UK risks 'losing science data'

Lynne Brindley, chief executive of the British Library, sought the support of the UK House of Commons Select Committee for Science and Technology yesterday (April 21) for a £12 million, 2-year investment at the library, to create a long-term national depository for digital scientific information and publications.

The depository is "absolutely critically to underpin science," said Brindley. "Without this, we will lose increasing amounts of scientific data, and it will mean the UK will not be world class in this area."

In a briefing document, the British Library said a large amount of scientific information is now being published exclusively in digital formats. "Much of this data is published informally and is inherently ephemeral. Unless archives begin to systematically identify, capture, and map this data, it will simply disappear."

The lack of a public record is also inhibiting the development of digital publications, said Brindley.

"The reason we wish to develop that expertise and keep it in-house [in the British Library] is that it is so strategically important for the nation and indeed for the future security of science. Indeed, some of the challenges of taking in this material, in all different formats, keeping it, providing access to it, and keeping it in the long term - by which I mean hundreds of years - are substantial."

This is not only a UK problem, said Brindley, and the British Library is working with the Library of Congress and other major research libraries; "but we are actually technically building that infrastructure."

The data will be kept on terabyte storage devices, she said. "We have archives stored securely in multiple locations, we have contracts for people to test out the robustness of our networking, we have very substantial recovery testing procedures, and contracts for recovery."

The process is much harder than with print publications, Brindley said. "And it's a much more active process, but we have to do it... We need to accelerate the building of this infrastructure over the next 2 to 3 years, and that's why we've put in a very substantial bid to the UK [2004] spending review for about £12 million."

This will be the "record of science, and indeed the intellectual memory of the nation, for the long term," said Brindley.

Peter Fox, University Librarian at Cambridge University, told the select committee that he echoed Brindley's sentiments. "There's potentially a gap growing between what is being produced and what is being archived, and we need to move quickly to ensure that gap is plugged, and for that we need support from the government."

Evan Harris, member of Parliament (MP) for Oxford West and Abingdon, asked Fox if there is a problem with losing access to the digital archive of a journal when a subscription is cancelled. "This is one of the disincentives to moving towards an electronic-only approach... You lose access to everything you've paid for," said Fox.

According to Brindley, however, "the situation is very messy - different publishers have different policies on supplying CD copies of the back issues that have been subscribed to... We need a national infrastructure to ensure those journals are always available."

Witnesses to the select committee also heard strong complaints about the widely varying contractual arrangements made by publishers, restricting access to digital versions of publications.

Ian Gibson, chairman of the committee and MP for Norwich North, read out a submission from the University of East Anglia: "Not only is the university restricted in giving access to its neighboring research, professional, and educational concerns, but also in our regional role as a major source of scientific information for the public. This goes against the government's desire to make science and its workings more open and available to the public. In hard copy, you have equal access provided you understand it; online presupposes privileged access."

"That's what we've been saying," said witness Frederick Friend, consultant to UK academia's Joint Information Systems Committee. "The answer to my mind is open-access publication... I'd urge the committee to recommend to the government that in any publicly funded research, the articles based upon that research be freely accessible over the Internet."

This week's hearing was the third of four being convened by the select committee on the subject of scientific publications. Earlier meetings heard evidence from publishers. The fourth session will be on May 5.