30

I am reading IEEE's policy on uploading a conference paper to other venues such as arxiv. Here is the current policy: https://conferences.ieeeauthorcenter.ieee.org/get-published/post-your-paper/. What is confusing me is that IEEE differentiates between preprint and author-submitted paper, where it is okay to put a pre-print on ArXiv, but not okay to put the author-submitted version.

Is the following understanding of the rules correct? One is allowed to upload any version of the paper to ArXiv except for (1) the version submitted for peer-review, prior to any changes being made to the paper, i.e., before addressing the reviewers comments, and (2) the final version of the paper, which is different from the accepted version, as it undergoes final styling and copy-editing.

Also, on a meta level, what is the logic between allowing preprints, but not author-submitted papers, and how much difference does there have to be between the two versions to not count as the same?

  • If your nickname should mean that you are in Germany, take a close look at §38 (4) UrhG gesetze-im-internet.de/urhg/__38.html - you may have the right to a secondary publication after an embargo period of 1 year. – cbeleites unhappy with SX Mar 23 at 22:58
  • 1
    pre-print = accepted, but not yet published. author-submitted = still under review? Putting the latter on arxiv breaks double-blind reviewing, which is standard among many IEEE venues. – Daniel Mar 24 at 11:03
57

what is the logic between allowing preprints, but not author-submitted papers?

There is no logic behind this. It is part of the culture of publishers who want to restrict access as much as possible, so they can profit from selling closed-access research.

If it were just up to publishers, I don't think IEEE would allow preprints, either. But they make an exception for arXiv because the culture has changed such that people want to make their papers available there. Additionally, note that it is valid to post any version of your paper other than the final copy-edited version online on your website.

(Note: if you ask IEEE, they might come up with some reason for it. Perhaps they treat submitting to arXiv as a competing submission, and since you can't submit elsewhere while a paper is under review, you shouldn't be able to submit to arXiv either. But I don't think this objection holds water, because arXiv and posting to a personal website have similar effects. Also, I don't think it is the real reason, which is that their business model relies on limiting paper access to some extent or another.)

What is confusing me is that IEEE differentiates between preprint and author-submitted paper, where it is okay to put a pre-print on arXiv, but not okay to put the author-submitted version.

It's worth pointing out that many researchers do wait to publish online until a paper is accepted. The idea is that papers go through lots of revisions, and the title, narrative, and main results of a paper could change after it goes through the publication process (possibly being rejected a few times). So, you may not want people to read the draft version and get a bad idea (or wrong idea) about where your work fits, and you prefer to make only the most impressive and clarified version available.

how much difference does there have to be between the two versions to not count as the same?

There is no rule here, as far as I know. It could even be the same version, but that probably means you ignored the reviews, which is a bad thing. Often for my papers, there is also zero difference between the preprint and the "final" version, other than it having a new template and formatting to look pretty, which doesn't really matter. (And I prefer to treat the preprint as the real "final" version.)

| improve this answer | |
  • 3
    Actually IEEE is a non profit professional organization, not a publisher per se. They publish things as a service to members and use any proceeds to support professional activities. – Buffy Mar 21 at 22:59
  • 12
    @Buffy While you are correct, they are the publisher for a large number of journals and conferences. I'm also not sure the non profit designation makes a real difference here. – 6005 Mar 21 at 23:07
  • 24
    About "the most impressive and clarified version available": For some (admittedly only a few) of my papers, that was the version before incorporating the referee's recommendations. – Andreas Blass Mar 22 at 2:37
  • 3
    I think this answer is correct (and spot on). There, I gave you a source to cite. :-) – Dan Romik Mar 23 at 2:55
  • 8
    @Buffy While the IEEE is a "non-profit organization," that just means that they don't return profits to owners or shareholders as cash (i.e., distribute dividends). They still, however, restrict much of the value they create or acquire (such as papers) to their members, rather than making that value available to the general public. – cjs Mar 23 at 4:10
-8

Well, by your own words, they want your paper to go through their peer-review before being unleashed on the world. On the one hand, they think it will make Arxiv, and the world, a better place. But more than that, they want to prevent headaches, drama, and confusion, whereby a hypothetically crummy version of your paper co-exists in this universe with a superior version that IEEE published. Imagine the bad version gets scores of vitriolic comments on reddit, which will never be forgotten by the Internet.

If the IEEE were 100% evil (as suggested by the other poster), they'd only let you publish the pre-review paper on Arxiv, to try to make Arxiv the repository of bad, unreviewed papers.

| improve this answer | |
  • 8
    Really bad answer... I'm not sure why some people think peer-review is just a god given thing... Also, some people think when something passed peer-review should be taken as 100% correct and any other thing which is not peer-reviewed yet should be taken as 100% incorrect. In my opinion. it doesn't matter if something is peer-reviewed or not, as long as it makes sense scientifically and could be reproduced, I'm happy to use it in my research, otherwise I don't care if something is published in IEEE. Nature, Science, etc. cause there are lots of junk science published in these elite journals – Alone Programmer Mar 22 at 19:07
  • @AloneProgrammer As per your username, I take it you also hate code review. – Aleksandr Dubinsky Mar 22 at 19:27
  • Where did I say "I hate review"? I'm just saying this culture of nothing should be disseminated unless it is peer-reviewed is pretty much dead right now... – Alone Programmer Mar 22 at 19:31
  • 9
    If this were IEEE's reasoning, then they would not allow publishing a preprint on your website before peer review either. As it stands, the policy has no logic to it. They aren't 100% evil (if they were, they would not let you publish online at all), but they are partially evil (still recovering from an outdated system). – 6005 Mar 22 at 19:31
  • 3
    The other answer does not suggest (or come even remotely close to suggesting) that IEEE is “100% evil”. – Dan Romik Mar 23 at 2:59

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.