The power of preprint

It's time to seriously consider faster, fairer models of publication

The Biologist 62(6) p7

How long did it take you to publish your last paper – from the moment you finished the draft manuscript until the day it appeared on the journal website? Three months? Six? A year? More than a year? It's unlikely to have been as 'quick' as three months and quite possible that it took more than a year to get into 'print'.

In the digital age, when publication can be instantaneous, the meaning of words such as 'quick' and 'print' is shifting. The customs of scholarly journals are an increasingly ill fit for the original purpose of the scientific paper: the rapid communication of new research.

If you really wanted to get your results out quickly, so that they could be dissected, discussed and taken forward, why didn't you upload your paper as a preprint?

Well, you will protest, journals won't take manuscripts that have already been published as preprints. And what about the important quality control mechanism of peer review? You might also admit that a large part of the delay in publication is because, for the sake of your career, you have to chase after the highest possible impact factors, which resulted in several rounds of rejection and resubmission as you worked your way down the hierarchy of prestige. These problems are familiar to all biologists. But the first two aren't actually problems and the last one exists only because we allow it to. Let me deal with each point in turn, briefly – I have written about them at length elsewhere[1].

Most major publishers accept manuscripts that have been uploaded as a preprint[2]. In many areas of physics, maths and computer science, the arXiv (arxiv.org) has long provided a mechanism for rapid dissemination of prepublication research reports. Those who use arXiv recognise that the preprint date establishes the first date of publication, and so determines who reported first, countering the notion that you risk being 'scooped' if you upload a preprint before your paper appears in a peer-reviewed journal.

Recently, preprints have come to life in the biosciences in the form of the bioRxiv (biorxiv.org) and innovations from open access journals such as PeerJ and F1000Research. Researchers such as the award winning cell biologist Ron Vale are increasingly seeing the advantages of this form of rapid communication[3].

Peer review is a useful tool for critiquing and sharpening research papers before they appear in print, but is subject to bias, poor at weeding out fraud and too often pressed into the service of maintaining journal brands. Preprints short circuit the delays due to journal processing but, crucially, do not cut out peer review. In fact, they have the potential to enhance it by allowing critique and comment by researchers from anywhere in the world. This is a form of publishing that harnesses the benefits of digitisation.

The problem of the impact factor chase is the most pernicious of the three spurious reasons that make people fearful of preprints. Its damaging effects on the speed of science and the processes of researcher evaluation have been highlighted by the San Francisco Declaration on Research Assessment[4].

We need to discover mechanisms that direct attention away from the name of the journals where we publish and back to the research itself. Preprints have an open, democratising quality that might just help us to do this.

So, for the sake of science and scientists, please think seriously about publishing your next manuscript as a preprint.

Stephen Curry is professor of structural biology at Imperial College London