Open access is ruined by perverse incentives
This summer I had couple of experiences which ruined my opinion of open access. More than ten years ago I have felt that openness is crucial for future fundamental developments, but now I clearly see perverse practices even in top journals and publishers.
SciPost experience
As an author, my paper [1] got rejected from SciPost. The funny thing was that reviewing took unreasonably long time (more than eight months), yet still was of a rather poor quality (despite SciPost being high impact and otherwise well regarded journal). I blame most of the delays on the editor (it took months for them to respond to some of my inquiries), while in their emails editor has blamed me for suggesting uncooperative reviewers. I am glad to help whenever I can, and I try to be fair when suggesting reviewers, but picking reviewers is not the authors' job, is it?
First round of reviews was reasonable. While the second round was quite an absurd experience. It contained a single review, which is uncommon in comparison to other higher tier journals. Furthermore, that single review was written in less than 15 minutes (based on the timeline visible in the SciPost editorial system). As a reviewer, it usually takes me at least couple of hours to read a manuscript before making my mind up, and starting to write a review. So, I guess it should be no surprise that review written in 15 minutes was a complete nonsense. At the time of receiving this review, I have decided against responding immediately and instead decided to wait for a second review. Yet, it never came. Instead, in few weeks time have received an email that my time for response was up (I was never notified to respond as soon as possible), and that the editor decided to reject the paper.
Yet the funniest thing of all is that SciPost supposedly emphasizes openness, but it extends only to accepted papers. So my terrible experience is conveniently swept under the rug. Only from the entries in arXiv it is visible that I tried and failed to publish in "prestigious" SciPost journal. The submission history is now only visible to me and the editors of SciPost. Though I can access only the history of the revised, not the original, submission. In my humble opinion things openness is useful only when it is not selective (obviously, within reasonable bounds), otherwise it is just a distraction.
So, SciPost = bad? I do not know. I have talked to couple of my colleagues about this experience. Some of them were shocked, as they had excellent experience with SciPost. While others had similarly terrible experience with SciPost.
MDPI experience
On the other hand, I had terrible experience as a reviewer at MDPI. In June I got a couple invitations to review papers, which seemingly fall within my expertise. The problem with econophysics and sociophysics, is that everyone thinks that social sciences are easy. But they are not, nor econophysics or sociophysics are social sciences. Hence, there are lots of low quality submissions and even accepted papers.
Those five papers I got to review in June were quite terrible. Some of them were ridiculously poorly written. Others presented poorly designed or conceptually wrong research. Some were terrible from both perspectives. So, I have written five quick negative reviews, outlining the major issues and providing quick overview of minor issues. Obviously, I have suggested rejecting all of them.
Funny thing is that I have received three or four of them for the second round of review. All of them are now published no matter what I did in the second round of reviews. I have written new negative reviews, I also tried reasoning with the editors, I have tried pointing out trivial mathematical mistakes as a sign that authors do not understand their own materials. It seems that all my comments were ignored.
Furthermore, I was notified about the papers being published despite ticking the box that I do not want to be informed about their status. Also in some instances I have directly told the editor that I am not interested in the manuscripts.
So, MDPI = bad? I can't say for sure. Yet I intend to limit my reviewing activity for MDPI. Namely, if the requested review is not for a special issue headed by someone I know and respect, or if the manuscript in question is not immediately relevant to me, I will not agree to review.
What I learned from this
I guess my main take away from these experiences with open access publishing is that it has slowly become perverted. It was a great initiative, but paying for publishing articles creates perverse incentives for publishers (publish more, earn more) and scientists (pay money, get your nonsense published, keep your position). SciPost tried to address this by not taking money for publishing, but their supposedly high standards stink of "scientific feudalism" (high impact scientists propelling their "underlings" towards success) and "cult of the hot".
References
- A. Kononovicius, B. Kaulakys. 1/f noise in semiconductors arising from the heterogeneous detrapping process of individual charge carriers. Journal of Statistical Mechanics 2024: 113201 (2024). doi: 10.1088/1742-5468/ad890b. arXiv:2306.07009 [math.PR].
