6

I have seen that some researchers in my field, that is Computer Science, do this:

  • They publish a paper, for example titled "The use of the Fourier transform for problem X". The paper gets accepted.

  • Then after a while they have another paper titled "The use of the Fourier transform for solving the problem X under these constraints". I read the paper and I see this is the same as before with only the title changed a little bit and published in another conference.

  • Again, the same paper in another language "l'uso della trasformata di Fourier...", which has the same content, but submitted to a local conference

Is this alright? For me it just seems that it should only be put like one publication under their CVs and not "resell" the idea over and over. I have this question because they had asked me to translate one of my papers to a local conference, but I do not consider that to be like a new publication. Should I translate and claim a new publication?

  • 6
    If (parts or most) of the text of the two papers is the same, this is self-plagiarism, and it is unhetical. Otherwise, if they added certain constraints for the second paper and they avoided writing the same text, it might be "less unhetical". It depends more on how much under these constraints is worth a publication. I do not know how it works when you translate a paper for a local/national university. It might even be allowed for what I know. – user7112 Dec 6 '13 at 13:29
  • To answer your question in a more general way: a publication is a publication. All publications can go to your CV. How much these publications are ethical is another issue. – user7112 Dec 6 '13 at 13:31
  • 4
    Well, sure, but that's like saying any car can go in your garage, whether you stole it or not. – JeffE Dec 6 '13 at 19:02
  • @dgraziotin Self-plagiarism is not merely unethical; it is a serious misconduct. – gerrit Dec 7 '13 at 18:00
  • Translated papers should go to "translations" section, but not "research papers" section. – Erkin Alp Güney Mar 13 '17 at 18:10
7

In computer science, papers are published in archival conferences, which makes things a little confusing. (In many other fields, papers are published in journals, and conferences are just for giving talks -- you can only publish a paper once, but you can give a talk about it many times).

a) It is unethical, in violation of the rules of the venue, and an incredibly stupid idea to publish the same paper more than once in an archival conference. This is self plagiarism, and is a serious academic offense.

b) On the other hand, in computer science there are also non-archival workshops to which people are invited to give talks. (These are more like the "conferences" in other fields). It is fine to give a talk at such a venue about a paper that has been previously published elsewhere.

Your question seems to be whether you personally should commit unethical and stupid act a). I would advise against it. (It is unethical for obvious reasons. It is stupid because anyone who seriously considers your publication record will be sure to spot such a move, and you will not be able to get a reasonable academic job).

  • which makes things a little confusing — No, not really. If the conference/symposium/workshop publishes proceedings with an ISBN or ISSN, it's an archival conference paper. Publishing the same paper in more than one such venue is unethical. – JeffE Dec 6 '13 at 19:02
  • @JeffE That's true. However, in my subfield of TCS, you quite commonly have 12-page proceedings that shorten/omit the proofs, and the very same results in a full paper (either in a special issue or as a standalone publication). This is not considered unethical at all, as long as the proceedings are really proceedings and the paper is really a blind-reviewed paper. – yo' Dec 10 '13 at 21:55
5

For the detail you've provided, it's difficult to tell.

  • Self plagarism is a big no-no, obviously.
  • However, when you're building on previously published work (e.g. by adding constraints), there's almost certainly going to be overlap.

I think what I'd look for is as follows:

  • Does the new work provide a meaningful contribution? (E.g. is solving the problem with constraints tricky to do, or does it give unexpected results?) This will depend on the field, and the standards of the conference.
  • Is the previously published paper cited by the new work? You don't specify whether this is the case or not in your question, and I think it's critical.
  • Are large portions of the text lifted verbatim? If the author is continuing with prior work there's bound to be some crossover (there's only so many ways you can clearly describe a problem using standard terminology, after all, and that's before we consider page limits), but it shouldn't be a cut&paste job.

If they pass all that then it's probably ok, even if the paper isn't exactly wonderful - but chances are then, if it's making it through the review stage then it's probably been submitted to a less prestigious conference, and people will pick up on that.

With regard to translating a paper for a local conference, it's not an issue I've ever faced and I'm a lot less sure about what the accepted etiquette is. My gut feeling though is the key test would be whether anyone is being mislead as to what's happened. Off the top of my head this including the local conference organisers, the conference that published the original version of the paper, any funding bodies who might ask how many papers you've published, and the people later reading your CV. If you're deceiving any of them in the way you present the publication, or even leaving an unreasonable opportunity for them to misunderstand, then no, don't. Otherwise, I can't see a reason to object.

0

Such a paper can be published, but it should not be treated as a new research output. Using the Australian HERDC specifications for research publications at 9.1 a publication is characterised by:

  • substantial scholarly activity, as evidenced by discussion of the relevant literature, an awareness of the history and antecedents of work described, and provided in a format which allows a reader to trace sources of the work, including through citations and footnotes
  • originality (i.e. not a compilation of existing works. See important notes below regarding the treatment of scholarly editions and scholarly translations)
  • [...]
  • increasing the stock of knowledge [...]

When dealing with papers, a new paper on a theme must still "increase the stock of knowledge" by "substantial scholarly activity" and be "original". Adding a new case study may fit the bill. Using a new technique of analysis may fit the bill. A new theoretical context may fit the bill. "Substantiveness" would be indicated by the portion of the work that is "original." If the original component would fly as a paper by itself, in the sense that the component of the paper that is original scholarly contribution to knowledge is reportable as research by itself; then it is fine. My dictum when dealing with these is "new evidence?" "new analysis?" "new conclusion?." If it meets one of these, originality has been met, and probably substantial scholarly activity.

Authors have a separate responsibility in terms of IP and copyright agreements that they may have previously made, and obviously have ethical obligations to cite prior works.

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.