ÖйúAƬ

Rising journal costs restrict access to scientific research. Martyn Bull reports on a campaign to get institutions to set up free internet archives

If the purpose of scientific publishing is to increase access to knowledge to enable further advances in science and technology, then researchers are not doing much to help. Too often they care more about the impact of publishing in a high-status journal than how many people might be able to read it.

That was the conclusion of the recent review of scientific publishing by the House of Commons Science and Technology Select Committee. MPs also think it is odd that academics do not take more interest in what happens to their research after publication, considering that authors publish to enhance their reputation and that of their institutions.

The report advocates radical change in the scientific publishing process to prevent escalating costs and restrictive publisher agreements choking access to scientific research. It recommends that authors disseminate their research for free on the internet by storing articles in institutional archives, as well as backing a switch from ¡°subscriber pays¡± to ¡°author pays¡± publishing.

Author apathy can be traced to a lack of awareness of the threats to journal access because the issues are left mainly to research libraries. The threats include increasing journal costs and decreasing library budgets, restrictive copyright agreements and the move to digital access. Consortium of University Research Libraries (Curl) figures show that the average price of journals rose by 58 per cent between 1998 and 2003, while the UK retail price index increased 11 per cent over the same period.

Institutional archives offer immediate access to the research of individuals and their institutions. Costs are minimal compared with the annual spend on scholarly journals. The software to set up an archive can be downloaded for free, and technical maintenance is minimal. The greatest costs lie in promoting and regulating a service.

Articles in archives are available free on the web and are maintained by the institution. Authors can submit their research findings to the archive either before or after publication. Quality ¡°kitemarks¡± attached to the articles signal if the work has appeared in a peer-reviewed journal. Archives also allow institutions to better reflect the scope and impact of their research, as well as helping to build brand identity.

There is considerable activity in this area in Britain. The Securing a Hybrid Environment for Research Preservation and Access (Sherpa) project, led by Nottingham University and funded by the Joint Information Systems Committee and Curl, aims to set up 13 institutional open-access archives. It is investigating the key issues in creating and maintaining collections, including intellectual property rights, quality control, collection development policies, business models and scholarly communication cultures.

UK archives vary in size. Some contain a handful of publications, whereas the Council for the Central Laboratories of the Research Councils has more than 48,000 articles.

Linking archives and providing meaningful search capabilities is a long way off. There is little point in readers having access to hundreds of papers unless they know what to look for. Unless archives are kept up to date and include contact details for the authors, they will not be meaningful. Some believe that archives need to move from list-based database searches to become a more user-friendly experience.

Canada has joined the UK, the US, Australia, India and Norway in recommending that researchers download their work into institutional archives themselves.