D-Lib Magazine
July/August 2004

Volume 10 Number 7/8

ISSN 1082-9873

The Role of ERPANET in Supporting Digital Curation and Preservation in Europe

Seamus Ross
Director HATII [1] and ERPANET [2]
<s.ross@hatii.arts.gla.ac.uk>

Introduction

At the conclusion of the Lisbon Council in March 2000, the European Union Member States agreed that by 2010 Europe should be 'the most competitive and dynamic knowledge-based economy in the world...' [3] Long-term access to reliable and authentic digital materials is at the heart of delivering this objective; they form the building blocks for learning and leisure, for e-government and public sector accountability, for scientific and historical research, and for business in the 'ambient intelligence landscape'. The European Commission and the Swiss Government, recognising that action was needed in the area of digital curation and preservation if this vision were to be delivered, agreed in 2001 to support ERPANET (Electronic Resource Preservation and Access Network). ERPANET (IST-2001-32706), led by the Humanities Advanced Technology and Information Institute (HATII) at the University of Glasgow (United Kingdom), and its partners the Schweizerisches Bundesarchiv (Switzerland), ISTBAL at the Universit à di Urbino (Italy) and Nationaal Archief van Nederland (Netherlands), works to enhance the preservation of cultural and scientific digital objects. Our initial funding stream of some 1.2 million euros runs for thirty-six months until November 2004. As generous as the funding from the Commission and the Swiss Government may seem, as we shall see below, without the commitment of professionals from across Europe, Australia and New Zealand, and Canada and the United States, ERPANET would not have been able to achieve its objectives. These experts have donated time, thought, effort, and knowledge.

ERPANET addresses the lack of awareness, fragmentation of knowledge and skills amongst the stakeholder communities about how to handle existing digital preservation problems, and the paucity of guidance as to how to plan effectively for the future. It addresses the lack of identification and focus on core research/problem areas, and brings coherence and consistency to activities in this area. ERPANET helps its community to develop skills and know-how to protect and to exploit to its full potential this intellectual wealth—Europe's next generation of renewable resources. ERPANET facilitates the realisation and exchange of knowledge on state-of-the-art developments in digital preservation and the transfer of expertise between individuals and institutions. More specifically ERPANET delivers research, content creation, advisory services, training seminars, thematic workshops, case studies, and outreach to both the information creation and user communities. It makes accessible tools, knowledge, and experience. After the team designing ERPANET had reviewed current practices, needs, and challenges posed by digital curation and preservation, while preparing our funding application in 2000 we set ourselves the following specific objectives:

• To identify and raise awareness of sources of information about the preservation of digital objects across the broad spectrum of national and regional cultural and scientific heritage activity in Europe.
• To appraise and evaluate information sources, to document developments in digital preservation, and to make available results of research, projects and best practice.
• To provide an online advisory service on digital preservation issues, practice, technology, and developments.
• To implement a suite of thematic workshops to bring together experts from a range of disciplines to examine key preservation issues and to initiate associated thematic discussion.
• To run training seminars based on best practice and to identify where and what further practitioner training and staff development initiatives might be undertaken.
• To conduct research through case studies that would improve our understanding of practices, needs, and future developments.
• To stimulate further research on digital preservation in key areas and encourage the development of standards where gaps and opportunities have been identified. For example, the ERPANET team in Glasgow has studied how to improve the ingest of digital materials into repositories.
• To stimulate ICT companies and software developers to incorporate some of the curation and preservation thinking into newer generations of software.

To achieve these objectives we established a work programme designed to support assessment of projects and publications, to enable us to run workshops, to facilitate the development of information resources, and to underpin research.

The Work of ERPANET

Four directors manage ERPANET on behalf of the main partners in the project. The Directors of ERPANET Niklaus Bütikofer (Schweizerisches Bundesarchiv), Maria Guercio (ISTBAL, Università di Urbino), Hans Hofman (Nationaal Archief Nederland, Den Haag), and Seamus Ross (HATII, University of Glasgow) work as a team to develop and refine our initiatives. Only a portion of the time each of the directors commits to the project is funded by either the Commission or the Swiss Government. As a result our institutions have made significant contributions to underwriting the costs associated with this project. The work of the directors is supported by Content Editors whose costs are fully met by our funders. There is one Content Editor based at each of our four nodes covering the British Isles and Nordic Countries, Northwest Europe, Mid-Europe, and Southern Europe. The Co-ordinator (Peter McKinney) is based at HATII (University of Glasgow) [4]. The ERPANET website (http://www.erpanet.org) is being developed as a key outreach tool and as a community resource. Although the project has a large number of activities (more details of which can be found on its website), here I will discuss three:

1. Assessments and Advisory Services,
2. Thematic Workshops and Training Seminars, and
3. Case Studies.

Assessments and Advisory Services

The quantity of literature on topics related to digital curation and preservation continues to grow quickly. It is increasingly difficult to decide what is worth reading; ERPANET is working to address this obstacle. Its team of Content Editors identify information sources, collect materials, and evaluate documents, standards, technical guidelines, research reports and web-published material. They review and appraise relevant research and development projects and related actions under EU Framework Programmes, US National Science Foundation (NSF) funding streams, and other national and international funding programmes. We regularly survey about ninety-five publications to identify literature which might prove a candidate for review (see the ERPANET website for the list). Each week the Directors and the Content Editors discuss candidate publications and select those that we will review. The resulting commentaries are designed to give the reader an understanding of the work in question, to set it in context, and to draw from it the benefits that it brings to an organisation, the new methods it suggests, or the new research that its results indicate should be conducted next. The objective is not merely to produce an abstract of the work in question, but to create a commentary on it that adds value through an analysis of work.

As well as using our knowledge to signpost key literature, the ERPANET team have established an online facility that allows those seeking answers to questions about digital preservation to post them. Wherever possible, the Content Editors and the Directors provide answers. The Directors of ERPANET have been drawing the attention of audiences to the service at all of our own seminars and workshops, as well as at the roughly seventy conferences at which we have represented ERPANET during its first thirty months. We are currently working on a redesign of this service to make it more effective and strongly encourage the community to take advantage of it.

In addition to our online advisory service, we have created best practice guidelines in the areas of preservation cost modelling, risk assessments, technology selection, and digital preservation policy design and development. We are constructing other tools: tools on ingest, audit and evaluation, and metadata are examples. These guides are not intended to be comprehensive, but have been designed as accessible starting points for organisations having little knowledge about digital curation and preservation and looking for initial guidance and thinking.

These services are now supplemented by ERPAePRINTS, which was launched at IFLA2003 (Berlin) [5]. ERPAePRINTS enables authors to make their work available through a reliable, managed, and OAI-compatible service [6]. The service aims to make accessible through a single portal literature related to digital curation and preservation whether the submitted work has been published or is grey literature [7]. Usage of ePrints services has been quite popular in some domains, while in other domains take up has so far been limited. Over the next couple of years we hope that usage of these services will continue to grow. We continue to look for new services we can add to our portfolio, and most recently we launched erpaDirectory, which provides brief descriptions of projects and resources related to digital curation and preservation. It is possible to submit links to this service, and if they pass the checks for content and relevance, they are mounted on the site.

Thematic Workshops and Training Seminars

Access to information meets some of the needs of the community; discussion and training meets others. To this end we have launched a range of thematic workshops and training seminars, and these have attracted just over 100 speakers and just under 800 participants. The workshops, usually three-day events, enable groups of experts to explore core preservation topics. So far these workshops have examined the relationship between digitisation, digital preservation and conservation (Toledo, June 2002), XML for Digital Preservation (Urbino, October 2002), the Long Term Preservation of Databases (Bern, April 2003), Trusted Digital Repositories (Rome, November 2003), and Audit and Certification in Digital Preservation (Antwerp, April 2004). The workshops aim to increase the level of awareness within the communities of digital information providers, practitioners, experts, and preservers. After each workshop, ERPANET produces a report that summarises the presentations and discussions, and makes recommendations for action, improvements to practices, and new directions for research. These reports are released on our website, and they provide an overview of cutting edge thinking on the issues addressed by the workshops.

The workshops give participants access to authoritative sources of information and experience that can be further analysed in different environments and enable participants to contribute to agenda setting exercises. ERPANET's Directors and the Content Editors develop the workshops over a number of months. We begin by identifying a problem domain. We then draw together our preliminary thinking. Next we build an initial list of speakers and engage in discussion with them about the issues. As a result of our internal discussions and those with potential speakers, we draft a briefing paper that forms the backbone of the workshop. This paper and the workshop agenda are available before the event on the ERPANET website. The presentations delivered at the workshop (and in some cases recordings of the presentations themselves) and the report of the workshop are added to the website after the meeting.

What is evident from these workshops is that the community is hungry for practical case studies and reports of real world experiences. The small size of the pool of experts makes it difficult to identify speakers with suitable technical expertise and presentation skills. The success of the workshops depends on the right mix of speakers and participating attendees; having, for instance, a cross-domain audience helps overcome the barriers between domains. Although we cannot prove it, we sense that in the thirty months since ERPANET started, the level of curation and preservation awareness of audiences at ERPANET events appears to have increased with each successive event.

Few practising librarians, archivists, museum professionals, and commercial records or digital object managers possess a sufficiently substantial body of knowledge about the technology and management issues to deal effectively with digital preservation. In an effort to eliminate that knowledge gap, ERPANET runs developmental seminars, which are two-day events focusing on policy and technical preservation issues. So far we have run seminars on OAIS (Copenhagen, November 2002), Policies for Digital Preservation (Fontainebleau, January 2003), Web Archiving (Kerkira, May 2003), Metadata (Marburg, September 2003), Appraisal of Scientific Data (Lisbon, December 2003), File Formats for Preservation (Vienna, May 2004), and Persistent Identifiers (Cork, June 2004). The quality of the training materials used in these seminars is of paramount importance if the participants are to be able to transfer the skills and knowledge they acquire to their home institution. We quickly discovered that the best experts have not always proved to be the best designers of training materials or the best communicators. As with the workshops, after the event we make available the seminar presentations and a report examining the key issues discussed.

The planned binary distinction between thematic workshops and training seminars has not always been maintained, reflecting, in part, the lack of stable knowledge in the preservation arena that can form the basis of training events focused on skills development. This has meant that presentations, discussions, and interactions at seminars create new knowledge in much the same way that similar activities at workshops do; this contrasts with the more traditional interpretation of a training event.

ERPANET is constantly looking for new ways to develop communities of practice and to promote developmental discussions. Just over a month after the Marburg training seminar in October 2003, we ran an online chat session that enabled speakers and participants to discuss issues further. We found this a very successful additional activity and consequently have conducted similar online events after other seminars and workshops. During the lifetime of the project the mechanisms that we can use to communicate with the community have changed, and the project has attempted to respond to emerging opportunities.

ERPANET has sponsored a number of events not envisaged at the outset of the project. In August 2002 we hosted a session at IFLA2002 (Glasgow) on Information for Life: Digital Preservation and Access; in April 2003, in conjunction with the Nationaal Archief van Nederland, we ran a workshop on Archives Online: moving archives into the digital era; and in September 2003 we hosted a three-day workshop with the DSpace Consortium at Glasgow on Implementing DSpace. In addition, we have already agreed to run ERPANET@SOA2004: Introducing ERPANET's Knowledge Base as a event before the 2004 Society of Archivists Conference in Glasgow; this event will introduce attendees to the knowledge we have developed through the seminars and workshops and will form the basis of a series of additional planned outreach activities. We are currently considering adding two further events to our pre-November 2004 work programme. One would focus on the preservation of contemporary digital art. The second would respond to the recognition by many participants in ERPANET events that as digital technologies are impacting on digital object creation and curation, these technologies are leading to fundamental changes in how archivists and records managers work. In an effort to develop a better understanding of this change, ERPANET will host a workshop in Bern (October 2004) to bring together experts to explore how the challenges posed by digital objects are reshaping the archival and records management disciplines. In the first thirty months of the project, the members of the ERPANET team have delivered more than seventy presentations at other events about digital preservation and the work of ERPANET.

Case Studies

Despite recognition in the library, archive, and records management communities that the survival of digital information requires action, casual discussion with professionals from these communities indicates that calls for action have so far not resulted in effective and commonly adopted digital preservation strategies. In an effort to understand what organisations are doing to promote preservation of their digital materials, the team at ERPANET have been conducting case studies. These studies are helping us to:

• build a picture of preservation methods and match the methods against institutional context to underpin guidance on good practice;
• accumulate and make accessible information about the experiences of different digital resource managing communities;
• identify issues that could benefit from new research;
• enable cross-sectoral comparisons of strategies and practices;
• provide sources of experience and methods to facilitate the development of assessment guidance;
• create material for training seminars and workshops; and,
• identify new areas for research.

Various sectors have been selected to represent a wide scope of information production and digital preservation activity. As of June 2004 we had substantive contact with over 400 institutions based in Europe. Less than 20% of these contacts have actually resulted in completed case studies. Completed case studies include, among many others, British Telecom Archives, Centraal Bureau voor de Statistiek, Council of Europe (Strasbourg), Deutsche Presse-Agentur, European Patent Office, Europol, International Labour Organisation (Geneva), Meteorological Office (UK), Universal Postal Union (Bern). To build as comprehensive and representative a picture as possible, we are interviewing archivists/records managers, information systems or technology managers, and business managers in organisations that agree to act as case study targets. This broad assessment of awareness and activity in organisations is providing us with detailed information about the extent of knowledge and practice in organisations, is giving us an indication of where ownership for the problem lies within organisations, and is offering us material to determine where digital preservation activity is likely to be promoted within organisations. In conducting our interviews we are examining:

• how to increase perception and awareness of risk associated with information loss;
• how digital preservation affects the organisation;
• what actions institutions are taking to prevent data loss;
• how institutions monitor these activities; and,
• what mechanisms institutions have put in place to enable them to define their digital preservation needs.

One of the immediate outcomes of our work is the case study questionnaire available at the ERPANET website that we hope others conducting similar studies might find a useful starting point [8]. The preliminary results of the case studies were outlined in a paper delivered at IFLA2003 (Berlin) [9] and were examined in more detail in a paper delivered in Ottawa in September of 2003 at the Canadian Conservation Institute Symposium 2003 on Preservation of Electronic Records: New Knowledge and Decision-Making [10]. Further publications arising from this research are in the pipeline. The research is providing new material that is enabling us to understand the digital curation and preservation problem better. The completed case studies are available on the ERPANET website for use by researchers and professionals looking for practical experience and approaches. During the summer of 2004, we will be supplementing these detailed studies with the results of a broader survey of current knowledge and practice in digital preservation across European companies.

Evaluation, Dissemination, and Collaboration

As is by now evident, ERPANET supports a diverse range of activities. We are concerned to ensure that they inter-relate closely and that they produce results and resources of the highest quality. To this end we have put in place quality control procedures covering the assessments we produce of literature and projects, the case studies we conduct, and the seminars and workshops we run. For seminars and workshops we seek user feedback to ensure that our events are of the highest standard and to enable the community to offer us information about their needs [11]. The reports of user evaluations of the individual seminars and workshops are available on the ERPANET website. Across the events held between June 2002 and June 2004, ninety percent of the attendees who completed an evaluation form reported that the events met their expectations. In the case of both types of ERPANET event, more than ninety percent of those attending reported they would be likely to attend another one [12]. Those who have attended multiple events—and there have been quite a few individuals who have done so—have helped to bring an intellectual continuity to our activities. We seek participants' opinions on such other topics as the documentation (e.g., briefing document, handouts), speakers, organisation, structure, and focus of the events. Across the board all aspects of the events are highly rated, but early on we did learn some valuable lessons from our evaluations. For instance at the first two events we did not provide copies of the overheads used by speakers to participants before the presentations. Participants asked for these in advance and now, wherever possible, we provide them. We have extended the amount of time at the workshops devoted to breakout discussion sessions in response to participant requests. In addition to evaluating each event, we are about to begin an impact assessment of the first seven events (Toledo, Urbino, Copenhagen, Fontainebleau, Bern, Kerkira, and Marburg) in an attempt to measure the benefits that attendance at those events had on the working practices of individuals and their organisations. This is being conducted as an independent research project led by a specialist in evaluation and impact assessments from the University of Toronto.

Our website is the primary ERPANET dissemination mechanism. It is from this service that users gain access to our products from our assessments to our case studies. As not everyone who is interested in attending our seminars and workshops can afford either the time or travel and accommodation costs, we encourage all speakers at our events to provide ERPANET with their presentation materials for distribution both at the events and on our website. Only one speaker has formally declined to do so. Wherever possible, we record the presentations. For some events we have made the audio recordings available online alongside the presentation slides. Reports of all our seminars and workshops are available online, and in 2004 we will be publishing the first nine of these reports in print. The papers delivered at the Marburg seminar on metadata will be published jointly with the Archivschule (Marburg) and those given at the Rome workshop on digital repositories will be published jointly with the Accademia dei Lincei. Papers from several of the other workshops have appeared in Archivi e Computer, including one by the directrice des Archives de France, Martine de Boisdeffre [13].

As a collaborative activity ERPANET depends upon synergy with other organisations and on engaging professionals. So far more than twenty European and international organisations have worked at corporate level with ERPANET to make our training seminars and thematic workshops viable and successful. In addition to the four core partners, these have included: The Biblioteca Nacional España, Caja Castilla la Mancha, The Royal Library (Denmark), Archives de France - Le Centre des Archives Contemporaines (Fontainebleau), Trivadis AG, Ionian University (Kerkira, Greece), Archivschule (Marburg), Accademia dei Lincei, Ministero per l'universitá e la ricerca (Italy), DSPACE Consortium (including the Cambridge University and Library and MIT), National Library of Portugal, CODATA, the Stadsarchief Antwerpen, Österreichische Nationalbibliothek (Vienna), University College Cork, Digital Curation Centre (UK), and Minerva. By June 2004 over one hundred professionals from nearly as many organisations contributed lectures and led discussions at these events.

The Research Libraries Group (Mountain View, CA) and the National Archives Records Administration (NARA) in Washington DC have helped build our programmes and have generously shared their expertise [14]. In mid-2003 we concluded a partnership agreement with the Preserving Access to Digital Information (PADI) initiative at the National Library of Australia and are in the process of redesigning our literature and project assessment databases so that we can automate the exchange commentaries with PADI and thereby eliminate duplication of effort. We are working to conclude other partnerships as well.

Collaborative Contributions

Preservation is a problem domain that demands collaborative action. ERPANET has worked to ensure it provides its expertise, and that of its community, to the work of other projects as well. For example, we played a role in the drafting of the Spanish Resolution on preserving tomorrow's memory - preserving digital content for future generations and ensuring its wider adoption [15]. To promote this resolution ERPANET contributed to the work led by the Italian Presidency of the European Union and the Italian Ministry of Culture, in cooperation with the European Commission and MINERVA [16], to develop an international conference on digital memory preservation, Futuro delle Memorie Digitali e Patrimonio Culturale (16-17 October 2003, Firenze). A major outcome of that meeting was the 'Firenze Agenda', which lays the foundation for ensuring that the Spanish Council Resolution is effectively implemented. The initiative is supported by an Experts' Workgroup on Digital Preservation [17] that engages many of the key players in Europe.

The European Commission's FP5 funded digital library activity DELOS and the National Science Foundation (NSF) established working groups to study the digital library landscape and define research agendas surrounding a number of core challenges facing digital libraries [18]. Among the eight working groups, one examined Digital Archiving and Preservation. Colleagues from ERPANET contributed to the development of its report, Invest to Save: Report and Recommendations of the NSF-DELOS Working Group on Digital Archiving and Preservation (2003) [19]. This report charts the research needed if digital libraries are to have access to the experience, methodologies, practices, and technologies necessary to ensure the long-term accessibility and usability of the digital assets they will hold. Members of the ERPANET consortium have joined DELOS2 [20] to press forward research in the area of digital preservation and digital libraries. The DELOS2 Preservation Task Group during its first eighteen months will investigate the development of a digital preservation testbed, the design, deployment and management of digital repositories, file formats and their classification, the viability of documentation of functionality and behaviour of digital objects, and the creation of preservation functionality modelling tools. These are all issues that ERPANET's case studies had found were core technical problem domains.

Conclusion

ERPANET's online resources, events, and contributions to research, awareness raising, dissemination, and action provides a substantial resource for the community. We have helped promote change in the community, but we now face our most substantial challenge. As our first phase of funding is coming to an end, we are examining business scenarios to find one that will give ERPANET long-term viability. We will decide in July 2004 how, if at all, ERPANET will be continued beyond November 2004. While we hope that ERPANET will continue playing a part in satisfying the need for reliable knowledge about digital preservation methods and practices and high quality research, the ERPANET team recognise that many more organisations of this kind are needed if we are to develop effective preservation capabilities across public, commercial, and consumer sectors.

Notes and References

1. HATII (Humanities Advanced Technology and Information Institute) home page, <http://www.hatii.arts.gla.ac.uk>.

2. ERPANET (Electronic Resource Preservation and Access Network) home page, <http://www.erpanet.org>.

3. Lisbon Council (2000), 'Presidency Conclusions: Lisbon European Council', <http://europa.eu.int/european_council/conclusions/index_en.htm> (from where there is a link). These conclusions were reinforced by the 'eEurope 2002 Action Plan' for which clear methods for measuring the success of its targets were established. See the 'eEurope 2002 Final Report' (2003), <http://europa.eu.int/eur-lex/en/com/cnc/2003/com2003_0066en01.pdf>.

4. It is worthy of note that the distributed nature of the partnership has raised many management and communication challenges, but it has also created a team rich in a diversity of skills.

5. 'Erpanet Announces ErpaEPRINTS at IFLA2003 Berlin', IFLA Express, issue 7 (8 August 2003), page 8. <http://www.ifla.org/IV/ifla69/IE7e.pdf>.

6. ERPAePRINTS, <http://eprints.erpanet.org>.

7. Subject, of course, to guarantees that the posting will not raise any intellectual property rights obstacles.

8. ERPANET Case Study Questions, <http://www.erpanet.org/php/studies/docs/ERPANET%20Case%20Study%
20Questions%20final.pdf>.

9. Programme and proceedings, IFLA World Library and Information Congress, Berlin 2003, <http://www.ifla.org/IV/ifla69/prog03.htm>.

10. S Ross, M Greenan, and P McKinney, in press 2004, 'Digital Preservation Strategies: The Initial Outcomes of the ERPANET Case Studies' in the Preservation of Electronic Records: New Knowledge and Decision-making, (Ottawa, Canadian Conservation Institute).

11. Summary evaluations for each event can be found at the ERPANET website.

12. For example in our cumulative evaluation report we noted that 'Ninety-five per cent of the participants returning survey forms indicated that they will definitely or probably attend another ERPANET event eighty per cent and fifteen per cent respectively.' See the Cumulative Feedback Report - ERPANET Workshops and Seminars (June 2004) on the ERPANET Website

13. 'Une introduction pour les politiques de la preservation électronique', Archivi & Computer, XIII Fascicolo 1-2/03.

14. Staff at the Electronic Records Archives Program at the National Archives and Records Administration (NARA) in Washington DC has been especially supportive of ERPANET's activities.

15. Spanish Presidency Resolution on Digital Preservation, Council Resolution of 25 June 2002 on preserving tomorrow's memory—preserving digital content for future generations, <http://europa.eu.int/eur-lex/pri/en/oj/dat/2002/c_162/c_16220020706en00040005.pdf>.

16. Minerva home page, <http://www.minervaeurope.org/>.

17. ERPANET Preservation Workgroup home page, <http://www.erpanet.org/www/workgroup/main.htm>.

18. DELOS Network of Excellence on Digital Libraries home page, <http://delos-noe.iei.pi.cnr.it/activities/internationalforum/Joint-WGs/joint-wgs.html>.

19. Invest to Save: Report and Recommendations of the NSF-DELOS Working Group on Digital Archiving and Preservation (2003) <http://delos-noe.iei.pi.cnr.it/activities/internationalforum/Joint-WGs/
digitalarchiving/Digitalarchiving.pdf>.

20. DELOS, <http://www.delos.info>.

21. DLM Forum'99 - European citizens and electronic information: the memory of the Information Society, <http://europa.eu.int/historical_archives/dlm_forum/doc/dlm-message-to-industry-en.pdf>.

22. Proceedings of the DLM-Forum 2002 @ccess and preservation of electronic information: best practices and solutions, Barcelona, 6-8 May 2002, (Luxembourg: Office for Official Publications of the European Communities, 2002), <http://europa.eu.int/historical_archives/dlm_forum/doc/dlm-proceed2002.pdf>.

Top | Contents
Search | Author Index | Title Index | Back Issues
Commentary | Next article
Home | E-mail the Editor

D-Lib Magazine Access Terms and Conditions

doi:10.1045/july2004-ross