David Haynes and David Streatfield worked together on a consultation exercise to establish who ought to be responsible for long-term preservation of digital material. They are joint authors with Monica Blake and Tanya Jowett of a report to the Digital Archiving Working Group. This article is based on the recommendations arising from the eLib supporting study described in that report (1). This article appears in the Web, and not the print, version of Ariadne.
During the supporting study the authors consulted various stake-holders in the creation, distribution and use of digital materials. A number of recommendations arose from this study and are presented in this paper. These recommendations are not official policy and represent the authors’ views of one possible way forward for ensuring that potentially valued digital materials are preserved for future study and use.
The JISC-funded study on ‘Responsibility for long term preservation and access to digital materials’ extended beyond rights holders to include all stakeholders in the production, exploitation, distribution and preservation of digital materials (see Figure 1).
Figure 1 - Production of Electronic Publications
Much of the current debate on digital archiving was started by the Task Force on Digital Archiving in the USA, which recommended the development of a national system of digital archives to act as repositories for digital information. The Task Force felt that:
Without the operation of a formal certification program and a fail-safe mechanism, preservation of the nation's cultural heritage in digital form will likely be overly dependent on marketplace forces, which may value information for too short a period and without applying broader, public interest criteria.
The RLG/CPA Report (2) was subsequently analysed by a team at Loughborough University (3) which came up with eight prioritised actions, including:
· appointing a National Digital Preservation Officer,
· establishing a National Digital Preservation Body,
· devising guidelines on practice and a digital preservation policy,
The researchers appointed for the joint BLRIC NPO study conducted a 39 face-to-face and telephone interviews and conducted a series of four focus group meetings to identify the main issues and to develop some priorities for topics to be addressed in any future archiving policy for digital materials. The focus was on who should be responsible for long-term preservation, how long the materials should be preserved for, who should have access to these materials, and who pays the cost of archiving.
Digital materials can include everything from electronic publications on CD-ROM to online databases and collections of experimental data in digital format. Although we excluded material that was published in printed form and subsequently digitised from the study, many of the same considerations apply to this material as well. For the purposes of this study the following definition of digital materials was used:
Digital archive material is text, data or documents originated in electronic form. This data may be made available to the public in electronic or printed form or it may result from publicly-funded research or be produced by public bodies. The scope does not cover conventional printed publication that re subsequently digitised for study or conservation. Compilations of data on CD-ROM, electronic databases and documents available via the Internet are included in this definition. Stationary and moving images and sound recordings held on digital media are also included in the scope of this definition.
Material which is suitable for archiving comes from a variety of sources including commercial publishers, the academic community, and government departments and agencies. Each of these groups has its own priorities (making a profit, being accountable to the public, dissemination of research results widely, keeping the nation’s intellectual heritage) and is therefore unlikely to subscribe to a common archiving policy. A co-ordinating body vested with the authority to formulate policy on digital archiving, advise government and carry out legislative or statutory requirements would be able to implement a consistent archiving policy.
The remit of the body would be to co-ordinate digital archiving activities in the public domain. A suggested title is the National Office of Digital Archiving or NODA (the name by which we will refer to this proposed body from now on.
NODA would provide a forum to represent the interests and views of the different stakeholders including rights holders such as authors of experimental data and publishers.
In our view the co-ordinating role should be separated from the archiving role which would be delegated to specialist agencies. The role of NODA would be to develop appropriate standards of service and to arbitrate where there are disputes about who is responsible for which materials. NODA would also provide a focus for liaison with digital archiving bodies in other countries and international agencies.
By setting archiving standards and selecting appropriate bodies to provide the archiving services, NODA will be taking on a policing role. It would be answerable to the government through its sponsoring department.
Digital materials are very diverse in their format and scope and can range from sound recordings through to experimental raw data in electronic form. Even text-based materials can range from html files with embedded Java scripts used for Web sites to Word processed documents incorporating graphics and spreadsheet data. Specialist skills are required to handle different types of material. Rather than centralising this activity we envisage a distributed archive, with the responsibility for archiving different materials being vested with specific bodies. Some of these may be public bodies such as the National Sound Archive, the Public Record Office, or the British Library for instance. Other areas could be handled by commercial organisations with the appropriate resources and expertise, on a contract basis, such as publishers, data storage and recovery companies and private picture libraries.
Some material of particular national interest could become the responsibility of the National Library of Scotland or the National Library of Wales.
This approach would necessitate the establishment of a national register of archived digital material. The starting point for this would be an audit of existing digital archives. Production of detailed catalogue records would be the responsibility of the individual archiving agencies, using agreed standards. The feasibility of using a common descriptor such as Digital Object Identifiers (DOI) should be explored.
Material should be divided into specific categories such as:
NODA would be responsible for the development of guidelines for retention and preservation of digital materials so that individual items can be identified and managed effectively.
There is on-going debate about the relative merits of preserving digital materials in their original formats or of converting them to one of a few well-supported formats. It is probably not possible to make hard and fast rules about preservation or conservation of materials in the long-term. In practice some materials may be kept both in its original format so that the future researchers have a better appreciation of all the proprietary features and attributes. In such cases conservation in a standard format would probably also be advisable with a migration policy to ensure future readability.
The majority of those consulted during the project felt that material selected for preservation should be kept for ever. Selection would probably be based on content rather than form, although some exemplars of a particular format may be preserved for future historians. Electronic publications could be subjected to similar criteria used for printed documents on legal deposit. Public records in electronic form fall under the public records Acts and are within the domain of the Public Record Office. Sampling may be appropriate for some categories of material.
Although some put forward the idea of periodic reviews of archived material we believe that this is incompatible with the idea of permanent retention of selected material. One of the problems of periodic reviews is that priorities change and selection criteria could change to such a degree that material of value to future users is lost.
Public funding is probably the only way to ensure continuity of archives. This could be through a variety of public bodies including the legal deposit libraries (if their remit is extended to cover electronic documents), research councils, and the Public Record Office. Recovering the full economic costs of archiving by charging users is unlikely to be viable, because the charges would probably be prohibitive and this would deter users from consulting historical digital materials. This in turn would reduce the number of users sharing the costs, raising the cost to individual users.
A National Office for Digital Archiving represents one possible way of carrying forward the agenda on digital archiving. The problem of preserving (or at least conserving) digital materials is growing and a coherent, co-ordinated national strategy needs to be put in place. Already potentially valuable material has been lost.
The next stage in this process should be to conduct a feasibility study for the development of a national digital archiving policy. This study would need to explore the main options that are available and provide some costings for these alternatives. This would then form the basis of a business case for the formation of NODA.
Material on this page is copyright Ariadne/original authors. This article last updated/links checked on 16-May-1998