A few days ago, fittingly in the context of Open Access week, we had an event to explain what are preprints and how they have the potential to change communication and career development in the biomedical sciences. You can follow the recording here: http://asapbio.org/event/preprints-biomedical-science-publication-in-the-era-of-twitter-facebook. The event counted on the participation of publishers, funders and users; a summary has been posted in The Node and I encourage you to look at it and contribute to the discussion. There is much to talk about in the wake of the event. Here I shall concentrate on a few issues in the context of publications which highlight the momentum for change.
Preprints are not peer-reviewed papers but importantly, they are nothing new as they have been tried and tested in the Physical sciences for over twenty years, with much success. Interestingly, at the time of the event EMBO J published an article from P. Ginsparg, the founder of arXiv, the Physics preprint server, calledÂ â€˜Preprint deja vueâ€™Â Â (http://onlinelibrary.wiley.com/doi/10.15252/embj.201695531/full); I recommend everybody to read it and reflect on what it says about the biomedical sciences for it shows us up (my reading, I hasten to say) as a small minded community, with a narrow notion of competitiveness and an overreliance on commercial publishing for the evaluation of our work.
Preprints and preprint servers are a rising culture and are emerging as a good first solution to some of the problems. But we have to get it right. There are two concerns that are often levelled to preprints: the worry about their lack of peer-review and the fear of being scooped. Both fail to recognize the essence of Science and the crisis of the peer-review process in the biomedical sciences. For example, on the first one, it would be foolish not to recognize that an important element fuelling the rise of the preprint culture is the degradation of the peer-review process which, up to now, has been the cornerstone of modern science. Nobody can deny that papers improve with peer-review but it is obvious that from a system of checks and balances and with the connivance of journals, the peer-review process has become â€“particularly at the high end of the market- a device to delay publications and, in the process, to give reviewers the power to determine the content of the work and, in some instances, use the anonymity of the process to produce unfair allegations and decisions. Much has been written about the anonymity (blind, double blind, open reviews) but it has always worried me that one of the arguments to preserve the status quo is said to be the protection of young PIs from retributions that established peers might launch in the face of the criticisms that might be levelled to them. How come we cannot own what we say and think? What this says really is that the reason for the anonymity is FEAR, a typical situation in totalitarian systems. How come fear is a justification for anonymity? It should give us pause for thought. Where have we taken our scientific culture: people afraid of signing what they believe in? It would be good if we could change the scientific culture, if we could encourage and practice more an open discussion of our work (which let us not forget, goes on in private, in journal clubs and cafeterias). Journals give you the option of commenting and discussing but only after publication. A preprint is there for discussion, so you can comment, openly, and influence its shape and help, rather than hinder, the authors.
But, in the end, we have to ask what is the purpose of peer review? It has been suggested that it is validation of the research and yet, if you ask (and we did at the event) how many people have failed to reproduce a published piece of work, the answer will be loud and clear: many, most. So, peer-review is not validation. It is a form of certification of the quality of work, like that of the rating agencies on credits and, as it was the case in the financial crisis, there are too many subprime papers AAA rated because they are bundled in HIF journals with a few quality ones. We have lost our bearings and it is unfortunate that we value our work not for its intrinsic merits but because of where it is published.
The issue of scooping cannot be separated from the strange contraption that the peer-review process has become. Here, again, we biologists have a very different understanding of a notion (scooping) that highlights our small mindedness. As Ginsparg puts it (see EMBO J above) â€œâ€œscoopingâ€ in the context of biology research appears to mean the race between laboratories working on overlappingâ€ and herein we highlight again that the form matters more than the content. There is little question that posting your work, whether in preprint or peer-review form, gives you priorityâ€¦ if there is anything to give priority forâ€¦â€¦â€¦â€¦ In Physics there is no question, again Ginsparg: â€œPosting work on arXiv gives authors a datestamped priority claim, which is accepted by the community, and gives immediate visibility to authorsâ€™ workâ€. I and many agree. The large number of papers with 0 citations in HIF journals is a sad comment on the huge amount of work that in the biomedical sciences goes into useful pieces of information but largely irrelevant pieces of science. How much money and pain would those authors have saved if they had posted their work in a preprint server! I shall leave you to take it from here but in my mind, there is little to fear about posting your work in a preprint server and much to be gained. In many ways, preprints are the â€œultimate open accessâ€. Preprints can create a more democratic, cost effective way of managing science and at the event it was excellent to see Journals promoting them and seeing their value.
There is much to work on to get the community to embrace preprints. We were lucky to have exceptional speakers at the event where the Wellcome Trust, as a funder, expressed their support for the culture and explained their own efforts towards it (Wellcome Open Research & the Open Science Prize). Interestingly, shortly afterwards we learnt that EMBO will allow for preprints to be cited as evidence of output (as long as they are accompanied by at least one peer-reviewed paper). I am aware of efforts along these lines in the US where many institutions already encourage applicants for jobs to cite preprints. This is good news because it does unshackle students and postdocs from the handicap of not being able to refer to their toilings when they are looking for fellowships and jobs. Preprints are here to stay. The reasons are many and being increasingly discussed and if you wanted a particularly one, this one was stated by Richard Farndale (Dpt of Biochemistry, University of Cambridge): preprints give us a way to let funders know that the work that they gave money to do, has been done. This simple statement alone, makes the point of what preprints are for and should lead funders not only to encourage preprints but to demand them. It should be the funders and not the journals, who decide whether the work has been done.
There is much being said and written about the current situation which really is an expression of movement. A particularly thoughtful piece by P. Walter and D. Mullins appeared recently in the ASCB: â€˜on publishing and the sneetches: a wake up call?â€™ (http://www.ascb.org/on-publishing-and-the-sneetches-a-wake-up-call-november-december-2016-newsletter/ ). Much to mull over here and many arguments for preprints. The article ends up on a note â€œThe end goal seems obvious: The knowledge that we produce in our publicly funded works belongs to humankind and must not be locked up behind pay-wallsâ€” newly submitted papers should be open-access and older ones open-archive. Our real challenge is to find the paths that get us there.Â But major change can happen, even if it seems impossible to imagine nowâ€ I say, let us use preprints. Not only use them but work, together, to shape the future of biomedical science communication.
Further readings on preprints from this blog: