Professor Duranti, who is also director of theInternational Research on Permanent AuthenticRecords in Electronic Systems, the InterPARES pro-ject, had doubts about ‘fixity’. She said that theInterPARES project had, at first, presumed fixity tobe an essential element of authenticity. She went on:‘But the reason for the InterPARES project 2 is thatwe are discovering that by stabilising records that, bytheir nature, are dynamic we, in fact, end up forgingthem.That is, we are eliminating their authenticity.’

She continued: ‘… should we have questions thatapply to all digital objects or shouldn’t we really haveseparate questions for different kinds of digital objects?Because, certainly, authenticity is not the same thingto music that it is to a legal record and I think thatthe primary concern should be actually separationnot unification.We should set out by thinking oftypes of digital objects separately, different characteri-stics, different solutions and different concepts.’

After further lengthy discussion on varying require-ments for the integrity of different digital objects,Hans Hofman suggested that from users’ perspectivesthe question was simply one of trust. ProfessorDuranti agreed but warned against archives’ past faithin creators. She said: ‘This is no longer true.The person who generates the material may trust it andmight be wrong. Because, with digital records, thefluidity of the record is such that if you don’t have

Fixed and Fluid‘If the utility of both the fixed and the fluid is recogni-

zed, the Web may develop much of its innovative powerfrom the possibility of producing documents that combineboth fixity and fluidity.Already, many documents retain aconstant text while their links are continually changed. …This interplay between fixity and fluidity, formerly possibleonly on the scale of collections, may now become a centralfeature of individual documents.’

John Seely Brown and Paul Duguid, The Social Life ofDocuments, Palo Alto, CA., Palo Alto Research Center(PARC), May 1996. http://www2.parc.com/ops/members/brown/papers/sociallife.html

‘The preserver has to take care that this digitalobject is carried through time to new users, currentusers. In different contexts it might be different per-ceptions of what authenticity is, and what we wantto achieve is a certain trust in the user that this is thedigital object that was once created with this identi-ty, because the identity is dependent on the creator.’

The Austrian National Library’s Max Kaiser, poin-ted out three different types of digital object: ‘Whenwe receive the object we have to decide what part ofit is to be preserved.Then we have to submit it toour archives and begin recording the changes thathave to be made to it – migration and other things.We have to record rights issues and then, based onthis complicated information package, we have todecide how we can disseminate it to our users.’

Professor Duranti put it: ‘So, in fact, the grave re-sponsibility for future preservation is with the creator.The creator has to make sure the object is identifiable;that it has metadata to ensure its integrity can beproved; that it can be seen, with access privilege; thatsort of thing.We therefore should also look atmethods that should be used by the creator to gene-rate the objects properly for future preservation.’

She said that the InterPARES project had createdtwo different sets of requirements for authenticity,one for the creator and another for the preserver.

Other members thought perhaps three sets ofrequirements were needed: ‘ingest’, ‘preservation’ and‘dissemination’ was Nils Brübach’s model.The groupdiscussed the role of the preserver as mediator between the creator and user.The preserver’s taskwas to create records, too, it was suggested, recordssuch as protocols describing what had been lost oradded in the preserving process, giving authenticityto what was left. Once these had been satisfied andan audit trail tracking process was in place then authenticity could be presumed.

In his summing up, Hans Hofman said the grouphad agreed that authenticity was not a static thingbut had to be approached from the contextual pointof view. He added: ‘We did not come up with a lotof criteria, but at least one is that an authentic objectis what it purports to be, and there are different play-ers: the creators, the preservers and the users, and allhave their own views that influence the way we dealwith authenticity.’

DigiCULT 13


14 DigiCULT

RELATED PROJECTS ANDA (SMALL) SELECTION

InterPARES - International Research on Perma-nent Authentic Records in Electronic Systems

InterPARES 1:The first phase of the project began in 1999 and

was concluded in 2001. It built on an earlier projectat the Vancouver, Canada, University of BritishColumbia (UBC), ‘The Preservation of the Integrityof Electronic Records’ (1996), which addressed issuessurrounding the creation and maintenance of authen-tic and reliable electronic records in their active, pre-archival state. See: http://www.interpares.org/UBCProject/index.htm

InterPARES 1 focused on the preservation of theauthenticity of records that are no longer needed bythe creating body to fulfill its own mission or purpo-ses. Results from InterPARES 1 included guidanceon conceptual requirements for authenticity, modelsof the processes of selection and preservation of authen-tic electronic records, a glossary, and several otherdocuments are available on the project website.

The final text of the InterPARES 1 findings willbe electronically published on its website by Septem-ber 2002 and published in book form by the ItalianMinistry for Cultural Properties and Activities in thewinter 2002-2003.

InterPARES 2:The key aspects distinguishing the second phase of

the InterPARES project from the first are describedon the project website as ‘dramatically innovative’.InterPARES 2 will not only address issues of authen-ticity but also reliability and accuracy. It will studythem throughout the records' life-cycle from creationto permanent preservation, unlike phase 1 that wasconcerned only with non-current records destinedfor permanent preservation. Importantly, InterPARES 2will focus on records produced in new digital environ-ments, experiential, dynamic, and interactive whereasphase 1 was concerned only with records generatedin databases and document management systems.Thefocus of InterPARES 2 will not be just on recordsresulting from administrative and legal activities, butthose resulting from artistic, scientific and govern-ment activities.

The InterPARES studies began in January 2002and will continue until December 2006.

http://www.interpares.org/

Reference Model for an Open ArchivalInformation System (OAIS)

The Reference Model for an Open ArchivalInformation System provides a common frameworkfor describing and comparing architectures and ope-rations of digital archives. It was developed by theU.S. Consultative Committee on Space Data, and hasbeen adopted as ISO 14721:2002.

A useful short presentation and review of the refe-rence model is provided in the Research LibrariesGroup and Online Computer Library Center(RLG/OCLC) report ‘Trusted Digital Repositories’(2002) highlighting the necessary functions of along-term digital repository.The RLG proposescompliance with this model as the defining attributeof a trusted digital repository.

The OAIS reference model was used, for example,by the Leeds University, England, Consortium ofUniversity Libraries’(CURL) Exemplars in DigitalArchives (Cedars) project (http://www.leeds.ac.uk/cedars/), whose participants discovered ‘the benefit ofadopting a shared vocabulary and set of concepts toallow implementation across a number of differentlocal situations’. In fact, Cedars has provided oneimportant demonstrator project based on this model.(cf. Russell, 2000; see: Literature)

Another major project using the OAIS referencemodel is the new digital library system of the BritishLibrary that is designed to provide long-term accessto digital collections.

Sources:NASA: ISO Archiving Standards: Overview (this site

contains links and background information to the refe-rence model and other archival standardization efforts)http://ssdoo.gsfc.nasa.gov/nost/isoas/overview.html

Blue Book of the Reference Model: CCSDS650.0-B-1: Reference Model for an Open ArchivalInformation System (OAIS). Blue Book. Issue 1.January 2002.http://ssdoo.gsfc.nasa.gov/nost/isoas/ref_model.html

Research Libraries Group: Open ArchivalInformation System (OAIS) Resources,http://www.rlg.org/longterm/oais.html

Helen Shenton: From Talking to Doing: Digital Pre-servation at the British Library. In: Preservation 2000,http://www.rlg.org/events/pres-2000/shenton.html

Photos: Beeld en GeluidAustralia and Victoria State MetadataStandards: the VERS Metadata Scheme

National Archives of Australia sees the VERS(Victoria Electronic Recordkeeping System) scheme asa reference tool for government agency, corporatemanagers, IT personnel and software vendors invol-ved in the design, selection and implementation ofelectronic recordkeeping and related informationmanagement systems. But, for Victoria’s State PublicRecord Office (PROVic), the VERS designer, its pur-pose is to represent information required for preser-ving records over a long period.

The national system (http://www.naa.gov.au/recordkeeping/control/rkms/summary.htm) defines abasic set of twenty metadata elements (eight ofwhich constitute a core set of mandatory metadata)and sixty-five sub-elements that may be incorporatedwithin such systems, and explains how they shouldbe applied within the Australian sphere.

The PROVic method differs (see: http://www.prov.vic.gov.au/vers/standards/pros9907/99-7-2s2.htm).

In its Standard for Electronic Records Management(PROS 99/007) the Office says: ‘The VERS approachis to fix records at (or close to) the time of creationusing digital signatures.Although the VERS approachhas many advantages over migration, it has one signi-ficant disadvantage; metadata that changes or accretes(e.g. use histories) over time is not well supported.Although it is possible to “layer” metadata to supportchanging or accreting metadata, this is not efficient forelements that are continually modified.’

The Victorian State Government earlier this yearauthorised expenditure of more than Au$8 million(c. 4.5 million EUR) to begin a VERS implementa-tion programme across State agencies that will even-tually cost Au$50 million.

DigiCULT 15

David among the Digital GoliathsThe Belgian Foundation for Scientific Research

project, Digital Archiving in Flemish Institutions andAdministrations (Digitale Archivering in VlaamseInstellingen en Diensten – DAVID), aims to producea manual of guidelines for archiving digital recordsby the end of 2003, supported by Antwerp CityArchives and the Leuven University’s InterdisciplinaryCentre for Law.The group has created a simple,stable e-mail archiving protocol with set data fieldsthat must be completed by originators and recipientsbefore archiving as microfilm or imaged hard copy.

http://www.dma.be/david/eng/index.htm

STANDARDS

Project Prism - Preservation, Reliability,Interoperability, Security and Metadata

Project Prism at Cornell University, (U.S.A.), is aninterdisciplinary research project started in 1999 andfunded by the U.S. National Science Foundation.A collaborative effort between Cornell‘s UniversityLibrary and its Computer Science Department, itinvestigates and develops policies and mechanisms toensure information integrity in digital libraries.

Prism focuses on five key areas, Preservation, Reli-ability, Interoperability, Security and Metadata, in thecontext of component-based digital library architec-ture with special attention to distributed collectionsand web content.The preservation component of theresearch is examining longevity issues for webresources using risk management methods.

Sources:http://www.prism.cornell.edu/http://www.library.cornell.edu/iris/research/prism/Kenney,Anne R, et al.: Preservation Risk Manage-

ment for Web Resources. In: D-Lib Magazine, January2002, http://www.dlib.org/dlib/january02/kenney/01kenney.html

‘The volume of information is growing at an unpreceden-ted pace.We already produce more information per year thanwe did in the whole period since we descended from thetrees.A lot of information is digital only and an XML docu-ment, for instance, is created while you view it. So, how doyou keep it?’

Ulrich Kampffmeyer, President of Project Consult, Germany

roles of both creators and users in the production oftheir work. He categorised the process as ‘work inprogress’.

He said: ‘The programme maker adds input, thejournalist puts stuff in and they take it back outagain. I call that, because I come from an industrialbackground, work in progress. Stuff also goes outthere to what we call clip sales so people are genera-ting money out of this work in progress.

16 DigiCULT

‘IDENTIFY METHODS, TECHNOLOGIES’

Forum Moderator, Hans Hofman, set the direc-tion in the experts’ search for technologies andmethods to ensure authenticity. ‘Are the cur-

rent information technologies able to achieve it? Ifnot, how should it be done and is there somethingthat should be done with standardisation?’ he asked.

The BBC’s Paul Fiander dropped in yet anotherpressing concern … costs. He detailed the broadca-sting Corporation’s holdings of radio and televisionmaterial, a collection growing exponentially as inter-active television comes on stream. MPEG compressioncompromised digital authenticity, he said, andresource constraint was ‘forcing us to change our selection and retention policy’.

So, another word entered the Forum debate ...appraisal. Director of U.S. National Archives andRecords Administration (NARA) Electronic RecordsArchives (ERA) Program, Kenneth Thibodeau, said itwas still the top criteria. Harking back to their earlierdiscussion, he reminded the experts: ‘Once you saythat authenticity is contextual you cannot validlypose the question “can technology save everything inan authentic way?”’

Luciana Duranti believed the problem really laywith creators who continued to ‘generate (records) inan inappropriate way’. She complained: ‘Each recordgenerator creates records in an idiosyncratic way notrespecting the many of rules.They make things verydifficult for the preserver.’ She wondered whetherthe creator ‘is doing it because he is not interested inpermanent preservation or because he doesn’t knowwhat he is supposed to do’.

The BBC’s Paul Fiander demonstrated how theCorporation solved the problem. In so doing, gave ahappy glimpse of the future for recordkeepers … nota place in the retirement sun but in the white heat ofthe technologies mediating between creators andusers, just where the Forum thought they should be.

Paul Fiander drew a diagram of the BBC’s produc-tion process – programme managers and journalistsmaking and using archivable material, fulfilling the

It goes many times because of the number ofchannels we have.We have people taking materialout to recreate new stuff and they put it back in.Finally, it goes out to playout.Where is the role ofthe archivist in this?’

All over the place?‘Exactly!’ said the Man from the BBC. Stabbing

the diagram’s Media Asset Manager ‘cloud’, heemphasised: ‘The role of the archivist is there, exceptwe don’t call them archivists anymore, because if youcalled them archivists nobody would let them nearthe place.We call them media managers.The skill ofthe archivist is to work in this cloud over here,because if they didn’t do their job properly here, wewould never find the material again.’

An incredulous delegate asked: ‘Is this your day today reality or the future?’

Mr Fiander was firm. ‘That is what we are doingtoday, we are putting people with library qualificationsinto that cloud there, calling them media asset mana-gers. I am talking about the archivist as the creator,being involved in part of the creation process.Thatis where the argument about changing – makingthe creator do something different – is being takencare of.’

Paul Fiander’s ‘Work in

Progress’ workflow model

Programme Makers

Journalists

WORK IN PROGRESS

Media Asset Manager

Clip Sales € Y Digital play out

Hans Hofman wondered how the archivists werecoping with this change of culture.

Paul Fiander said they had to be able to work withjournalists and programme makers.They did notsimply write new rules for the creators, though. ‘Youhave to work with them.You have to be immersedwith them if you want them to change.’

Netherlands’ archivist Annemieke de Jong saw howarchives could aid the integration of asset manage-ment systems in an organisation. ‘In the BBC model,you could see that programme makers and journalistsuse the same procedures, structures and metadata forcopyright that is being developed in the archivingworld. So, it is not just the system.You should makethe rules in the archive and then distribute them tothe creators and producers.’

This would function working directly with thecreators, she said, but she wanted to know: ‘If youwant to preserve material that is being producedoutside of your organisational model how do youmaintain this form of control?’

The consensus was that market pressure could havea positive effect although systems vendors were often

Persistent Object Preservation ‘Question: How are the government’s electronic records going to be

preserved over multiple generations of technology so that future archivistsand historians can access them? Answer: Nobody knows yet.

But Kenneth Thibodeau, director of the National Archives andRecords Administration’s Electronic Records Archives (ERA) programin College Park, Md. thinks he’s on the trail of a solution. It’s calledpersistent object preservation’.

‘State it simplyIn the persistent object method, the structure of a record and of

aggregates of records is described in plain language—simple tags andschemas—so that any future technologies, and people, will recognizethe essential properties of the record and be able to access it, he said.

That gives managers the ability to change hardware and software overtime with no significant impact on the records that are being managedand preserved.

“What San Diego is telling us is that records in this format should begood for 300 to 400 years,”Thibodeau said.’

From: ‚For the record, NARA techie aims to preserve‘, by RichardW.Walker. In: GCN magazine, July 30, 2001;http://www.gcn.com/vol20_no21/news/4752-1.html.

For a description of NARA's Electronic Records Archives Programsee Thibodeau, Kenneth (2001): Building the Archives of the Future:Advances in Preserving Electronic Records at the National Archivesand Records Administration. In: D-Lib Magazine, No. 2,Vol. 7,February 2001, http://www.dlib.org/dlib/february01/thibodeau/02thibodeau.html.

DigiCULT 17

hard to influence.The Forum discussed technologicalsolutions, migration, emulation and ‘persistent objectpreservation’4, a process being researched by DrThibodeau’s NARA project at University ofCalifornia’ San Diego Supercomputing Centre.

Delegates were dubious about system emulation(‘It remains an ethical question,’ said ProfessorDuranti) and migration (‘You are entirely controlledby the software industry,’ said Dr Thibodeau).

‘So there is no technology actually that can reallydeal with what we require, is that the conclusion atthis moment?’ asked Moderator Hofman. ‘Shouldn’twe then move in earlier in the creation process inorder to influence the way things are created forexample open source, creating standards, etc.’Thatwas the way the BBC had done it and not just forthe archives.

Dr Brübach was all for telling the software indus-try ‘Hey, folks. Build in an interface which we canuse to export stuff to one of our archiving formats.Make the process as easy as possible with metadatacollected clandestinely so the user does not evenknow what is happening in the background’. Otherswere doubtful if the industry would comply butagreed that archivists could achieve some successclandestinely.

Summing up, Hans Hofman told the Forum:‘The aspects we have been discussing are mainly therequirements.We have also the technological issuesbecause technology is the reason why we are nowsuddenly facing all these issues around authenticity inpreserving digital objects.There are cultural aspectsas well. How do we convince people that they haveto have a different attitude towards what they arecreating and preserving?’

‘But what I also hear is that there are differentcommunities and may be different perceptions inauthenticity although there still might be a moregeneric idea of authenticity.This may lead to differentsolutions because the requirements are not always thesame.’

4San Diego Supercomputer

Center: Collection Based

Persistent Archives,

http://www.sdsc.edu/

NARA/Publications/

collections.html

The group requested a survey by the DigiCULTSecretariat of research projects and case studies over thelast three years.Annemieke de Jong suggested the surveyshould concentrate on subjects she had looked for in thepast and rarely found, things like interactivity, multi-

Address the Challenges Cooperatively‘The importance of maintaining the viability and accessi-

bility of digital objects over the long term underscores theneed to develop infrastructure in support of these objectives.Given the many shared challenges associated with digitalpreservation, preservation metadata among them, there istremendous scope to address these challenges co-operatively… to advance the imperative of preserving digital objectsover the long term.’

Preservation Metadata and the OAIS Information Model:AMetadata Framework to Support the Preservation of Digital Objects.

OCLC/RLG Working Group on Preservation Metadata,June 2002.http://www.oclc.org/research/pmwg/pm_framework.pdf

Short Time Horizons on Authenticity‘A great deal of technology and infrastructure now being deployed

will be useful in managing integrity and authenticity over time.However, these developments are being driven by commercial require-ments with short time horizons in areas such as authentication, electro-nic commerce, electronic contracting, and management and control ofdigital intellectual property.’

Clifford A. Lynch, Authenticity and Integrity in the Digital Environment:An Exploratory Analysis of the Central Role of Trust. Washington, DC:Council on Library and Information Resources, 2000.http://www.clir.org/pubs/reports/pub92/lynch.html

18 DigiCULT

‘MAKE CONCLUSIONS, RECOMMENDATIONS’

The DigiCULT Forum discussions precededthe OCLC/RLG Working Group by amonth, but the two gatherings reached the

same conclusion … co-operation is needed betweenstudy groups. OCLC/RLG published their report inJune5 acknowledging the ‘tremendous scope’ toaddressing co-operatively challenges of digital preser-vation, especially development of metadata protocols.

The DigiCULT experts discussed the RLG’s earlier report on Trusted Digital Repositories6, theAustralian Victorian Electronic RecordkeepingSystem (VERS)7, InterPARES8, and the projectPrism at Cornell University.9

The group wanted a survey of these and otherreports and initiatives with which the Forum couldcollaborate. Members considered they should be bet-ter informed on existing study infrastructures like theEuropean Commission-backed digital libraries net-work DELOS10, the ERPANET project11 andMoReq specifications12 , and the German stateNordrhein-Westphalia's VERA Project13, anInternet-based archives administration project.

Group members asked also for the inclusion ofprojects implementing long-term access to digitalcollections based on the U.S. Reference Model for anOpen Archival Information System (OAIS) that has beenadopted as an international standard ISO 14721:2002.

Hans Hofman suggested that such a collaborativeeffort should ‘make available knowledge of what ishappening in all these institutions’. He said: ‘Every-body who is doing something in the area of digitalpreservation and authenticity should be involved andshould be connected to that network to help identifyresearch issues and solutions, as we have today.We haveshown how difficult it is to identify those issues.’

He said the Forum had identified a need for greaterarchival influence at the moment of the creation of

digital objects. Now the group had to identify whereit would work.

Luciana Duranti agreed it would work in publicenvironments. ‘People I have spoken to in the libraryfield who are publishers etc, do want guidelines onhow to do the things right.You cannot impose themon them, but they would be very glad if you gavethem criteria because they do have a problem at thesource, at the creation of the digital works.’

Other delegates thought some of this guidance wasalready available from standards, notably the InternationalStandards Organization’s Records Management work,ISO15489.

5Preservation Metadata and

the OAIS Information Model:

A Metadata Framework to

Support the Preservation of

Digital Objects. OCLC/RLG

Working Group on Preservation

Metadata, June 2002

http://www.oclc.org/research/

pmwg/pm_framework.pdf.


media, audio-visual objects and ‘complete digitalresources’. Case studies should be selected accordingto the types of digital objects the archives preserved ‘sothat the object is the criteria, not the institution itself ’.

Dr Thibodeau described a ‘very strongly argued’paper by Coalition for Networked Information (CNI)Executive Director Clifford A. Lynch14 asserting thatdigital authenticity depended on trust.TheDigiCULT survey should include ‘two aspects of thetrust issue’. He identified these as:‘What are the majorfactors that drive trust decisions, a willingness toplace faith in person or institution or solution, thatapply in the digital world.And secondly what are therisks entailed with that, those acts of faith.’

It all added up to a lot of work for Forum Secretary,John Pereira, and his Salzburg Research colleaguesbefore the second DigiCULT Forum in Essen, Germany,in September 2002. But at the end of the long, longday, he was still smiling and continuing to ask delega-tes: ‘Is this something we should add to the list?’

Moderator Hans Hofman, reviewing the forum

DigiCULT 19

6Trusted Digital Repositories:

Attributes and Responsibilities.

Mountain View, CA: RLG and

OCLC, May 2002.

http://www.rlg.org/longterm/

repositories.pdf.7Victorian Electronic

Recordkeeping System.

Melbourne: Public Record

Office,Victoria,Australia, 1999.

http://www.prov.vic.gov.au/

vers/standards/pros9907/

99-7-2s2.htm.8InterPARES: International

Research on Permanent

Authentic Records in

Electronic Systems,

http://www.interpares.org.9Prism, http://www.library.

cornell.edu/iris/

research/prism/10DELOS Network of

Excellence on Digital

Libraries, http://delos-

noe.iei.pi.cnr.it/.11ERPANET, European

Resource Preservation and

Access Network,

http://www.erpanet.org/.12MoReq, Model

Requirements for the manage-

ment of electronic documents

and records management

systems,

http://www.cornwell.co.uk/

moreq.pdf.13Frank M. Bischoff:

Staatsarchiv Münster, Das

Projekt VERA in Nordrhein-

Westfalen: Nutzung der

Internettechnologie für die

Erschließung und archivüber-

greifende Verwaltung der

Bestände. Munster, NRW.

http://www.uni-marburg.de/

archivschule/bischoff.pdf14Clifford A. Lynch:

Authenticity and Integrity in

the Digital Environment:An

Exploratory Analysis of the

Central Role of Trust.

Washington, DC: Council on

Library and Information

Resources, 2000.

http://www.clir.org/pubs/

reports/pub92/lynch.html.

debate, commented afterwards:‘The discussion showedthat the notion of authenticity is still a difficult subjectthat is being interpreted differently by people withdifferent backgrounds and different perceptions.There was some agreement, however, that the creator,the preserver and the user each play important roles inidentifying and maintaining the authenticity of digitalobjects be they records, publications or audiovisualmaterial. It was also clear that ensuring authenticitystarts at the creation of the digital object itself.’

He thought that the concept of authenticity was stillseen as confusing. He said: ‘It is difficult to get holdof, let alone to approach or deal with it.The Forumasked for a survey of existing initiatives that try to dealwith the preservation of digital objects to discoverhow they approach the notion of authenticity. It wasemphasised that closer collaboration between differentdisciplines or communities is necessary, with a moreprominent role for the archival community and itsperception of authenticity. So the final word has notyet been spoken on this issue.’

20 DigiCULT

Undercurrents at some stretches of theAtlantic coast are so strong that peopleonly venture into the water while holding

hands with each other.That image springs to mindwhen you hear Dr Ulrich Kampffmeyer talk aboutthe digital flood that threatens to drown us. ‘Unless’,he says, ‘the ICT industry and the public sector areable to co-operate in developing solutions for docu-ment life cycle management.A life cycle, that notonly comprises the generation and use, but also thelong-term availability, and the guaranteed authentici-ty of documents.’

Dr Kampffmeyer is director of PROJECT CON-SULT, a consulting company for both industry andthe public sector, in the area of document relatedtechnology. He is also member of the board ofAIIM, the Association for Image and InformationManagement, and chairs the ICT-committee ofDLM, an abbreviation that used to stand for‘Donnees lisible par machine’ (machine readabledata), but has been changed to ‘Document LifecycleManagement’, representing the wider scope of thesubject. DLM-forum is a network initiated by theEuropean Commission, to stimulate co-operationand technology development in archiving.

According to Kampffmeyer, the traditional pro-blems of archiving, are the kilometres of paper docu-ments, whose integrity is threatened by their fragilityand a fading consistency. ‘These problems seem smallwhen compared to the archiving problems createdby the use of new technology’, he says. ‘The volumeof information is growing at an unprecedented pace.We already produce more information per year thanwe did in the whole period since we descended fromthe trees.A lot of this information is digital only,meaning it has no physical representation.That makesit much more volatile.An XML-document for instanceis created while you view it. So how do you keep it?’

Although generating volatile information, govern-ments want to use these technologies to make theirpolicies more transparent. Everyone must have accessto our books, comments Romano Prodi, the presi-dent of the European Commission. He is not alone;even the authorities of the smallest village want tobe accessible to its citizens via Internet and email.Kampffmeyer: ‘A lot of projects are put into gear onall levels of government, and most of them will fail,I’m afraid. Firstly, because the new transparency can-not be created by technology alone; a fundamentalchange in the administrative organisation is needed.Secondly, because there is not enough money provi-

ded. Ill-defined projects are tendered and awarded tocompanies offering the lowest price.Which, is ofcourse no guarantee for success.’

‘Thirdly’, Kampffmeyer continues, ‘the technologyis not ripe yet, in the sense that it cannot liveup to the demands of good governance.You can create electronic documents andyou can keep them in an electronicarchive. But, for example, an electronicsignature to authenticate a document isinvalid after three years; the migration ofdata and the accompanying loss isstill an unsolved problem,and the history of elec-tronic records can be tam-pered with quite easily.People from the indu-stry can show you anice graphic or run anice project, but that isnot the point.The pointis, that technology hasto be embedded inorganisations and pro-cedures, and that is cer-tainly not the case yet.’

According to Kampffmeyer, the technology is stillin its infancy, or, as he puts it, ‘at the beginning of itslife cycle’. ‘Archives of physical documents have beenin existence for over 6000 years, whereas, electronicdocuments have only been around for the last twen-ty years, or so.The people who develop the techno-logies, haven’t made the mental transition yet, fromcreating and using a document to the long termavailability of that same document.’

To mature, the technology has to be fostered, notonly by companies but also by the public sector. ‘Theindustrial approach is always one-dimensional’, hesays. ‘When there is a demand, they look for techni-cal ways to fulfil that. If the public sector goes onissuing ill-defined projects it becomes a downwardspiral of failure and frustration.The sensible thing todo is to try and develop standards, predefined structures, metadata and interchangeable formats,through co-operation between the public sector andindustry. Not on a local or national level, but on anEuropean and international level. Only through co-operation and co-ordination, can we realise the goalthat “electronic archives are the memory of theinformation society”, as Commissioner ErkkiLiikanen has put it.’

Only co-operation can prevent us from drowning

Interview with UlrichKampffmeyer,PROJECT CONSULT, Germany

by Joost van Kasteren

Photo: Project Consult

EVERY PROGRAM MAKER HAS TO BE BECOME A PART TIME ARCHIVIST.AN INTERVIEW WITH ANNEMIEKE DE JONG, NETHERLANDS INSTITUTEFOR SOUND AND VISION

‘In a production environment that becomesmore digital by the day, like the broadcastingorganisations, everyone has to become a part

time archivist.At least if you want to be able to makethe programs you want.’ Raising the awareness of themanagers of the twenty-eight public broadcastingorganisations in the Netherlands and of the producers,directors and other people involved in making tele-vision, radio and internet programs is an important taskof Annemieke de Jong, head of the Information PolicySector ICT of the Netherlands Institute for Soundand Vision (former Netherlands Audiovisual Archive).

‘At the moment, a program after broadcasting isarchived and made available for reuse and/or research’,she says. ‘A batch process with steps that are only

loosely linked. In the digital environment, the processof information production, broadcasting, archivingand reuse is almost a continuum.A digital workflowconsisting of bits and bytes from countless internaland external sources which can be tapped by anyonewith the right credentials.’

The Netherlands Institute for Sound and Vision(Sound and Vision) was founded in 1997 (under thename of NAA) out of a fusion between several organisations. It contains 600,000 hours of television,radio and film, 2 million photo’s and half a millionmusical recordings. Every year it grows by 5000hours of TV and film-images and about 15,000 hoursof audio.That is not all that is broadcasted. Soundand Vision selects the material according to criterialaid down in a handbook. Selection is based onhistorical importance, national interest, and customerdemands.The material is recorded and made availablefor reuse by the public broadcasting organisations andother professionals as well as for research.

DigiCULT 21

ARCHIVING IN THE AUDIOVISUAL PRODUCTION CONTEXT

A CASE STUDY ON THE NETHERLANDS INSTITUTE FOR SOUND AND VISIONBy Friso Visser and Pieter Kop


Every year Sound and Vision handles 100,000 appli-cations, most of them for reuse of material.

De Jong believes that the ongoing digitisation ofthe production, distribution, broadcasting and archi-ving of radio, television and (more and more) webprograms will lead to an unprecedented growth ofboth size and complexity of information. ‘To controlthe flow of information you have to have an infra-structure.You could compare it with the infrastructureof dykes, sluices and other works to control the flowand level of our rivers and streams. So, at the momentwe are developing standards and protocols to laydown format and content of digital data. In the nearfuture these standards will help to make and keep thedata accessible during all stages of production. Fromthe first idea of a program maker to the storage andpossible reuse in twenty-five or fifty or a 100 years.’

The standards and protocols will differ enormouslyfrom current practice. De Jong: ‘Traditionally we areat the end of the chain. Everything that is broadcastedis itemised and described by our specialised archivists.Not only the content of the program, but also theunderlying documents and the footage that has notbeen used. Quite a cumbersome work that in thelight of the growing information flow has to beautomated. Replacing a person by a robot will notsuffice though.What we aim for is a way of archivingwhereby metadata for storage and reuse are gene-rated during the whole process from idea to broad-cast and are linked to the content.This content,including related material like documents, will haveto be described and categorised in such a way that itcan be stored in a database and made available to our clients in a format that suits them.’

When asked, De Jong says that the informationwill not be stored in one single archive database, butin several databases for television, radio, photographs,musical recordings, catalogue descriptions, and props,such as material items that have been used in programs,like a clowns suit or the hat of a police inspector in apopular series. For easy access it is essential thoughthat these databases are linked through a single inter-face and that the metadata are compatible. Sound andVision is in the process of developing a standard formetadata, which is suitable for the audiovisual archivedomain.

These metadata differ from metadata of text in thesense that audiovisual content carries more implicitinformation (colour, shape) and by nature is moretemporal (time-dependent). Furthermore, the meta-data contain information on technical aspects likeformat and resolution.That is why Sound and Visionis heavily involved in developing standards for audio-visual metadata both internally and in co-operationwith other organisations in EBU (P/Meta project).

22 DigiCULT

Browser: P/Meta - Metadata Exchange SchemeP/Meta website, http://www.ebu.ch/pmc_home.html (Task forces)SCHEMAS activity report on P/Meta by A. de Jong (Review date: 2002-01-31), http://www.schemasforum.org/registry/desire/activityreports.php3Richard Hopper: EBU Project Group P/Meta. In: EBU TechnicalReview,April 2002, http://www.ebu.ch/trev_290-hopper.pdf

De Jong: ‘Developing of international standards is aslow process. On the other hand technology develop-ment is driving you forward. So we develop our ownstandards but that process is closely linked to thedevelopment of international standards.’

Pho

to:B

eeld

en

Gel

uid

A virtual archive can only function if it is strictlymanaged.At the Netherlands Institute for Sound andVision De Jong is involved in developing a digitalmedia archive management system that controls thedatabases, servers and the network, and the actualprocessing of data. De Jong: ‘You could say that itmanages the life cycle of the media objects, be itshots, items, summaries, key words or the object assuch.’A system like that has to be rather versatile. Ithas to be able to register and index any media object(image, sound, text) in such a way that it can beidentified; to support any existing and future mediatype; be able to migrate the content between mediatypes; and to be accessible in different ways either viacontent or metadata. It also must contain the neces-sary information on copyrights and authorisations.That is quite important because for most of the con-tent Sound and Vision does not own the copyright.

Implementing standards and protocols for archi-ving is not easy. Partly because of the way the Dutchbroadcasting is organised, with twenty-eight indepen-dent broadcasting organisations, which all have to bepersuaded to use the same standards and protocols.Partly also because programs are made by creativepeople who lack the discipline for archiving. DeJong: ‘If it would be only for archiving in the tradi-tional sense, people would not be motivated to storethe content they produce in a formalised way andadd the necessary metadata. But, because they expe-rience more and more the difficulties in retrievingdigital information, they recognise the need of stan-dards and protocols for storing data. I think if we cangive them easy-to-use tools they will be very eagerto co-operate, because it makes their work mucheasier.’

rely on the services provided by Sound and Vision’s160 employees for their publications and research.To do so, Sound and Vision keeps an impressivecollection of audiovisual material of which most iskept at the Mediapark in Hilversum, where theDutch broadcasting industry operates.2

The combined collections of Sound and Visioninclude over 600,000 hours of footage and sounds,which date back to the very beginning of cinema.Dutch television and radio programs were collectedfrom the first rise of these media until today.Apartfrom the objects in the Museum of Broadcast, thereare 300,000 movies, 125,000 videotapes, 2 millionphotographs, 17,000 hours of radio-broadcast, 60,000recordings of concerts and festivals, 100,000 compactdisks, 250,000 LP’s and 35,000 recordings about thehistory of broadcasting.Today, one of the largestmusic collections in the world, the collection isupdated on a daily base with new material.Preservation of the old material is an important issue.Better conservation techniques are being implemen-ted to safeguard the collection against decay.Anincreasing part of the collection is being digitised.TheDutch radio broadcastings are completely recorded and archived, with the exception of thebroadcastings of the two music stations.There is noneed to record music, which is already present in thearchives.The institute acknowledges, however, thatthe way music is being presented on the radio is infact part of cultural heritage.Therefore, twice a year,an entire week of broadcasting is recorded.

DigiCULT 23

The Netherlands Institute for Sound andVision (Sound and Vision)1, until veryrecently known as the Netherlands

Audiovisual Archive, was founded in 1997 after amerger between four organisations.These organisa-tions – the Archives of Public Broadcast, theNetherlands Government Film Service, the Museumof Broadcast and the Film Research Foundation – alljoined forces to preserve and exploit the Dutchaudiovisual heritage. Responsibilities vary from thoseof a cultural heritage institution to a source of infor-mation and material for the broadcasting industry.Amongst Sound and Vision’s main customers arepublic and commercial broadcasters, producers,journalists, schools and universities.These customers

ABOUT SOUND AND VISION

1Sound and Vision website:

http://www.beeldengeluid.nl2Within a so-called ‘Digital

Platform’, Sound and Vision

cooperates with Dutch public

broadcasters

(http://www.nos.nl) as well as

producers

(http://www.nob.nl) on issues

such as metadata standards.

The various departments within Sound andVision still use different cataloguingsystems. In the future, those different systems

have to be more closely related, sharing at least thesame metadata schemes and thesauri. Implementingthese changes have to be done incrementally.Thecatalogue needs to be accessible to clients seven daysa week, from early in the morning to late at night.Material for news items sometimes have to be deliv-ered within minutes, so an ‘under construction’ signis unacceptable.An advantage of incremental alterna-tion is the possibility to implement temporarily theagreements made in the discussions on internationalstandardisation issues. Possibilities to test new theo-ries are provided in this process.A clear disadvantage,more on the organisational level, is the psychologicalimpact on employees who have to work with fre-quently changing systems.The workflow basicallyconsists of acquisition, conservation/preservation andaccess.

Various publications on audiovisual research areissued in the form of annual reports, press releases,books and guides. Students and researchers have thepossibility to access part of the archive with help ofdigital catalogues.This still has to be done within thepremises of the institute and copying of the materialrequires permission of the holder of copyrights.Thestudy centre provides opportunities to find informa-tion about various aspects of the history of media.It holds readings on AV archiving and the use of AVmedia for purposes of historical research, magazines,guides, catalogues and various other documentations.Recent educational developments consist of onlineapplications of parts of the archives at schools anduniversities.

Most of the audio materials are held in thePhonotheque. Of each newly released album, thephonotheque acquires two copies, one of which isfor borrowing purposes. Each year, 11,000 disks areadded to the collection.The general public, as well asbroadcasters, have access to the vast collection withthe possibility to borrow almost any kind of song,stock-music or sound-effect.The phonotheque alsointermediates in copyright issues between creatorsand users.

Originating from the Museum of Broadcast,Sound and Vision has a permanent exhibition withabout 20,000 objects displayed for the general public.An important part is formed by artefacts that arerelated to the history of radio and television in theNetherlands.The focus, however, is shifting to a

more interactive environment, where AV-materialscreate an atmosphere of nostalgia.At the moment, abuilding is being constructed at the Mediapark tohouse both the museum and the professional servi-ces.This new home for Dutch audiovisual heritagewill be opened in 2005 (see model on page 23).

24 DigiCULT

SERVICES

INTERNATIONAL

Activities of Sound and Vision are not limitedto the country’s borders.The institute ispresently a board member of the

International Federation of Television Archives(FIAT-IFTA) and chairs its Committee onDocumentation. International cooperation occursmostly with European public broadcasters like theBBC, RAI, ORF, SVT and SWT. It also takes part intwo workgroups of the European Broadcast Union(EBU), where new digitisation standards for produc-tion, broadcasting and archiving processes of radioand television are being developed. Multi MediaContent Description Interface, or MPEG-7, is one ofthe recent projects on standardisation and expectationsare high. Other international projects in whichSound and Vision operates are AMICITIA3, ECHO4

and PRESTO5. Most of the cooperation is focusedon development of metadata standards, automatedarchiving systems and forms of digital preservation.6

Participation in ECHOECHO is an acronym for European Chronicles On Line.The projectstarted in 2000 as part of the European Commission’s ‘Fifth Framework’series of projects.The goal of ECHO is to develop a universal softwareinfrastructure for support of digital AV-archives. Four national AV archi-ves, as well as technical and academic organisations, participate in theproject.The archives are from Switzerland (Memoriav), France (InstitutNational de l’Audiovisuel), Italy (Instituto di Luce) and The Netherlands(Sound and Vision). Enhancing access to non-english archive materialsis the focus of the project.Access to such materials is often difficult,although the conserved and disclosed materials are no less of importancethan the english ones. ECHO provides a forum for cooperation betweenAV archives through standardised retrieval protocols. Selected materialwill be accessible via a web interface. Images are easier identified withadvanced search methods and multi-lingual access.Automatic extractionof annotations from the context uses a combination of speech indexing,language interpretation and image recognition, which requires the con-tent to be digitised for analysis. Before finishing the project in August2002, the tasks of Sound and Vision involve formulation of user-demands, designing a suitable corpus, assisting in the development ofmetadata standards and coordination of the evaluations.

Storage, and subsequent retrieval of thousands ofimages, sounds and films, requests an orderly datamanagement system. For this, metadata are vital.Themerger of the four organisations posed problems forthe exchange of information. Systems were obsolete,incompatible and have to be standardised.To accom-plish this, metadata standards are being introduced tothe cataloguing systems.

Metadata could be defined as ‘data about data’.They describe sources of information and classifythem.The type of metadata used, depends on thedomain it describes and its purpose, such as identifi-cation or retrieval of data. Ideas about the grasp ofmetadata differ from purely data on content todescription of the entire process of creation, organi-sation and software architecture

Exchange of digital information requires the use of standards to allow interoperability of systems anddata contained in them. In the first place, standardisa-tion within an organisation is essential. Secondly, asglobalisation continues, metadata have to be standard-ized on a national and even international level toguarantee permanent sustainability. Organisations,who develop standards for AV-metadata, come fromthe world of broadcasters,AV-producers and archi-vists.Their operations can vary from a local levelwithin specific fields, to all-including and globalstandardisation of metadata.

Recent developments regarding standards are:° Development of metadata taxonomy to facilitate

communication about metadata.° Development of ontologies, linked to metadata

features and description of data-elements like fieldnames, types, classification and semantics.

° Defining a central structure of registration in order to facilitate the mapping between different metadata schemas.

° Defining a set of tools with generic functionalities to develop and exploit metadata.

Vision, as the metadata, which is added by the broadcasters (creators), is increasingly tuned to themetadata standards designed by Sound and Vision.Although the creator has no direct advantage in usingthe same metadata standards, in the long run thiscooperation facilitates reuse of the material,preserving its authenticity at the same time.Workingwith the same metadata standards, the character ofthe metadata differs.The creator should generatemetadata, which has to identify the record, whereasthe preserver needs to add its own metadata thatdescribes the processes of migration and dissemi-nation of the record.

The concepts of authenticity and integrity of datain relation to AV materials are not very strictly defined.The original data are kept in their raw form as muchas possible. However, it is a matter of fact that thematerials are often edited.The news, for instance,might be recorded including the footage of a TVcomment or also including those items not broadcastat all. Integrity is something that was tampered within the past. Being a corporate archive, when materialswere used, they would be often returned (or not atall) in its edited version. Nowadays, the original isalways kept in store, while viewing or lending copiesare made available.

The following example indicates how Sound andVision tries to preserve authenticity of its non-digitalcollection.When the program ‘Big Brother’ appearedon Dutch television, the public opinion was dividedabout its consequences. Psychologists and sociologistscondemned the program, pointing at potential dangersfor both viewers and participants, while the generalpublic embraced this new way of making television.Millions of people would watch the show every dayand the ones that did not, found it increasingly difficult to strike up a conversation with one of theviewers.This controversial program began to look atthe relationship between privacy and television in anew way. For this contribution to the cultural heritagealone it was worth preserving the concepts, pro-grammes and related material.The project leader ofSound and Vision’s museum, exploring the possi-bilities of an exhibition dedicated to Big Brother,came up with the idea of showcasing one of thecamera’s that was actually used in the program. In hisattempts to find the authentic cameras, he experiencedsome difficulties in explaining the concept of authen-ticity to the employees of the technical service,which exploits the equipment.They did not under-stand why this one specific camera, out of hundredsof identical cameras, had to be identified and displayedin a museum.

DigiCULT 25

INTEGRITY AND AUTHENTICITY

Integrity and authenticity are abstract concepts,but preserving them is imperative in every stageof the workflow. Criteria for authenticity are

based on the components of a digital object, the kindof object (such as text, audio) and the type of usagesuch as scholarly use or entertainment. In preservinga digital object’s authenticity, there is a need for twodifferent sets of requirements, for both creator andpreserver.This has been acknowledged by Sound and

3AMICITIA website:

http://www.amicitia-project.de4ECHO website:

http://pcerato2.iei.pi.cnr.it/

echo/5PRESTO website:

http://presto.joanneum.ac.at6An impressive guide on

metadata and standards is a

publication of Annemieke de

Jong, who works for SOUND

AND VISION. Unfortunately

this is only available in Dutch.

Jong,A. de (2000):

Metadata in de audiovisuele

productieomgeving. Hilversum,

The Netherlands.

The BBC has one of the biggest audio-visualarchives in the world, analogue mountains ofarchival material to which on an increasing

scale of digital material is added, because for aboutfour years, everything created in the BBC has beenon a digital format.

As Paul Fiander, Head of BBC Information &Archives, described the situation in the Forum inBarcelona:‘If you look at the BBC, we started seventy-five years ago collecting material. In the modern era theBBC has traditionally transmitted from two televisionchannels and five radio channels.The BBC now hasseven television channels and will soon have over twentyradio channels, in addition to a new media site that createsthousands of new pages every month.We are also justabout to start permanently broadcasting in interactivewhere for every traditional stream of linear content,lets take a sport, we now have six streams, such ascamera position, information, etc. So, whatever prin-ciples we agree on retention and archiving, suddenlyour job is expanded by factors of ten, twenty or even100.We employ at the moment within the BBC 250people solely related to the archive. If you multiplythe content without finding some automatic way ofdoing it, you can see that the number of people werequire will get beyond any number reasonable toemploy.Think for a minute, we are just talking aboutnew content being created today. I have not evenmentioned or proposed a method to deal with theanalogue mountains created in the past.’

According to Fiander, they have spent about 35million Euro so far in preserving analogue material.The BBC will probably spend up to 90 million euroon the whole preservation process, and a lot morethan that in managing these assets.

BBC Information & Archives is a business unit thatnot only has to demonstrate to UK licence fee payersand professional archivists the quality and standards ofhow it looks after the BBC’s broadcast material. It alsohas to demonstrate to the management and stakeholdersthat it is supportive in providing value for money in thearena of public service TV. In doing so, bringing analoguearchival material into the digital world is essentially aboutfacilitating better accessibility, reuse, and valorisation.

In the light of volumes of digital material that arenearly on an exponential scale of growth, Fiander callsfor clarification of principles and pragmatic solutions.

For example,‘authenticity in an audio-visual world,does that say we should not compress, what do we loseby MPEG-1-2-21?’ Some national archives do notaccept material that has been compressed (becausethey cannot guarantee that it can be rendered to itsoriginal state). But, if you are the BBC and facingsuch an enormous growth of material, how do youaccommodate to that? Can you still keep everythingthat is deemed valuable, or do you have to adapt yourselection and retention policy, to match your funding?

Actually, the BBC does not keep everything,because ‘lack of space, lack of funds whatever criteriayou want to apply, is forcing us to change our selectionand retention policy. If I give you an example fromtelevision, we will keep all of the landmark series, allthe good programmes, but , we won’t keep all thecookery programmes, and we certainty will not keepall the quiz shows, we will keep examples of those.The content volumes are rising, so we will inevitablybe more selective unless we can find technologicalsolutions that enables us to reduce the unit cost ofselection, retention and storage.Without these tech-nological solutions you may be tempted to leave it tothe market alone, such as those items that have thehighest short term re-use value in which case wewill focus highly on sports, the wealthiest user group.’

In the Forum discussion, Nils Brübach stated that‘the appraisal policy should not be simply based oneconomic matters’, but be defined beforehandinstead of ‘seeing it as a budgetary reflex’. KenThibodeau added, that the digital environment mightwell lead to the beneficial situation that players like,for example, the Universal Studios ‘keep a lot morematerial than they ever did before, because they canslightly repurpose it and create new markets. So thearchives become a revenue generator.’ Fiander affirmed, that representatives of such companies see ‘a new exploitation track, but who wins again, I amsorry but we are back with sport again.’ (-gg)

26 DigiCULT

Pho

to:B

eeld

en

Gel

uid

THE DIGITAL

DigiCULT 27

BROADCASTER EXAMPLE: BBCAPPRAISAL, INTEGRITY AND AUTHENTICITY IN THE LIGHTOF A TREMENDOUS SCALE OF GROWTH OF VOLUMES:

The mission statement of Sound and Visioncan be divided into two parts. On the onehand, it is the central archive of Dutch

public broadcast, managing a broad selection of television and radio programs, music and requisites.On the other hand, Sound and Vision has set itselfthe difficult task to preserve AV-material and securethe permanent access to the collection for the generalpublic.This hybrid system posed some problems andmisunderstandings in the past. For instance, a journa-list wanted to have one of the recently broadcastnews items removed from the archive and changed,because he had made a mistake and was not satisfiedwith the results of his creation.To his surprise, thisrequest was not granted.

In the 1960’s, a popular serial for children was broad-cast on Dutch television.The tapes, on which thematerial was stored, were expensive and the decisionwas made to reuse the tapes for new productions.The serial was lost forever.A decade later, this amnesiaof memory institutions returned at the expense ofthe popular serial ‘Dr.Who’, which mysteriously disappeared from the BBC archives.

Also because original materials were borrowed andnever returned, information from the archives havebeen lost.The only recordings of Queen Beatrix’scoronation went missing from the archives for half ayear, but were fortunately recovered. Some materialthat was borrowed and never returned is now proba-bly part of someone’s private archive.These incidentsmade Sound and Vision change its policy towardsborrowing and, with the exception of material ofthe phonotheque, original sources are first copiedbefore beeing borrowed.

next few years. Some of its elements are consideredessential and urgent enough to realise within twoyears in a basic system.This pilot will consist of acatalogue module with an interface to a new auto-mated thesaurus.Avail is the name of the presentcataloguing system of radio and television, whichruns under UNIX. Both, Sound and Vision and thepublic broadcasting industry use this system, so thenew catalogue in iMMix has to be developed incooperation with the broadcasters.

A current development in the AV environment isso-called tape less production. In the ideal situation,the production, broadcasting, archiving, retrieval andreuse, should be fully digitised.The barriers betweenthis future situation and the present one are notablytechnological. Digital storage is only feasible for asmall part of the AV collection.There is simply toolittle capacity. Given the contemporary technologies,digitisation would mean replacing tapes by disks, orbetter, expanding the existing collection with disks,since the original will never be discarded.Furthermore, this material cannot be accessed onlinecontinuously, which makes additional human inter-vention necessary.

The ever-increasing pace of change in computerhardware and software is not the only thing threaten-ing the collection.Also the players of AV-materialhave been evolving. For some old tapes, there is onlyone player left in The Netherlands, with the onlysource for spare parts being East-German tape record-ers. It is clear that Sound and Vision is an organisation,which will always be confronted with changingsituations.The Institute is gradually playing a moreimportant role as an intermediary, increasingly settingitself proactive tasks in production and exploitationof new AV material. Fighting for preservation of arapidly growing collection, Sound and Vision willhave to develop a way to sustain and handle theworkflow within a highly dynamic environment.

For further information on Sound and Vision please contact:Netherlands Institute for Sound and VisionMrs.Alette AhsmannCommunication ManagerMedia Park - Sumatralaan 45Postbus 1060NL-1200 BB HilversumPhone: +31 (0)35-6774720Fax: +31 (0)35-6773208E-mail: [email protected]

28 DigiCULT

LESSONS LEARNED

FUTURE SCENARIO

Integration is the magic word for the future ofSound and Vision. Integration of standards,methods, technologies and services will absorb a

lot of time. iMMix is an asset management systemwhich supports management of rights, metadata andessence. It will enhance acquisition, conservation,indexing, retrieval, borrowing and Customer Relation-ship Management (CRM). Even the financial admin-istration will be integrated in the system.The project‘Store-it’ deals with the creation of an infrastructurefor digital storage and distribution of essence.Theresults will also be implemented in iMMix.Theinformation system will be implemented during the

DigiCULT 29


Reading Lists on Authenticity & Integrity:

Association of Canadian Archivists,ACA Institute2002:Approaches to the Preservation of ElectronicRecords. Reading list, http://archivists.ca/institute/2002/readings.htm

PADI (Preserving Access to Digital Information)website:Authenticity,http://www.nla.gov.au/padi/topics/4.html

Readings:

Authenticity in a Digital Environment,Council on Library and Information Resources, May2000. http://www.clir.org/pubs/reports/pub92/pub92.pdf.This report is based on a workshop orga-nised by CLIR in January 2000.The workshopaimed to clarify the meaning of ‘authenticity’ whenrelated to digital information. Papers by five authors- Charles Cullen, Peter Hirtle, David Levy, CliffordLynch, and Jeff Rothenberg address questions askedby CLIR about what authenticity means.

Bearman, D. and Trant, J. (1998):Authenticity ofDigital Resources:Towards a Statement ofRequirements in the Research Process, D-Lib magazine,June 1998, http://www.dlib.org/dlib/june98/06bearman.html

Brodie, N.:Authenticity, Preservation and Access inDigital Collections. In: Preservation 2000:AnInternational Conference on the Preservation andLong Term Accessibility of Digital Materials, 7/8December 2000,York, England. Conference Papers,http://www.rlg.org/events/pres-2000/brodie.html.

Day, M. (1999): Metadata for digital preservation:an update (12/1999).Ariadne 22,http://www.ariadne.ac.uk/issue22/metadata/intro.html

InterPARES Authenticity Task Force, AuthenticityTask Force Final Report, http://www.interpares.org/

OCLC/RLG Working Group on PreservationMetadata (2002): Preservation Metadata and theOAIS Information Model:A Metadata Framework toSupport the Preservation of Digital Objects. June2002. http://www.oclc.org/research/pmwg/pm_framework.pdf.The report is a comprehensiveguide to preservation metadata that is applicable to abroad range of digital preservation activities. It isintended for use by organisations and institutionsmanaging, or planning to manage, the long-termretention of digital resources.

Russell, Kelly (2000): Digital Preservation and theCedars Project Experience. In: Preservation 2000.Conference Papers, http://www.rlg.org/events/pres-2000/russell.html

Waters, D. and Garret, J. (1996): Preserving DigitalInformation. Report of the Task Force on Archivingof Digital Information, commissioned by theCommission on Preservation and Access and theResearch Libraries Group.Washington, D.C.:Commission on Preservation and Access.http://www.rlg.org/ArchTF/ This report arose from a decision at the end of 1994by the Commission on Preservation and Access(CPA) and the Research Libraries Group (RLG) tocreate a Task Force charged with investigating andrecommending means to ensure ‘continued accessindefinitely into the future of records stored in digi-tal electronic form.’This was a watershed exercisewhich generated thoughtful discussion on this topicworldwide.

30 DigiCULT

LITERATURE

Essential Reading Trusted Digital Repositories:

Trusted Digital Repositories:Attributes andResponsibilities.An RLG-OCLC Report, MountainView, CA., Research Libraries Group (RLG), May2002.http://www.rlg.org/longterm/repositories.pdf‘Just a few years ago, the development of trusteddigital repositories seemed far in the future, buttoday it is an immediate challenge.The expert involve-ment and community consensus developed duringthe course of this work suggest that organizationsand funding agencies will need to work together inthe very near future to address the needs articulatedin this report.’RLG and the Online Computer Library Center(OCLC) have issued a report on how to ensure reliable access to digital information over the longterm that is freely available at the RLG Web site.Thereport addresses the attributes trusted, reliable,sustainable digital repositories must have and discuss-es requisite responsibilities at both the higher orga-nisational/curatorial level and the operational level.nFinally, the report looks at how repositories can becertified and summarizes key recommendations.Anappendix provides technical overviews of theReference Model for an Open Archival InformationSystem (OAIS) — a common framework for descri-bing and comparing architectures and operations ofdigital archives. Compliance with this model is pro-posed as a defining attribute of a trusted digital reposi-tory. RLG is particularly interested in the design andimplementation of a certification program for trusteddigital repositories.This will produce tools and guidancefor institutions responsible for digital collections –whether they are creating their own repositories,working with publishers, or planning to contract forthird-party services.The report also includes a basicoperational responsibilities checklist for trusted digitalrepositories.The report is a major step in definingtrusted digital repositories and providing the basis foreffective action. It is primarily intended for culturalinstitutions such as libraries, archives, museums, and

scholarly publishers and is specifically aimed at thosewith traditional or legal responsibilities for the pre-servation of cultural heritage. It is written to aidsenior administrators as well as those implementingdigital archiving services.

This summary has been abstracted from the RLGand OCLC Issue Final Report on Trusted DigitalRepositories:Attributes and Responsibilites,http://www.rlg.org/pr/pr2002-repositories.html

DigiCULT 31

Julien van Borm

Dr. Julien van Borm has been Director of UIALibrary University of Antwerp since 1989, and acts asmanaging director of Anet, the library automationcenter of the University of Antwerp, Belgium. Otherduties include for example: President of the Councilof Flemish Public Libaries, Member of the FlemishCultural Council, Member of the BelgianConference of University Libraries and the RoyalLibrary, Member of Sabido, the libraries workgroupof the Dutch-Flemisch Language Union. Dr. Bormhas worked on EU level in the framework of theIST-program and for the Belgian Focal Point for theEC IST program.

Nils Brübach

Dr. Nils Brübach is senior lecturer for archivalScience (Archivoberrat) at the Archivschule Marburg,Germany. He is a delegate to ISO/TC46/SC11‘Information and Documentation’, Member of theEditorial Board for the ISO 15489 (‘RecordsManagement’), and Member of ICA's committee ondescriptive standards.

Luciana Duranti

Luciana Duranti is Chair of and a Professor in theMaster of Archival Studies Program (MAS) at theSchool of Library,Archival and Information Studiesof the University of British Columbia, Canada,where she has taught since 1987 and has occupiedthe position of Associate Dean Research for theFaculty of Arts. She has been President of the Societyof American Archivists, and is active nationally andinternationally in several other archival associations.She is presently Project-Director of InterPARES.

Paul Fiander

Paul Fiander is Head of Information and Archives,BBC, United Kingdom. He first joined BBC World-wide in 1992. In June 1998, he was given the task ofpositioning the business unit Information & Archives(I&A) for the digital age and stabilising it’s losseswhich were running at approximately £2m per year.His main responsibilities are to give I&A financialstability, to take a keen interest in operational issuesof the department, and to demonstrate to the outsideworld, both licence fee payers and professional archi-vists, the quality and standards of how I&A looksafter the BBC’s broadcast material. One of his mainfunctions with the management team is to formulatethe strategy for I&A’s future.To ensure it builds theskills, systems and technology to put I&A in the bestposition to help it’s customers, stakeholders and staff.

Hans Hofman

Hans Hofman is working as senior advisor for thegovernment program ‘Digital Longevity’ at theNational Archives of the Netherlands.This programhas the objective to establish policies for electronicrecord keeping within government, including digitalpreservation. On the international scene he has beenMember of the Committee on Records in an Elec-tronic Environment (CREE) of the InternationalCouncil on Archives, Member of the DLM-MonitoringGroup within the European Union (since 1996),co-investigator and representative of the NationalArchives of the Netherlands in the InterPares researchprojects, and co-director of the recently startedEuropean project ERPANET (Electronic ResourcePreservation and Access Network) on digital preser-vation. Since 2000 he represents the Netherlands inthe TC46/SC11 for developing the ISO RM stan-dard 15489.

32 DigiCULT

THE BARCELONA FORUM PARTICIPANTS

Annemieke de Jong

Annemieke de Jong is Head of Information Policyat the Netherlands Institute for Sound and Vision(until July 2002: National Audiovisual Archive). Inthis function she designs strategies for positioning AVarchives in the digital era and develops policies forpreservation and reproduction of digital audiovisualmedia. Before, she worked as senior archivist at theCentral Archive of the Public BroadcastingCompanies (FBA), 1982-1990, and manager of itsTelevision Archives, 1990-1997; she also served asHead of Catalogue of the National Audiovisual Archive,1995-1998. Her international activities include:Member of the Executive Council and Chairpersonof the Documentation Commission of the InternationalFederation of Television Archives (FIAT/IFTA),1994-2001; Member of the Media ManagementCommission of the FIAT/IFTA, 2001-; projectmanager of the National Audiovisual Archive (NAA)for the IST-Project ECHO, 1999-; Member of theP-Meta project group of the EBU, 2000.

Max Kaiser

Max Kaiser is researcher at the Austrian NationalLibrary. He holds a master's degree in german studies and philosophy, and since 1995 worked in research projects for the University of Vienna,German Department.At the Austrian NationalLibrary he is responsible for the Austrian InternetPortal for Literary Archives (KOOP-LITERA,http://www.onb.ac.at/koop-litera/) and the EU-projects MALVINE (http://www.malvine.org) andLEAF (http://www.crxnet.com/leaf/).

Adelheit Stein

Adelheid Stein is Senior Researcher at theFraunhofer IPSI, Germany, where she has beeninvolved in several European IT projects, and univer-sity teaching. She is currently the head co-ordinatorof the COLLATE project (Collaboratory forAnnotation, Indexing and Retrieval of DigitizedHistorical Archive Material; IST-1999-20882).

Kenneth Thibodeau

Kenneth Thibodeau is Director of the ElectronicRecords Archives (ERA) Program at the NationalArchives and Records Administration (NARA).TheERA Program is building the archives of the futurefor NARA, virtual archives capable of preserving andproviding access to historically valuable records ofthe Federal Government. Mr Thibodeau has twenty-six years experience in archives and records manage-ment and is an internationally recognized expert inelectronic records. He has taught at the University ofNotre Dame and was Chief of the RecordsManagement Branch of the National Institutes ofHealth before coming to NARA in 1988. In 1996,he served as the Director of the Department ofDefense (DoD) Records Management Task Force,which revised DoD’s Records ManagementInstruction and developed the DoD RecordsManagement Application Standard, 5015.2-STD.AFellow of the Society of American Archivists, he haspublished over thirty papers and spoken at more than120 conferences around the world.

DigiCULT 33

DigiCULT is an IST Support Measure (IST-2001-34898) to establish a regular technology watch thatmonitors and analyses technological developmentsrelevant to and in the cultural and scientific heritagesector over the period of thirty months (03/2002-08/2004).

In order to encourage early take up, DigiCULTproduces seven Thematic Issues, three TechnologyWatch Reports, along with the newsletterDigiCULT.Info.

DigiCULT draws on the results of the strategicstudy ‘Technological Landscapes for Tomorrow’sCultural Economy (DigiCULT)’, that was initiatedby the European Commission, DG InformationSociety (Unit D2: Cultural Heritage Applications) in2000 and completed in 2001.

Copies of the DigiCULT Full Report and Exe-cutive Summary can be downloaded or ordered athttp://www.digicult.info.

Project Consortium of DigiCULT:

Salzburg Research Forschungsgesellschaft (Projectco-ordinator) http://www.salzburgresearch.at

HATII – Humanities Advanced Technology andInformation Institute, University of Glasgowhttp://www.hatii.arts.gla.ac.uk

Pricewaterhouse Coopershttp://www.pwcglobal.com

For further information on DigiCULT pleasecontact the team of the project co-ordinator:

Mrs.Andrea Mulrenin,[email protected]: +43-(0)662-2288-304

Mr. Guntram Geser,[email protected]: +43-(0)662-2288-303

DigiCULT Forum secretariat:Mr. John Pereira, [email protected]: +43-(0)662-2288-521

Salzburg Research Forschungsgesellschaft m.b.H.Jakob-Haringer-Str. 5/IIIA - 5020 Salzburg Austria Phone: +43-(0)662-2288-200Fax: +43-(0)662-2288-222http://www.salzburgresearch.at

Project partners:

HATII - Humanities Advanced Technology andInformation Institute, University of Glasgow Contact: Seamus Ross,[email protected]

Pricewaterhouse Coopers ConsultingContact: Mr. Friso Visser,[email protected]

The members of the DigiCULT SteeringCommittee are:

Philippe Avenier, Ministère de la culture et de lacommunication, FrancePaolo Buonora,Archivio di Stato di Roma, ItalyCostis Dallas, Critical Publics SA, GreeceBert Degenhart-Drenth,ADLIB InformationSystems BV,The NetherlandsPaul Fiander, BBC Information & Archives, UnitedKingdomPeter Holm Lindgaard, Library Manager, DenmarkErich J. Neuhold, Fraunhofer IPSI, GermanyBruce Royan, SCRAN, United Kingdom

34 DigiCULT

DIGICULT: PROJECT INFORMATION

This Thematic Issue is a product of theDigiCULT Project (IST-2001-34898).

Authors:Guntram Geser, Salzburg Research Joost van Kasteren, Journalist Pieter Kop, PwC Consulting John Pereira, Salzburg Research Seamus Ross, HATII, University of Glasgow Michael Steemson, Caldeson Consultancy Friso Visser, PwC Consulting

Images:If not otherwise indicated, the images in thisThematic Issue are provided by and reproduced withkind permission of the Nederlands Instituut voorBeeld en Geluid / Netherlands Institute for Soundand Vision, Hilversum,The Netherlands. –Photography Liek Bouma

Graphics & Layout:Peter Baldinger, Salzburg Research

Printed in Austria©2002

DigiCULT Thematic Issue 1 builds on the firstDigiCULT Forum roundtable that discussed the sub-ject of Integrity and Authenticity of Digital Objects.This expert roundtable was held in Barcelona onMay 6th, 2002, in the context of the DLM-Conference 2002.

DigiCULT Thematic Issue 2 will follow theexpert roundtable on Digital Asset Management forCultural Heritage Institutions, that will take place inEssen, Germany, on September 3rd, 2002, in the con-text of the AIIM Conference @ DMS EXPO.

IMPRINT

DigiCULT 35

DigiCULT Consortium: