Preprints seem to be having a moment. Last week, the registration agency CrossRef announced that they will soon allow members to assign DOIs (digital object identifiers) to preprints, just as they do for published articles. In making this change, CrossRef is acknowledging that preprints are an important part of the scholarly publishing ecosystem. In addition, back in March, a group of biologists made it into the New York Times for advocating for the use of preprints in their own discipline. At the same time, many academics still don’t know much about preprints or why they matter.
In general, a preprint is a piece of scholarship that has not yet been peer reviewed (and thus, hasn’t been published in a scholarly journal). It is related to a postprint, which has been peer reviewed, but has not been properly formatted by the publisher. Confusingly, the term preprint is sometimes also used to describe a postprint. Preprints have a long history, but people have been trying to collect and distribute them in a more formal way since the 1940s. The first online archive for preprints, arXiv, was launched in 1991 by Paul A. Ginsparg, a physicist at Los Alamos National Laboratory (Ginsparg is now a professor at Cornell University). Ginsparg hoped that arXiv (originally called xxx.lanl.gov) would help “level the research playing field,” by granting anyone with an internet connection access to the latest scholarship in high-energy physics, for free. He also knew it would help researchers get their work out into the world faster than ever before. Almost twenty-five years later, arXiv hosts over 1 million preprints from disciplines including mathematics, computer science and statistics. As New York University Professor of Physics David Hogg noted in a recent Wired article, “When I give seminars, I give the arXiv numbers for my papers. Why? Because I know that my arXiv papers are available to any audience member, no matter what their institutional affiliation or library support.” Thanks in part to the success of arXiv, scholars in other disciplines are now considering making drafts of their work public, including those in the humanities. In CORE, the Modern Language Association’s new digital repository, 25% of the articles are preprints or postprints.
So, why should academics, particularly those outside of the sciences, care about preprints? These days, more and more scholars are sharing copies of their work online (see our recent post on Academia.edu). Since most scholars do not own the copyright to their work, however, they may not have permission from the publisher to do so. One way to get around this is by sharing a preprint. While the vast majority of publishers will not allow a scholar to make the final version of their article (also known as the publisher’s version/PDF) freely available, they often allow the sharing of a preprint and/or postprint through an institutional repository or a personal website. According to SHERPA/RoMEO, a database of journal policies, 79% of publishers formerly allow for some kind of self-archiving.
It’s important to point out that not everyone in the academy agrees that the posting of preprints is a good idea. Some scholars worry that if they share their ideas too early, they might get stolen. Others correctly note that a preprint is not a substitute for a peer-reviewed journal article (which remains the gold standard for getting tenure). Finally, there are more general concerns about sharing work before it has been thoroughly vetted or revised. However, one recent study compared over 9,000 preprints from arXiv to their final published versions. The authors ultimately found that there were very few differences between the two versions.
Have you shared a preprint of your work online before? Why or why not?