Dear Dr. Gibbs,
I enjoyed your essay and agreed with many of the points: peer review as currently practiced could be greatly improved; the academic "game" is rigged against those without academic affiliations or with academic affiliations that are not from R1 universities; the idea of open access is good, but the exact way in which it will be funded is still unclear (this I think was the point of your Elsevier story on page 5 where Elsevier's stock price went up *after* open access since now they were getting paid by the government plus still getting library subscription fees (it was the government that paid right? not the individual authors).
My only question/quibble is in the larger picture will having a better peer review system really have that big an impact on the progress of science, steering humanity to a "good" state, etc.? Aren't there more important factors to focus on? I just say this to play devil's advocate since as a theoretical physicist my main output in science is through what I publish so this is an extremely important point to me. But I am also reminded of a student-athlete at my university who once told me that the best thing to advance the goals of the university would be if there was *more* support for athletics on campus. This kind of statement may seem strange unless you spent some time at an American university which not only are academic centers, but also amateur athletic training institutions.
Anyway I like the idea you mention of having peer review be public. Something like post the paper to some electronic archive and then people would post comments/critics of a given work. You might still need to keep the names of the people leaving critics anonymous since people, even those trained in the scientific method, often fall in love with their own ideas and have a hard time hearing even well intended negative things about a given piece of work. It doesn't help that often the criticism is in a tone that can be unnecessarily negative. Also the more people that give some comment on a piece of work the better. Something some journals have done is to list the editor in charge of the particular article along with the final published version. The idea is this then forces the editor to take some responsibility for the quality of those articles accepted. It give another layer of oversight into the quality of the paper since both the referee(s) and the editor have given the paper the OK and the editor even puts their name with the paper. Of course the reverse does not happen -- no one publishes the rejected articles with the name of the editor that rejected them. So there is some asymmetry on this point.
Finally, you might find interesting a recent study of NIH R01 grants published in
N. Danthi, et al. "Percentile Ranking and Citation Impact of a Large Cohort of NHLBI-Funded Cardiovascular R01 Grants", Circulation Research, 114, 302656 (2014)
this study found that there was virtually no difference in the scientific impact of those grants which were ranked Category 1 (excellent grants according to the grant reviewers and which therefore got funded), Category 2 (very good grants according to the grant reviewers but which just missed the funding mark) and Category 3 (fair or marginal grants as judged by the reviewers which also did not get funded). Grants in all three categories published about the same number of articles and received the same number of citations. This is not necessarily the best measure of scientific worth, but it is the metric by which NIH says it judges the grants.
Anyway and interesting and thought provoking essay.
Best,
Doug