31

At our institute it is not common to upload a pre-print of your submission to arXiv. I got told by a friend that it is basically a no-brainer and there are virtually no downsides in doing so. I checked with the journal I want to publish in, they even allow submitting arXiv links, so I am safe from there.

Since nobody at my institute seems to upload their pre-prints, should I ask my advisor before doing so? He is also co-author of the paper.

My field of research is materials science (and to some extent more general physics).

  • 2
    "virtually no downsides" may not be true, otherwise you would probably not be asking this question. – Bitwise Jan 17 '17 at 9:30
  • 1
    The only downsides I know of are that some few publishers do not accept papers that have been previously published on arXiv and that you basically cannot ever completely remove it. But if you stick to journals that have less strict policies and are confident about your research, I think not much can go wrong. – Ian Jan 17 '17 at 9:34
  • 6
    Why wouldn't you ask? – user2390246 Jan 17 '17 at 11:03
  • 1
    @JoãoMendes good point, I edited the title. – Ian Jan 17 '17 at 11:07
  • 16
    @user2390246 I guess it was just general insecurity about asking allegedly obvious questions. – Ian Jan 17 '17 at 11:08
76

Yes, you should check with any co-authors before publicly posting a preprint of work that is not yet published.

They may prefer that you not post it at this time (e.g. if they have a particular timeline in mind for when they want to publicly share the work, given other related things they are working on). Even if they have no reason to object, you should check with them as a courtesy.

  • 29
    As you suggested I wrote an e-mail to my professor and he called me afterwards. He said he never heard of it before and asked what it is. After explaining everything I knew, he agreed to submitting it to arXiv. So asking was probably the right way to inform my supervisor about what arXiv is. Thanks! – Ian Jan 17 '17 at 9:32
  • 27
    @ian_itor Now I have become very curious what field you are in which is within the scope of arXiv but where it is possible for a professor to not even have heard of it. – Tobias Kildetoft Jan 17 '17 at 10:42
  • 3
    My area is materials science. I guess arXiv is not as common as in other physics-related fields, but it is there. – Ian Jan 17 '17 at 11:05
64

Submitting to ArXiv is a form of publication. You always need the consent of all coauthors to publish anything, anywhere.

  • 12
    This is the best and simplest answer. It's not a courtesy to notify your co-authors, it's a professional requirement. – David Jan 17 '17 at 16:47
  • 1
    I love when a one-liner answer falls together. – Mindwin Jan 18 '17 at 10:24
14

About "virtually no downsides": the following is a real story. A mathematician proves his best theorem ever. Writes the paper, sends it to a very good journal, and uploads to the arxiv. The very good journal takes a year to referee the paper, and rejects it (happens all the time, at least in my area). During that year, a group of other mathematicians read the paper in the arxiv, worked on it, proved a generalization, and got it published in a journal. So, a year later, the original paper is virtually unpublishable, and our friend has to scrap it.

As you can imagine, this person never again uploaded anything to the arxiv.

  • 6
    But at least the paper from the other mathematicians has to cite his arxiv publication, haven't they? – Trilarion Jan 17 '17 at 16:35
  • 6
    @MartinArgerami If people are only citing the more general follow-up article, I'm pretty sure that's because it's more general, not because your friend's article is only on ArXiv. – David Richerby Jan 17 '17 at 16:55
  • 7
    And how would the story be different if he published the paper and soon after, the other group would had proven the generalization? – Nick S Jan 17 '17 at 17:06
  • 20
    This is a very sad story, but I wonder if "never again upload anything to the arxiv" is the right moral. Would the mathematician really have done better to keep his important work absolutely secret for several years? He could find himself in the same situation at the end. I would rather blame several other aspects of mathematics publication culture. In particular, if someone writes a paper explicitly generalizing your theorem, then that should enhance the value of your paper, not diminish it. The entire community should understand this and be sympathetic to it. – Pete L. Clark Jan 17 '17 at 19:04
  • 5
    @Pete: I learned about this story many years ago, and I kept using the Arxiv, so I understand its value. But I can understand the frustration if that were to happen to me. I fully agree about the publication culture. – Martin Argerami Jan 17 '17 at 19:51
3

It looks like you are afraid that he would refuse, so you are asking us whether you should ask him, when you could have asked him directly in the first place.

Since he is the co-author you should definitely ask him. Perhaps the question you should be asking us is how to persuade him that uploading pre-print to arxiv is a good idea.

  • Thanks for your comment. You are partly right, I could have just asked, but you know how it is with asking professors, you tend to overthink stuff. I just thought maybe it is such a common thing to do, that it goes without saying. – Ian Jan 17 '17 at 9:31
  • 2
    Not should ask the coauthor; must! – David Richerby Jan 17 '17 at 10:46
  • 1
    @DavidRicherby: I think in English logic we have the tautology ( should definitely ask = must ask ). =) – user21820 Jan 17 '17 at 12:04
  • 1
    @user21820 I agree that "should definitely" is stronger than "should", so my comment was a little unfair. However, I think "should definitely" is still at the level of advising somebody to do something; I would say that there is a moral obligation to consult all coauthors before publishing something in their name. – David Richerby Jan 17 '17 at 12:14
  • 1
    @DavidRicherby: Yup I'm not disagreeing at all. English is weird; the "definitely" definitely leaves (at least) me no doubt that there should not be any case whatsoever in which you have reason not to ask him. But then "should" is the outer operator... – user21820 Jan 17 '17 at 12:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.