FAQ

Below are a set of responses to FAQs intended as a community resource. These responses are a work in progress. Please leave your thoughts on the responses and suggestions for other FAQs in the comments below.

1. What are preprint servers?
2. Why do people use preprint servers?
3. What preprint servers are available to geneticists and biologists more broadly?
4. What are journal policies on preprint servers?
5. What are best practices for preprints?
6. How do citations work for preprints?
7. The journal I’m interested in has unclear/negative policies about preprint servers, what should I do?
8. Can I/should I publicize my preprint?
9. How should journalists view this site?
10. Do preprints represent prior publication?
11. Am I more likely to get “scooped” by placing my article on a preprint server?
12. When should I submit to a preprint server?
13. Without peer review, how do I know a paper isn’t flawed?
14. What if my list of target journals includes some that do not accept preprints?


1. What are preprint servers?
A preprint server is a centralized permanent database of manuscripts in a field. These manuscripts are submitted by researchers to archive their work with little to no editorial oversight, and include both peer-reviewed and submitted manuscripts.


2. Why do people use preprint servers?
There are a number of reasons people use preprint servers:

  1. Having a manuscript available allows you to get feedback from a broader pool of people prior to official publication than you otherwise would. See the comment threads here and here for examples.
  2. Preprints disseminate new results and ideas more quickly. For example, many preprints have been widely discussed on blogs; see examples here and here.
  3. In many fields that have adopted arXiv, posting a result to a preprint server is what establishes priority for a discovery. However, note that this depends on the field.
  4. The arXiv provides a repository of all versions of an article; in some cases it is interesting to see how a manuscript has evolved over time in response to comments.

See posts by Titus Brown, Graham Coop, Joe Pickrell, and this Nature News article for more.


3. What preprint servers are available to geneticists and biologists more broadly?
The majority of population genetics preprints for the moment are appearing in the quantitative biology section of arXiv and bioRxiv. Other options are figshare (which has a section for posting of complete manuscripts), peerJ, or the Social Science Research Network. Some older preprints are hosted on the now-closed preprint server Nature Proceedings.


4. What are journal policies on preprint servers?
You can search for journals preprint and copyright policies here at SHERPAS romeo. The online community has also been building a wikipedia page of (mostly biology) journal policies. A number of high-profile journals have made strong statements of support for preprint servers (e.g. Nature, PNAS). One thing to be careful of is that many journals that allow preprints do not allow the final published version to appear on preprint archives.


5. What are best practices for preprints?

  1. Make sure that your submitted preprint is machine readable. This allows for text-mining, allowing trends in the scientific literature to be tracked (e.g.) and broad analyses to be performed.
  2. If you submit tex files to sites like the arXiv, be aware that the tex is accessible. Be sure to clean up comments, etc., to avoid embarrassment.
  3. Update your preprint as it proceeds through rounds of peer review (if this is allowed by the journal where you have submitted). This avoids the confusion of older versions being cited/read.
  4. When the article is published by a journal, update the preprint to include a bibliographic reference and a link to the journal’s online publication via a DOI (“digital object identifier”) number (see here for instructions for arXiv). These steps will allow readers to track versions and will encourage journals to support the preprint system.
  5. While many journals stipulate that submitted manuscripts include double-spaced lines and figures separated from captions separated from text, we suspect that most readers and indeed reviewers prefer to read papers without such features. We suggest, if possible, submitting single-spaced papers, preferably with figures and captions embedded in the text.
  6. Make your submission to the preprint archive as complete as possible. Include the supplementary information or links to it as part of your submission. If you are submitting tex files to the arXiv, the arXiv tex processing will combine multiple tex files allowing you make a combined pdf of main text and supplement see the conversation here. Alternatively supplementary information and figures could also be uploaded to figshare.


6. How do citations work for preprints?
Many, but by no means all, journals allow the citation of preprints. You should consult the editorial staff if you are unsure, and encourage them to allow citations to preprints if they don’t already.

Google scholar keeps track of citations to/of preprint papers. Google scholar can then fairly seamlessly integrated with the citations of papers when the articles appear in press [you can merge entries if your paper changes title, see here].

See here for information on how to cite arXived articles.


7. The journal I’m interested in has unclear/negative policies about preprint servers, what should I do?
If you are unsure about a journal’s policy on preprint servers, you should contact the editor. Many editors/journals just haven’t had much experience with preprint servers. Some journals and editors are understandably hesitant about preprint servers for various reasons. In these cases we recommend that you write a polite email to the journal clarifying confusion over preprints and explaining the benefits. Also talk to others that make up that journal’s community, and get them to add their voices through emails or blog posts etc. Journals can and do change their policies once they realize that their community is in favor of preprints (e.g. ESA).


8. Can I/should I publicize my preprint?
This is somewhat of a grey area and depends on the journal you submit to. You are definitely free to point colleagues and members of your community towards your preprint. However, most journals ask that you do not directly contact the press over your article. This certainly means that you should not send out press releases or hold press conferences announcing preprints. However, what this means with respect to blogs and other social media is often unclear. A number of journals have policies about press and preprints, although they are still somewhat cryptic (e.g. Nature, PNAS). PLoS journals stand out as being refreshingly clear on this issue (e.g. PLoS Genetics).

We view the “Our paper” posts on Haldane’s sieve as seeking community discussion/engagement on a paper. We ask the authors of these posts to highlight areas/questions they want feedback on, and not to shy away from discussing technical details. Thus we see no difference between blogging about a paper here, and giving a talk at a conference [especially as many videos/slides are now online]. Authors should refrain from mentioning what journal the paper is under-review/will appear in. Before writing an “Our paper” post authors should carefully read the journal guidelines and consult with editorial staff if necessary.


9. How should journalists view this site?
We ask journalists to respect the community discussion format that Haldane’s Sieve represents. We hope you benefit from seeing a open discussion of papers, and that this discussion shapes the future reporting of articles. Journalists are obviously free to ask questions via the comments. However, we prefer that you directly contact the authors about papers that you have a professional interest in, and alert an author if you plan to quote from their posts or comments on this site.


10. Do preprints represent prior publication?
A natural concern of many is that publishing on a preprint archive represents prior publication. This is not usually a copyright matter but can violate a journal’s policy (the so-called Ingelfinger rule, not to be confused with Goldfinger). In the view of many journals, submission to preprint archives does not represent prior publication, but we advise you to check individual journal policies.


11. Am I more likely to get “scooped” by placing my article on a preprint server?
If you post a result on a pre-print server, it will likely be seen by other groups working on similar topics. One possibility is that these other groups could then rush their work through the formal publication system and be published “first”. The preprint will of course be publicly date-stamped as having appeared before the “scooping” article, and likely has a head-start in the review process. It is thus up to the community to determine whether a result posted to a preprint server is considered to have priority over one published in a peer-reviewed journal. In most communities that have adopted arXiv, a post to a preprint server is the moment at which a result is presented to the community, and thus is considered to have priority. However, this will depend on the attitude in your community. We encourage you to give fair credit to preprints that have shaped your understanding of a subject, just as you would do for published work, to help spread this attitude.


12. When should I submit to a preprint server?
People submit to preprint servers when they feel their results are mature enough that they would like to share them with the community; in general, this is at the same time that a paper is submitted for review. Early preprinting offers the maximum benefit, allowing you maximum time for feedback, however submitting at any point during the review cycle is also okay.


13. Without peer review, how do I know a paper isn’t flawed?
Posting to a preprint server is not a substitute for peer review. The goal of effective evaluation of research is separate from the goal of rapid distribution of research; preprint servers only address the latter. Papers posted to preprint servers are often not subject to peer review (at least before their initial posting) and so there is extra burden is on you, the reader, to determine whether you are capable of evaluating the claims in them.


14. What if my list of target journals includes some that do not accept preprints?
Unfortunately, submitting your work to a preprint server may preclude you from later sending it to any journal that does not accept preprints. One way around this is to send your work exclusively to journals that support preprints; in our view, the benefits of quickly disseminating and openly discussing new research may often outweigh the costs of somewhat limiting the range of target journals. Even if you are not willing to take this step, you can support preprint systems by contributing to the discussion in forums like Haldane’s Sieve and by asking journal editors to adopt favorable policies toward preprints.

Graham Coop, Bryan Howie, and Joe Pickrell

5 thoughts on “FAQ

  1. About point (11) — perhaps having some ambiguity about whether a preprint server counts as “priority” will help get over the silly idea that whoever publishes first gets all the credit, even if the work was done in parallel. A preprint server also serves as a good way to get a scooped paper out earlier, making it more obvious that the work really was independent.

  2. Pingback: All the cool kids are on arXiv and Haldane’s Sieve .. why you should be too

Leave a comment