Sir

I have added Peter A. Lawrence's Commentary on ”The politics of publication“ (Nature 422, 259–261; 2003) and David Colquhoun's incisive Correspondence response ”Challenging the tyranny of impact factors“ (Nature 423, 479; 2003) to a very small collection of items on this subject which are de rigueur reading for everyone in my group, no matter how senior or copiously cited. The two others are ”Is science losing its objectivity?“ (J. Ziman, Nature 382, 751–754; 1996) and ”The martial art of scientific publication“ (E. N. Parker, Eos 78, 393–395; 1997).

Asking the young scientists for whom I have responsibility to read these articles is the best way I can think of to meet Lawrence's question of what we can do about the situation that he so clearly diagnosed. The audit culture and the bragging attitude it sustains are advancing on all fronts; accountants and administrators armed with networked computers are the downside of the advances that have transformed the way in which we are able to analyse scientific data. There seems to be no appreciation of the once unchallenged argument that the scientific goose that lays the golden eggs needs some tall green grass, privacy and free choice of nesting sites. It does not respond well to weekly or quarterly requests for deliverables and activity reports.

I fear that electronic publishing, while conferring obvious benefits such as connectivity and searchability of the literature, may exacerbate the situation. I have had a paper to review which cited about 30 references, hardly any of them more than 10 years old. When I commented that this was unfair to earlier workers, I received the response that older references were not available on the e-accessible database at the author's institution!

This anecdote typifies a detectable trend and, I believe, is one that risks causing serious erosion of the knowledge base — not to mention the fact that the pioneers in a subject often stated the principles and problems with far greater clarity than many modern authors do.

One suggestion would be for a team of senior scientists, or perhaps national academies around the globe, to produce a handbook on responsibilities, rights and privileges as regards publication. It would be for all scientists but specifically aimed at entrants to the research enterprise. Even if one could not get agreement from those who successfully operate the present system to their advantage, at least the debate, if public, would raise consciousness.