Joan Comenge, Marie Curie Fellow in the group, has responded to my previous post. I reproduce below his comment as I hope it might trigger further discussion:
Very useful post in which (some of) the problems of academic publications are nicely pointed out. Agree 100 % in how stupid is a publication system in which the writer(s) and corrector(s) are paying to read their work. Even worst, as you said, public funds are wasted in paying private companies for whom we (public workers mostly) work for free. However, I think the most important problem regarding the lack of scientific value on the publications (lack of reproducibility, overstatements, etc.) is not due to that perverse system, but to the pressure researchers have to publish, and to a lack of ethics (maybe both are linked…).
First, you have to be the first in publishing something if you want some impact. No matter how good and reproducible is your work if someone has published something similar before (and probably because they didn’t care about reproducibility). Here, the problem is your personal ethics. No reviewer could ever check if the results you are presenting occurred just once because that day was raining.
Then, you have to have a number of publications if you want to get the next job. Normally, positions are 2-3 years, so you can imagine. You want to publish 5-6 papers. You wouldn’t mind to repeat the same work and change some word to make it slightly different. Again, I don’t think open access would solve this lack of ethics. Neither open reviews. If I have to publish 5-6 papers in 2 years, I would not spend my time reviewing 15-18 papers (I assume at least 3 reviews per paper would be needed and I assume everyone would like to publish 6 papers). Quality should come first than quantity, but who thinks this is happening? Not me. And I don’t think this would be solved with open access/ review as well. It could get even worst.
So what? I totally support open access (in fact, the main publication of my PhD is in PlosOne), but I think the scientific problems would be still the same if we don’t fight for a better evaluation system and, mainly, we are not conscious about our responsibility of not publishing shit.