Preprints – What’s the worst that could happen?

Early career scientists are laying the building blocks of their career. Giving away your best ideas for free and everyone to read can seem risky. Pre-prints, on the surface, may look just like that, but are they?

Pre-print servers like EarthArXiv are sprouting in many scientific disciplines. They proclaim the benefits of research being cited more over time and being accessible. But pre-prints are neither peer-reviewed nor do they have an impact factor. As a PhD student, like myself, that has to submit at least three peer-reviewed journal papers this seems backwards. Obviously, getting more citations out of one article is excellent. Making the research accessible to anyone who would like to read it, also increases reach. But if it were this utopia of free reach, surely everyone would do it, right?

I had conversations with several mid-stage or late-stage scientists. The responses are mixed. People are wary that high-impact journals will look for novelty content and reject pre-prints due to this fact that the pre-printed article might not sell new issues. This is less risky for researchers that have an established body of work. For an EC researcher, one rejection on the basis of an earlier decision can have a higher relative impact. The difference between zero and one published paper is relatively higher than the difference between nine and ten papers. The promise of increased citation can seem like a siren call.

Essentially, it comes down to risk-benefit analysis, so let’s do that!

The Risks

Generally, preprint servers are very new in geosciences, with two general geo preprint servers (EsSOAr, EarthArxiv) and the paleontology-based PaleoArxiv, all opened in 2017. These follow in the tradition of the long-standing ArXiv from the physical sciences, founded in 1991. In physics and machine learning, preprints are extremely common and serve as the primary venue of public dissemination of research. Nevertheless, we don’t know how these will be accepted in Earth sciences, so here’s a list of concerns:

  • Publication is rejected, due to prior publication
  • Publication gets scooped
  • Open Access to paper may attract criticism
  • Giving away research for free lowers the perceived value of the publication
  • Early access to publication hinders exploration of subsequent ideas

Rejection

Generally, most large publishers accept pre-printed articles, as a form of scientific communication, similarly to conference publications. However, many journals in geoscience are powered by professional societies that may oppose pre-prints. So realistically, until further notice, some journals may, in fact, reject publication after pre-printing. It is, therefore, essential to consult with the journal of choice. In my field, a publication with the Society of Geophysicists is not possible after pre-printing (at the time of writing), whereas the American Geophysical Union is the driving force behind EsSOAr. Essentially, that means publication should be possible, but if you have a specific target journal, you may ask them for a change in policy or may have to refrain from preprinting / publishing with that publisher. Rejection of publication, therefore, can be mitigated beforehand. Additionally, one should consider that when a journal considers their value proposition so little that a pre-print takes significantly from their post-print, whether this is a venue worth publishing with.

Scooping

Apparently, there are people out there that have a tough time developing their own ideas, while still trying to pass themselves off as scientists. Scouring pre-print servers would be a possibility to look for yet-to-be-published work and pass it off as original research. However, with the growing acceptance of pre-prints as a legitimate form of scientific communication and the added benefit of proper DOIs on most platforms, an outright scoop have devastating effects on the scooper. No one wants a redaction under their belt. Effectively, it seems that an open pre-print may even be a safer route than most conference submissions. With EAGE I know, a combination is possible, which is a great way to increase the visibility of budding ideas and attract collaborators.

Criticism

Fear of criticism for publishing on pre-print servers or criticism of the work is granted. Criticism of the work is original peer review, and EarthArxiv encourages dissemination of syn-review work with the comment system. Criticism for pre-printing can also happen. After all, it is some form of communication of non-peer-reviewed work. However, as acceptance grows, the weight of these arguments decreases. The European Research Council now accepts pre-prints as publication type in an applicants track record. The ERC steps into the rows of many other major funding agencies there. Once again, if you have a particular career in mind, or grants you want to align with perfectly. You should check with them beforehand, whether they accept or encourage pre-prints. However, it seems that pre-printing may be beneficial in creating a strong track-record for early career scientists, instead of hindering their advance.

Value

Humans have a giant set of biases. Some of these biases are concerned with the value of things. Some individuals see Gucci and Prada as more valuable than other brands. A Tesla seems more desirable than a Volkswagen in the eye of the masses. It is not a mere coincidence that these luxurious items are much more expensive than the increase in quality would suggest. Judging the quality of scientific publications is very difficult, especially before reading. We rely on metrics like impact factors, glamorous institutions, and citation count. I have never heard of the consideration of price, though. The discussion has always been the other way around, whether we can afford expensive subscriptions. Therefore, I believe, the perceived value will not decrease, while the reach of information may increase.

Loss of follow-up

Sharing ideas as early pre-prints increases citation count, I hear touted as one of the main benefits. That implies that derivative works or further developments flourish from early and open access. When someone sits on an idea a bit longer, that means a certain reluctance to pre-printing may exist. Exploring the concept further may be interesting. Especially in a PhD thesis setting, publishing several papers out of one central idea may be crucial to the success of the student. Realistically, incremental research is alive and well, especially due to the “publish or perish”-culture. This may be a problem, but it also implies that someone is able to take the pre-printed research and develop the next increment before the supposed domain-expert could. This sure is a risk, albeit a small one.

The benefits

Let’s move on from the risks to benefits in pre-prints for early career scientists.

Flowers in a row of testubes with different colored water decoration and science experiment concept

This was mentioned before, large funding agencies such as ERC and the US-based NHS accept pre-prints to allow scientists to “accelerate dissemination and enhance the rigor of their work”. Certainly, a reason to pre-print.

Increased citation count for early career scientists can be precious. Citation counts often grow proportionally to the time one has spent in a field. Pre-print servers can act as a central agency to collect scientific contributions. ArXiv, for example, is the one-stop shop for anything that pertains to innovation in high-energy physics and machine learning. In geoscience we have so many societies with differing policies and foci, a central source of information would indeed be valuable for research to be discovered. One or two additional citations may well be worth it in the beginnings of a career.

The open review can be a benefit of pre-prints. For early-career scientists, this is both a learning opportunity, as well as a way to ensure high quality of early contributions to science. Someone like me that has to publish to obtain a PhD also benefits from the added rigour to the publication. This may well defuse some of the harder questions in the defence, which I sure appreciate.

Networking as an early advocate of open science. Once again, depending on the career prospects, positioning oneself as an open scientist may be beneficial in getting their name out. If you are looking for some particular industry sponsors, this may not be for you. However, it might increase credibility in your work that you are willing to share it at an early syn-review stage.

Concluding, it seems that there are few downsides to pre-printing and some significant upsides, such as increased citation count. The disadvantages can be mitigated and aren’t career-limiting, whereas the upsides can significantly accelerate a young scientist’s career. It seems like the hardest part for young geoscientists is deciding whether to pre-print on EsSOAr or EarthArXiv.

The following two tabs change content below.
... is a geophysicist by heart. He works at the intersection of machine learning and geoscience. He is the founder of The Way of the Geophysicist and a deep learning enthusiast. Writing mostly about computational geoscience and interesting bits and pieces relevant to post-grad life.

Latest posts by Jesper Dramsch (see all)

Posted in Easy Reading, Science and tagged , , , , , .