The issue of publications of science manuscripts is reaching breaking point. Breaking in the sense of tearing down the enthusiasm of young investigators, the patience of seasoned ones and generating a lot of debate at institutional level. A few days ago, a survey by Nature News showed that more than 35% report taking 1-2 years to publish a manuscript and 20% 2-5 years (https://www.nature.com/news/does-it-take-too-long-to-publish-research-1.19320). What the bottom line of the article hid is that the time does not mean time sitting in the same journal –though some times it does- but time from submission to the first journal. For a more informed case study of the topic you should look at several posts in the always excellent S. Royle’s blog (https://quantixed.wordpress.com/); in the context of this post, I particularly like this one https://quantixed.wordpress.com/2014/04/15/some-things-last-a-long-time/. The point is that in science, but particularly in the biological sciences, something very fundamental has changed. Scientific publishing is an evolving enterprise but over the last twenty years it has moved, imperceptibly, from a medium scientists use to report their findings to a consumer good that uses scientists to develop its business model. It could be argued that this should not be a blanket statement but, actually it is pretty much that, particularly in the non-academic tier. Compound this with a sociological trend of too many scientists, disenchantment with career prospects and the emergent importance of media and this explains why scientific publishing has become a source of employment and a business that, as has been repeated many a time, uses scientists at all ends –from the bench to the editorial desk- of its fabric.
The problem is that the essence of the enterprise: finding-report-publication, has not evolved and what we have is a system that, basically, cannot cope with its essential fuel: science. Too much is being produced, all needs to be peer reviewed and peer review has evolved into an interesting sociological contraption that at some point should be the subject of some PhD thesis. The aim of peer review in its current incarnation is not to improve the work (this is the gimmick) but to ensure that a paper takes the longest possible time to be published and that its cost escalates to the point that what is considered good research is, basically, expensive research. And so it is against this backdrop (a much simplified version of it) that last week a meeting took place at HHMI headquarters with the cunning acronym of ASAPbio (Accelerating Science and Publication in biology: https://asapbio.org/). The meeting gathered an interesting mixture of senior scientists, administrators and editors with the aim of discussing preprints and preprint servers in Biology. For those of you who have not heard of preprints (and surprisingly there are many out there who haven’t and people at ASAPbio who acknowledged that they did not know they existed before the meeting!), a preprint -in the context of science communication- is a manuscript which has not been subject to external peer review and which is posted in a site for public viewing; it even has a Wikipedia page (https://en.wikipedia.org/wiki/Preprint), whose first words are “Publication of manuscripts in a peer-reviewed journal often takes weeks, months or even years from the time of initial submission,…..â€, it says all.
ASAPbio was broadcast live on the internet and you could follow it on Twitter and social media; some of us did and you can see the videos now at the site (https://asapbio.org/). It was interesting. The gathering was representative of what we could call the establishment of science, principally the US establishment, and gave careful consideration to the culture of preprints and its future. Perhaps the most surprising thing to come out of it is that we have to fight a fear to preprints as forms of unvalidated science, preprint postings as a danger to “true research†and, surprisingly, some of these voices from qualified scientists (yes, the ones who have created and control the system we enjoy –or endure, as you choose-). There were exceptions amidst the old guard. The objection raised by those who worried was that preprints are not ‘controlled’, are not reviewed (forgetting really what lies at the heart of a manuscript at the time of submission), vetted. One had the impression that preprints are some kind of modern samizdats (in case you have not heard this word: https://en.wikipedia.org/wiki/Samizdat) forgetting that samizdats were a good thing in the cold war era. In fact, as a response to ASAPbio some went as far as posting tweets wondering whether what is in preprints is true or safe for consumption!. The answer to this question is simple: read the paper. Furthermore, physicists have been using preprint servers since 1991, particularly in the form of their very successful arXiv (https://arxiv.org/) home to over one million e-prints on all aspects of physical and mathematical sciences. Paul Ginsparg, founder of arXiv, participated at the meeting providing details of its history, value, mechanics and validation of and making clear that in many ways, preprints (e-prints at arXiv) are the bread and butter of the physics community. He also reminded (in some cases revealed) to the audience that arXiv introduced a section on quantitative biology which is proving popular. Along these lines a few years ago under the initiative of John Inglis, CSH started a parallel server, bioRxiv, which aimed at catering to do for Biology what arXiv does for Physics and Mathematics. There are other preprint servers of different flavours but bioRxiv aimed from the beginning to be arXiv like i.e. community oriented, user friendly, data rich. As it was reported at the meeting (and you can read here https://asapbio.org/biorxiv) bioRxiv is succeeding and that many of us –who have been using bioRxiv from the beginning- believe that it is a seed for and leads the way to the future of bioscience publishing.
The problem, for now, is that Biology lags culturally and practically behind Physics and this is reflected in many practical aspects, not only the publication ethos. Biology, as a science, hiding behind mindless collection of data has become more a culture of glamour than of content, of haves and have nots in which a good idea or a good point is likely to go amiss and where many bad ones get too much fleeting attention by virtue of where they are published. The question people most often ask you when you try to tell them about your work is not WHAT is it about, but WHERE have you sent the manuscript or, often, how big is your group. The science of the bioscience is a means to an end and so, it is not surprising that publishers have latched on to it and make the most of it. I don’t want to be misunderstood, there is a lot of good science around, it is just that not all that appears as good science is good science (though all data are useful). And thus it emerges the current tangled web of bioscience publishing. You, like me and many, have felt the dismay of starting the path of trying to publish something in what some people would call a ‘decent journal’. By now it is clear that the demands of peer review are not very different in most of those ‘decent journals’, that the difference between them is that in NCS you get rejections with arrogant and patronizing demands and messages from the editors, whereas in other journals you just get a directive to answer the often arrogant and patronising comments of the reviewers. I shall not bore you with details, we all have our stories. As has been said before, when people finally publish, there is a sigh of relief more than of elation. There are exceptions and I can only cite, within the realm of what I know, EMBO Press and The Company of Biologists. I wished I could cite eLife and PLoS but I am afraid that each in their own ways have failed to live up to their expectations (PLoS ONE has done a lot of good but has become slow and cumbersome as reviewers ask for more and more work, and PLoS Biology has become a place where people go when their papers have been rejected from NCS; in the case of eLife, with much that is good, it is unfortunate that so much power and resources are being channelled to beat NCS at their own game rather than in being really creative –for those who praise their excellent reviewing methods of eLife, you should know that EMBO journal had been using them for several years before eLife.
So, in terms of publication time and effort, things are getting worst all the time (see above). Enter preprints, that apparently dangerous and unsafe way of disseminating science. I must say that to label a preprint as dangerous or scientifically unsafe is to miss the point that publications in NCS can be worst because, as has been pointed out many times having been ‘peered reviewed’ they show the cracks of the system and the pitfalls of the glamour world those magazines promote: Arsenic life, the STAP problem or cold fusion come to mind and, in the case of STAP, don’t forget it led to the humiliation of a world renowned institute and the suicide of a much respected scientist with little consequence to the journal which published the manuscript in the first place. To say that preprints are dangerous is to overlook where the real danger lies and where the pursuit of glamour over science has led and leads. For those who say that preprints will be honey for garbage I shall say that having followed bioRxiv from the beginning, one thing has surprised me: there has been almost zero crank science. BioRxiv has had a steady increase in submissions and all of them -up to now it is possible to follow most of them- are sound science, no less controversial than what is published after peer review. If anything, one of the valuable aspects of preprints –if properly used- is that they represent raw science, before it is tampered with by editors and other scientists, who often have more competition in mind than the advancement of he science. It is clear that preprints in Biology are here to stay and that in the future will be the best way to present your science. I suggest that you go to the ASAPbio site and read through some of the views of the meeting and even if you have time, watch some of the discussions. More importantly, make sure that you use preprints to disseminate your science.
There were two other important issues discussed at the meeting. The first one was whether preprints should be allowed as an evaluation of scientific research. In theory, and in the light of DORA, this is a no-brainer but the discussion is important as the people gathered at ASAPbio represented a significant element of the establishment. Despite some concerns, it was also reassuring that, for the most part, people saw this in a positive light and many pledged not only to accept preprints (in preprint servers) as evidence of scientific quality, but to encourage their citation in job and fellowship applications. Many young scientists will appreciate this and I suspect that we shall see more of this in the future. The second issue was the mechanics and detail of such a cultural change. You cannot move this idea forward without attention to the organizational mechanics of the server, something arXiv and now bioRxiv know well. There were several references to this and it is important that in the enthusiasm of the move, we do not forget this important aspect of the enterprise.
ASAPbio was led by Ron Vale, whose excellent https://biorxiv.org/content/early/2015/07/11/022368.article-metrics served as a fuse for the meeting. There were well known people in the audience, like Michale Eisen, who have been doing a very open campaign for open science. However, while Eisen is good as a ramming bat against the establishment, we should not forget the many who in a quiet way have been adding to the momentum of preprints by convincing students and postdocs of the value for their careers and science of preprints servers. We need the chief scientists and administrators on board but, more than that we need the inertia created by people submitting their papers to preprint servers commenting on those servers (and don’t forget that you can comment in PubMed through PubMed Commons). It is clear when you come to think of it, that not only there is nothing to be lost by using e-print/preprint servers, but much –all- to be gained, particularly if you are a young Scientist. What is good for physicists cannot be bad for biologists.
ASAPbio was a formal recognition that something is changing and hopefully a formal start of a much needed move into new directions. I can see the worries of so called major publishers about the watershed that is coming, they might lose their stranglehold on the Science they own and shape, but a date comes to mind, 1789, and a voice that reportedly said “let’em eat briocheâ€. There will be no blood here but much needed change and a recovery of what is scientists’ privilege, the control over their work. I get the impression that something is moving. I sense not the dawn of a new era but the dusk of a short lived one based on much that is wrong and that we need to turn around. As it is being said these days, make sure that your next paper goes to a preprint server as it goes, or before it goes, to peer review. One senses that this is the way to solve, not only the emerging tyranny of journals but also peer review itself.
“many pledged not only to accept preprints (in preprint servers) as evidence of scientific quality, but to encourage their citation in job and fellowship applications.”
The key point here is not to accept a listing of a preprint as evidence of scientific quality (that gets back to the awful JIF issue) but that by submitting a preprint, the submitter receives a DOI that allows reviewers, search committees, etc. access to that manuscript so they can better judge recent productivity by reading the manuscript. I know (via Twitter) that many younger scientists and researchers in smaller labs worry that preprints may work against them as it increases the possibility of being scooped by bigger labs. It certainly may reset the clock such that people will not speak at meetings or show new poster data without having the preprint “in hand” for fear of this, but that is surely better than waiting until you get the “in press” notification 6 months to 2 years after submission.
My feeling is that younger scientists will realise the advantages of preprints and publish sections of theses and smaller advances that they cannot pursue as a means to better demonstrate their accomplishments. After all, a paper is a relatively crude way to summarize what is usually years of work. That’s where the real possibilities will emerge and such uses will challenge conventional journals as they are not engineered for individual scientist empowerment.