From Demetris slides (here) I cite some from Schulman E.R.'s humorous paper:
"Scientific papers … are an important—though poorly understood—method of publication. They are important because without them scientists cannot get money from the government or from universities. They are poorly understood because they are not written very well.
and three statements which could be interpreted as truisms, but they are not (with my comments):
- Reading other good papers is much more useful than reading guidelines about how to write and publish papers (but you have to know the basics of writing a scientific paper, and someone that dissects it for you would help)
- Writing a good paper presupposes good understanding of the subject studied (In my view not entirely true ... you just need to have an intuition and pursue it. Having it all clear can take decades, and in the long range we are all dead. As D.K. says too, a good scientific paper is something in the flow of knowledge. Do not wait too much to write it)
- Publishing the paper presupposes good understanding of how the peer review process works (definitely true. Some opportunistic behaviour is necessary to survive).
To sum up, D.K. presentation is a "must read".
Looking at publishing from another side, papers are not all the outcomes of a research, but one among others, as are models, patents, books, data, and other stuff, especially if one considers that not everybody is a professor in real life (is here is sort of an equation ? - academy=publishing / not academy .. : do not care ?). Certainly writing a paper and going trough a review process can challenge your certainties, and refine your knowledge. IMHO doing research and writing are two different jobs, that meet only in good papers (and with the further note that unpublished research does not produced shared knowledge, and therefore science^2^2b).
Finishing with who started the thread, Luca B. also tries to point out (the papers in figure are from his own CV^1) that publications history, and peer review process can be weird and, at the light of history, even wrong (as a brief report of Keith Beven^3 included in D.K presentation also tells). However, being consistent, he could publish. Being consistent about consistency, besides being smart, is certainly a key of success.
^1 - I could also put the same comment on some of my papers,and, I could further include a paper that was never published after good reviews, just for the decision of the AE. Never mind: peer review is the worst form of selecting papers, except all the others form that have been tried from time to time (paraphrasing W. Churchill).
^2 - Something is still transmitted through oral communication though.
^2b - Science, however, is not just shared knowledge. For instance, it is substantiated by the possibility of checking (well, falsifying) assertions with experiments, yes, at the uncertain light of probability.
^3 - I admire K. Beven work, and I almost completely agree on what Jeff McDonnell (Google scholar here) says on him (reported in D.K slides). However, I am urged to remind, without offending anyone, that there is a certain difference between Saul Bellow and Stephen King. The first has certainly less readers than the first, but ...