Sites for citations?

While Tim Birkhead (Columnist, 4 January) is right in that mis-citation is a serious problem, his focus only on the training of researchers downplays the importance of access to the literature in finding the appropriate citation. The ever increasing number of journals, coupled with the hugely above-inflation increases in journal subscription rates over the past 20 years, has meant that too many journals are not available in one's local library. A proper literature search to track down the three, four or more links from an individual's exposure to a concept to its origination requires access to each of the links in the chain of referencing. One must follow all the links to find where the precise idea referenced originated (for example, while Darwin first suggested sex selection as an evolutionary concept, there have been numerous variations and improvements on this basic concept, and the exact variant referenced should be acknowledged as well as, where appropriate, the original idea).

Luckily, there is a solution to the access problem, which is for all researchers to self-archive the final refereed version (the author copy is fine as the exact layout is not needed, just the final content) in their own institutional repository. As well as taking advantage of the reduced workload for researchers able to check online the papers they need without waiting for and paying for inter-library loans or expensive per-article charges on publishers' online databases, the cross-linking of citations and other value-added services that such repositories will allow to emerge will further improve the ability of scholars and scientists to correctly cite the work of those on whose shoulders they stand.

Andrew A. Adams,
School of Systems Engineering,
University of Reading.