One of the first and most acute causes of culture shock in physicists who have fallen among biologists is the relative role played by preprint servers in the two sciences. The arxiv has been a standard part of physics for years; a place where preliminary results can be displayed, critiqued and refined. In contrast biology preprint servers like biorxiv have only recently begun to attract attention and widespread use. For those who don’t yet know about preprints, this perspective is a quite excellent introduction. I have some specific comments that follow below, and I would appreciate the author saying something about them and maybe changing the MS (although it’s not necessary in all cases by any means).
The MS is very strong on educating the reader about preprints and their advantages for an individual researcher. But much less so when it comes to the advantages to the field in general. I think this is important, because the value to the individual and the field are entwined with how preprints are viewed, and the respect in which they are held, and this article is an opportunity to push the dial in the direction of more respect.
Physics and biology are very different enterprises. The position of arxiv in the former is not (only) due to a more open-minded attitude among physicists, but the history of the science and the esteem in which preprints are held. So in a highly competitive field such as Fermi gas microscopy, researchers upload their work as soon as possible and their priority is recognized. Similarly, dense theoretical work can be trialed online in front of a tiny community of peers who can suggest alternatives. These may be extremes/caricatures, but it is true that the important factor is the status of preprints in the community. And while preprints have all the advantages stated in this article, unless their status changes they will remain underused.
In a high status preprint world, as the MS states, publishing on a preprint server does get your paper out and could be seen as ‘scooping’ competitors. But we are a very long way from that at present. Why cite a preprint that came out well over a year ago, and is accepted but still waiting to see the light of day, when there’s an alternative in a glamor journal that came out a few months later? The preprint might get some nice comments on twitter, but nobody is going to see it as having priority.
Also in a high status preprint world, folks would be more likely to comment. At the moment, only a few preprints get many comments, which limits the value of the approach quite a lot. I know the MS says this, but it could say more about it as without this many of the advantages claimed for preprints fall away – so what if you can show your preliminary findings to the world if the world doesn’t give you feedback? I suspect it is also true that the numbers of comments are not only down to the inherent interest of the content, but also the profile of the authors. If Nick Loman (for example) puts something on twitter about his preprint, it will get looked at more. Similarly a Very Eminent Scientist* has tried to convince me that the large amount of attention and comments their preprints receive is wholly down to their extraordinary high quality, and has nothing to do with the fact that they are member of the National Academy with a very high profile. Such distortions will never go away, but might be less likely if preprints were consumed with the same interest as ‘proper’ journals.
So I urge a little light rewriting to raise these issues –perhaps explicitly recommending that funders mandate that papers should be put up as preprints at the same time they are initially submitted for publication. I have the following specific points I noted as I was reading.
*Not suggesting Nick is not a Very Eminent Scientist, but this person is even more so than he is.
Line 14 – these other things like blog posts seem interesting, but quite different from preprints. I’d suggest ditching them or making the focus clearer
64 repetitions of landscapes fractured etc
82 and beyond. Several issues are raised here that are never fully dealt with. The most important is DURC. I think you should say a bit more about how this can be handled responsibly, and the potential for preprints to be subject to minimal review to prevent some of these things.
186 Could preprints not attract adversarial comments as well? I am not sure scientists are fully free from the attributes of BTL commentators. What about the possibility of vexatious criticisms? This could be especially true for ‘political’ topics.
249 I wonder if preprints are more common in fields with a high proportion of physics types?
286 the public, or other scientists?
294 impact factors are ‘more traditional’ than altmetric, but more ‘controversial’? I have to say this made me giggle. I also confess I have no idea how to interpret these scores so this bit was usefult o me.
301 The comparisons with mBio are really neat, but there is a skew evident in the comparison of the means and median citations. It looks like the majority of preprints get very few citations, but a few get many more. The mBio citations look more normally distributed. Can you say more about this?
321 I think this is a bit misleading as review is still needed before publication. Sidestepping is not possible.
365 brining? ☺