Sir

Much nonsense has been written about the ‘E-Biomed’ initiative to launch a global website for much of the biomedical literature, announced by Harold Varmus, director of the US National Institutes of Health (NIH). If it were the case that the present system had served science well for 300 years, for example, the outrage over the present state of scientific publishing felt in some quarters would not exist (see Nature 398, 735 & Nature 399, 292; 1999).

Happily, dialogue has been under way worldwide for some time, involving people from various disciplines, to investigate how electronic publishing can reach the broadest possible audience. Some new model will emerge from this process. Harmonization of goals and means is difficult, and benefits greatly from the involvement of a broad pool of perspectives and contributions. In this setting, the Varmus initiative is welcome, as are the critical evaluations to which the proposal will be subject.

Some of the valid concerns are the preservation and organization of materials; cataloguing; provenance (whether to peer review); copyright; and economic impact (including access in less developed countries). To advance the ‘E-Biomed’ proposal, I suggest that a moderated listserver is set up to initiate documentation and development of the larger community of interest. The NIH would be an appropriate host. These groups should be encouraged to participate: authors and researchers; archivists and librarians; Internet specialists; publishers and professional organizations; and readers.

I am initially mistrustful of what appears to be (largely) an electronic replication of existing methods of publication. One possibility is that ‘E-Biomed’ could host papers vetted by ‘alternative’ editorial boards, whose composition would be publicly stated, and whose editorial manifesto would be published, but whose membership would not be controlled by a central governing board.