Ad verba per numeros

Tuesday, January 26, 2016, 11:11 AM
A short entry to recycle a long email I've just sent to a colleague. The topic of discussion was about releasing preprints of papers that are meant to be published in journals.

I assume that some of you may be very well aware of all of this stuff so if you can tell the relation between sherpas, Romeo, preprints and postprints you may skip this post. If you cannot then go on reading.

Some researchers (particularly those still in graduate school) seem to think that once they have submitted the final version of their accepted paper and the copyright transfer are no longer owners of their work. Putting a copy of that shiny paper on a website is, of course, out of discussion.

That's pitiful because by putting their papers behind a paywall they are getting null visibility, recognition and, needless to say, citations.

I'm not going to discuss here about using citations (or downloads) as quality metrics but I assume you agree that writing something that is not going to be read is futile (and somewhat stupid).

So, you may just break the rules and simply publish that author's copy not for redistribution; needless to say I do not recommend that.

You can also pay a good money to keep your research open. I don't want to discuss Open Access here, however and, moreover, I don't have the money to pay for that.

Hence, what are the available options to keep your papers open, free (gratis) both for you and your readers, and legal?

You have two choices: preprints and postprings/author's versions.

In the first case you publish (in your website or in a preprint server) the draft you are about to send to a journal (or a conference). In the second case you publish your manuscript as accepted for publication, sometimes including modifications made following reviewers' comments, but without the journal format and prior to copy editing (if any).

Needless to say, you must be sure the journal(s) where you want to submit your paper accept that. Here is where Sherpa/ROMEO enters. There you can check if you can or cannot publish your manuscript as preprint/postprint.

(For conferences it is wise to check with the program chair before attempting to release the preprint.)

Before going on with the advantages (and caveats) of putting preprints you need to know a little about preprint servers. They are major repositories for that: preprints. They free you of the problems of handling a website and they are very well known so your work is quite likely to be read (or at least skimmed).

Given my STEM background I tend to use arXiv but there are others more suitable for other areas of knowledge (for instance, SSRN for Social Sciences).

Now, what are the good parts of publishing preprints, postprints and author's versions? (1) visibility for your work and you, (2) higher probabilities of being cited, (3) providing free access to research (that is usually funded with public funds).

What about the caveats?

First of all, if your preprint is eventually well known it is impossible for it to be blind-reviewed.

If you still believe in Santa and the Tooth Fairy that may be a problem for you. However, if you assume that most of the time reviewers are very aware of who is submitting the manuscript they are reviewing you should have no worries.

Of course, I know that some people are quite strict with this but it would be a matter of discussion about procedures with the Chief Editor or the Program Chair. I highly doubt that papers are rejected because of being available of preprints (well, at least in most of journals, I'm not talking about Nature or Science ;)

A second problem is that your preprint may be more cited that the eventual published paper; particularly if the draft had a different title. It is annoying but difficult to handle (although arXiv provides a way to indicate the DOI of the eventual publication).

From my point of view advantages of preprints/postprints outbalance the disadvantages and, thus, I highly recommend everybody to release their research as preprints, submit only to venues which are preprint-friendly and put author's versions as soon as the real paper is available in the publisher site.

As usual you can find me on Twitter if you want to further discuss this.