by PhilipJ on 26 February 2006
A week or so ago, I quoted a letter to Nature written by Lindomar B. de Carvalho on worthless papers, and whether they were a bigger detriment to science than the high-profile scandals that plague our profession. He suggested we
[slow] down the paper-publishing machine by limiting the number of journals that publish original research, asking more peer-reviewers to read preprints and opening up preprint manuscripts for public discussion.
I admitted at the time that I thought the first suggestion was a great idea, but didn’t see the immediate effect the other two would have. Lindomar found my post and replied, but comments were automatically turned off for that story at the time, and so his reply might have gotten missed by those who were interested in his ideas. I’m reproducing it here:
if you have more people knowing what you do, the way you do, criticizing your work, etc. Soon or later you may be kicked out of the picture. In the other way around, if you just keep on submiting papers to journals of rank A to rank Z, until your paper get accepted! There will be no way to slow down the so called paper publishing machine.
So the idea of the two other options is to help clean up the â€œtrash scienceâ€?.
Thanks for the reply, Lindomar! Now I think I see where you were going—if all papers were first submitted as preprints (to, for example, the arXiv) and everyone had a chance to see your work, it would be harder to get things published to begin with. As it stands, a rejected paper can be submitted to the next best journal, and if it gets rejected again it can be submitted to the third best journal, etc, until it gets published somewhere. If the process were more transparent and the community at large had an opportunity to discuss your results and methods prior to publication (as opposed to a small handful of reviewers), the opinion of the community at large may help weed out results that are truly irrelevant. Perpetually attempting to publish substandard work would become harder if everyone had the opportunity to critique the work being done.
While I think it is a good idea, I see one big pitfall—how do we make the community’s response count when a paper gets submitted to a journal? The only clear method would be for the arXiv to become a publishing house itself, where the community response is the peer review. This doesn’t seem so far fetched an idea for the physics community*, but I can’t imagine all scientific disciplines moving to this kind of mechanism any time soon.
* I think the field of high energy physics already thinks of the arXiv as such, in a way.