Holding the past in our digital future

Ask academics and librarians to list their most useful and frequently consulted reference sources, and a good many will now include, somewhere near the top, the accumulation of emails and discussion-group postings saved in files on their PCs. These informal electronic information stores of previously answered questions and online debates with never-met colleagues are increasingly matched by vast structured archives that also exist only in digital form. In the past year, the Library of Congress has, for example, established a freely available collection of archived digital documents covering the terrorist attacks on the US on September 11 2001 ( http://web. archive.org/collections/sep11.html ). The collection contains more than 5 terabytes of data - a terabyte being approximately one thousand billion bytes, or 1,000 gigabytes. A gigabyte is roughly the amount of data required to encode a human gene sequence; or, looking at it another way, a terabyte is more or less equivalent to 600 million pages of text from a standard paperback.

Not all of the Library of Congress archive is made up of text and, of course, not all of it is of more than ephemeral relevance. But it is inconceivable that future generations of historians, researchers and the casually curious will be able to gain adequate insight into early 21st-century culture and society without being able to read such digital archives.

As yet, rather more thought has gone into constructing digital collections, either from specially commissioned electronic documents or from digitising printed texts and records, than in ensuring that they can continue to be read, even in the relatively short term. But for some years Maggie Jones and Neil Beagrie have been central players in the moves to address the issues of preserving electronic information sources. Two workshops held at Warwick University in 1995 and 1999 helped to open up the debate on the development of digital preservation strategies. Preservation Management of Digital Materials: A Handbook is the best example yet of turning discussion into action.

The handbook, sponsored by Resource: The Council for Museums, Archives and Libraries, is the result of a continued collaboration between the Joint Information Systems Committee of the UK Higher and Further Education Funding Councils and the National Preservation Office based at the British Library. It is to be commended for managing to draw together expertise from these various bodies into clearly argued statements of recommended practice, without feeling as if it were written by a committee.

In matters of broad principle, there is overlap in developing programs for preserving digital resources with traditional strategies for paper-based materials, not least in recognising that preservation concerns need to be addressed by those creating the documents as well as by those charged with looking after and providing access to them. But digital documents are dependent on machines and have to cope with the dangers of technological obsolescence - hardware and software that is no longer supported, file formats that can no longer be recognised - in addition to the problems of physical deterioration.

The threats to digital documents are drastically more pressing. If ways of maintaining access to an electronic information resource are not investigated at the earliest stage, then it is in danger of becoming unusable in a very short space of time: "Timeframes during which action needs to be taken is (sic) measured in a few years, perhaps only two to five, as opposed to decades or even centuries we associate with the preservation of traditional materials."

Jones and Beagrie's approach to helping avert losses is multi-pronged. A strategic overview of the whole range of concerns of digital preservation is aimed at all those who ought to be acquiring a general awareness of the subject, and assumes (and requires) only a novice level of expertise. More targeted strategies, standards and best-practice guidelines will be of use to those with technical and non-technical responsibilities for maintaining and providing access to electronic information sources, as well as to creators and publishers of digital resources. Summaries, glossaries and bibliographies are all as clearly organised as one would expect.

The issues of preserving digital materials may not yet be as publicly emotive as the debates on preserving traditional print archives of books, periodicals and newspapers, but they are just as vital. This handbook does an excellent job of highlighting them. An electronic version, including additional supporting materials, is available on the web ( www.jisc.ac.uk ). Updates and further case studies are promised; it is from these that the momentum for looking after the digital future will best be maintained.

Christopher Phipps is reader services librarian, London Library.