F1000, who publish the extremely pricey “Faculty of 1000″ reviews of scientific literature are now throwing their hat in the open-access publication ring with F1000 Research.
They are planning to do post-publication review,which is the current darling of the open-access advocates, as it pretty much guarantees that anything someone wants to publish will appear—the only barrier being the price the author has to pay to get it published. F1000 has been doing post-publication review (of work from any scientific journal) for years, so they have some ideas about how that should go.
I’ve written some Faculty of 1000 reviews, and I found their editorial constraints rather limiting. They never wanted any negative comments about a paper, just superlatives. Even the very good papers I read generally had a few flaws, and they kept wanting to remove mention of those parts from my reviews. I’ve pretty much given up doing reviews for Faculty of 1000 for several reasons:
- Their subscription prices are so high that the University of California Library has come close to unsubscribing a couple of times, and I don’t want to be providing free content to a company then charging so much for the content.
- Their editorial policy of “superlatives only” does not match my more balanced (or perhaps just more negative) reviewing style.
- I’ve lost interest in the field for which the originally recruited me. I find it very difficult these days to read most papers on protein structure prediction and related fields. Mostly people are rehashing ideas that have been around for a decade or more, rarely improving on them. There are probably only a handful of good papers a year in the field, and I don’t have the patience to find them. (That is the job of F1000 reviewers, and if I were eagerly reading dozens of papers I’d be glad to share the occasional gem, but I’m not eagerly reading papers, so I almost never encounter the good ones.)
Given F1000′s history, I’m worried that the new F1000 Research will be very expensive and will quickly become a repository for trash papers. In a standard open-access journal that problem can be avoided with strong editorial control (PLoS Computational Biology generally publishes decent papers, for example), but with only post-publication review, the venue is likely to get clogged with stuff too bad to be published elsewhere.
Who is going to want to do post-publication review of a stream of junk? It is hard enough reading the mediocre papers that have already been through peer review already—reading the ones that even the authors think would fail peer review would be excruciating. If no one competent is reading and critiquing the papers, then the junk will not be filtered out and the whole process will degrade into a vanity press. There are already dozens of vanity open-access journals that will print any sort of s**t if the author pays the fee (even going through a pretense of peer reviewing), and I did not see anything in the F1000 Research announcement that suggested that they would avoid this trap.
Note that what F1000 is proposing is very different from the open access in arXiv, the (mostly) pre-publication archive heavily used by the physics community. “Open access to 731,853 e-prints in Physics, Mathematics, Computer Science, Quantitative Biology, Quantitative Finance and Statistics.” The arXiv site does not provide discussion forums, but many researchers put papers into arXiv for pre-publication discussion before submitting them to peer-reviewed journals. There are enough respected scientists doing this that people do regularly look through arXiv for interesting material in their field. Because depositing a paper in arXiv gets it read by a few people, but does not count as publication for promotion and tenure, there is little incentive to clog arXiv with junk.
People do sometimes stick papers in there to establish that they got an idea first, even if the paper needs some work still before it is publishable, and sometimes the early drafts of a paper have flaws that are revealed before publication, but most of the papers in arXiv are published in peer-reviewed journals eventually.
Also, note that the fields most rife with sloppy research and outright fraud are not covered by arXiv (medical research seems to have the lion’s share of scientific fraud, probably because of the amount of money involved). Although there are some crackpot papers in arXiv, for the most part it is real science. I believe that F1000 wants to cover a wider swath, including fields in which fraud and crackpot theorists are a more common problem. Relying on post-publication review seems a risky endeavor.
Disclaimer: as an unfunded researcher, I have mixed feelings about open-access publication. I like getting papers for free, and I like the idea of distributing scientific papers as freely and cheaply as possible. But most of the business models call for the author to pay for publication, which is fine for those who have grants that can pay publication charges, but deadly for those of us without funding. Open-access publishing may, in fact, restrict publishing to a smaller group of authors than the traditional publication model—those with money. Requiring those with grants to publish in open-access venues makes some sense to me as a taxpayer, but if it results in the loss of subscriber-pays journals, many researchers will be excluded from publication, and everyone will need to spend even more time writing grant proposals instead of doing research.
What I’d like to see are more things like arXiv, which provide free access funded neither by the authors nor the readers, but by organizations with an interest in free distribution of scientific literature (granting agencies, governments, and research libraries, for example). I doubt that F1000, as a for-profit company, plans anything like that.
- The Faculty of 1000 (johncarlosbaez.wordpress.com)
- F1000 launches fast, open science publishing for biology and medicine (blogs.nature.com)
- ArXiv submission? (rrresearch.fieldofscience.com)
- Faster-than-light neutrino update: What’s going on behind the scenes? (boingboing.net)
- On Elsevier (michaelnielsen.org)