Letters |
The Physiological Laboratory, University of Cambridge CB2 3EG, UK
E-mail: rhsc1{at}cam.ac.uk
Peer review is indeed ... a flawed process, full of easily identified defects with little evidence that it works, as Richard Smith concludes in his thoughtful article (April 2006 JRSM1)but even flawed refereeing is better than not being peer reviewed at all. Increasingly, prestigious journals put submissions through a preliminary triage before they are even allowed to be seen by real scientists. This pre-selection process aims to identify sexiness; however important a paper in terms of advancing its subject, if it is not regarded as sufficiently topical or is too intellectually demanding (perhaps it has some maths) it will be returned with a computer-generated letter to the effect that it would be better in a more specialized journal. Typically, a request for further information about exactly how the paper failed to meet the pre-selection criteria is met with further boilerplate generalities: the process is secret and unregulated.
Of course, journals such as Nature and Science are under pressure from the huge increase in submissions that has resulted from the whole-scale adoption of impact-factor bibliometrics in the research assistant excersise. But sooner or later scientists are going to ask whether it is worth wasting time on this demeaning and dispiriting ritual, and whether perhaps the internet is now grown up enough that we can cope with web publicationand dispense with the luxury of peer review. Once publication per se confers no particular prestige, the cancerous over-publication that afflicts us all will be stopped in its tracks. The metastatistics generated by Google and Amazon demonstrate how one can use the extraordinary power of the internet itself to create systems that prevent one drowning in unreliable information. Would a free-for-all be so very terrible?
Footnotes
Competing interests None declared.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||