I'd like to mention here two possibly related facts:

(a) Soon after this question was answered, there was an arXiv paper posted providing the same answer

(b) Soon after this closely related subsequent question was answered, the post was deleted by the OP. In this case, the answer was quite elaborate. (It's been copied here (*) by the answerer).

The OP are distinct but anonymous with low reputation, which suggests they could be the same or related (moderator can have more info about this?) Edit: Jeremy's link points to a paper by the same named author, by the second OP. So it seems quite clear that both accounts belong the same person. [Further edit, by YChoi: it has been confirmed by a moderator that the two accounts share the same IP address.]

Note sure what to ask at this point, except that the second question should be undeleted [edit: now done by the OP after peer pressure :), so the additional post (*) has been deleted too as redundant].

  • 22
    There is another arXiv paper by the same author arxiv.org/abs/1802.03631 whose content is mostly Eric Wofsey’s answer to this MSE question math.stackexchange.com/questions/2589180 pretty much word for word and unacknowledged. – Jeremy Rickard Jul 8 at 12:42
  • 3
    Very related: meta.mathoverflow.net/questions/3523/… – Federico Poloni Jul 8 at 13:04
  • 1
    @FedericoPoloni yes. I left the post because while the deletion problem is settled , the other aspect (arxiv unacknowledged posts) is not. – YCor Jul 8 at 13:05
  • @YCor Sure, I didn't mean it as a negative comment; I agree with you that this question deserves some attention. – Federico Poloni Jul 8 at 13:08
  • 4
    I just opened a feature request post on Meta Stack Exchange: meta.stackexchange.com/questions/312315/… – Federico Poloni Jul 8 at 13:23
  • 6
    @YCor many thanks for bringing this to everyone's attention. (I have misgivings about something like this potentially happening with some TCS flavoured questions, but I admit I have no evidence it has actually yet happened) – Yemon Choi Jul 8 at 15:45
  • 1
    Interesting idea ... suppose I post an answer in MO using a pseudonym. Then if I later write a paper containing that same thing, must I "out" my pseudonym? – Gerald Edgar Jul 9 at 12:14
  • 3
    @GeraldEdgar in the given cases the answer are not under a pseudonym, and the author of the arXiv papers is rather suspected to coincide with the OP. If you post under a pseudonym (and try to stay anonymous), it's quite incoherent to publish a paper based on such an anonymous post, since, even if you don't quote the MO post, people would probably assume you coincide with the answerer. [This does not technically answer your question. To make it fun, you might in addition complain for plagiarism under your nickname :)]. – YCor Jul 9 at 14:32
  • 1
    @Gerald, this just underscores the importance of proper attribution, especially if there is value in operating under a pseudonym. Gerhard "Uses Own Name For Pseudonym" Paseman, 2018.07.09. – Gerhard Paseman Jul 9 at 16:42
  • 10
    arxiv.org/abs/1802.03631 has now been deleted by the author, and in arxiv.org/abs/1806.03735 an acknowledgement to the MO posting has been added – Carlo Beenakker Jul 10 at 8:32
  • 3
    @CarloBeenakker thanks for the update (the arxiv paper is not and can't be deleted, but rather withdrawn: one can still access the posted version by clicking to the previous versions) – YCor Jul 10 at 8:51
  • 4
    @CarloBeenakker Thanks for this information! I did not study attentively enough either the MO answer or the arxiv preprint to judge, but my impression is that the acknowledgement in the latter is very far from being adequate. – მამუკა ჯიბლაძე Jul 12 at 19:36
  • 5
    Just to give a perennial link to the update mentioned by Carlo and მამუკა ჯიბლაძე it's arxiv.org/abs/1806.03735v4. The update barely consists in the sentence "I would like to acknowledge the discussions in the Math overflow question as mentioned in [4]", [4] being the ref+link to the MO question. – YCor Jul 12 at 20:14
  • 11
    And the removed paper (arxiv.org/abs/1802.03631v2) gives attribution to the MSE question with extreme weasel wording: "It appears similar argument is also in the answer to the question [this https URL]", as if it were some kind of accident that the proof was cut-and-paste. – David Roberts Jul 15 at 12:39
  • 2
    And another question by the same person: mathoverflow.net/questions/306488/… (no answers as of yet) – David Roberts Jul 20 at 23:59

While I'm not a lawyer, the part that was until recently unsettled, namely the use of answers in an arXiv preprint without proper attribution, falls squarely under the terms of the Creative Commons licensing, summarized here, and it seems to me those terms had been violated.

My understanding is that proper attribution in this case is legally enforceable. A moderator message to the OP is in order.

  • 16
    Do MO mods have any plans to contact the arXiv mods? – Yemon Choi Jul 8 at 15:50
  • 4
    Good point, Yemon; thanks. Looking into that as well. – Todd Trimble Jul 8 at 16:09
  • 4
    What if it is further submitted and accepted for publication in a journal? Must not be easy to watch all such cases... – მამუკა ჯიბლაძე Jul 8 at 21:13
  • 22
    Isn't the underlying problem rather academic misconduct than a copyright violation? -- While copyright comes into play when actual text is copied, copyright does not prevent stealing mathematical ideas. That means, copyright merely forces the offender to formulate the proofs etc. in their own words, not more -- or am I mistaken? – Stefan Kohl Jul 8 at 21:36
  • 4
    @StefanKohl Jeremy Rickard noted another instance where it seemed to be plagiarism pure and simple. I haven't looked closely enough at the current case to see whether it approaches such a level of sheer copying. – Todd Trimble Jul 8 at 23:08
  • 16
    one of the offending arXiv submissions has now been deleted and the other has an acknowledgement to MO; at least this thread had an effect, hopefully the author will have learned that this is not the way to proceed in scientific research. – Carlo Beenakker Jul 10 at 8:41
  • 3
    I would not say unattributed use "falls squarely under the terms of the Creative Commons license", but rather "is in violation of the terms of the Creative Commons license" - the former phrasing to me sounds like it's something that could be allowed. – David Roberts Jul 15 at 12:34
  • @DavidRoberts Taking my response as a whole, I don't see how anyone could mistake my meaning. I'd prefer to parse it as "the part that is still unsettled... falls squarely under the terms of the Creative Commons license". – Todd Trimble Jul 15 at 13:01
  • @ToddTrimble actually I had to do a double take to make sure I'm getting your response correctly --- I did make the same interpretation as David originally, but given the wider context of the question and answer, did the double take to re-interpret. – Suvrit Jul 18 at 11:03
  • @Suvrit Does this edit make my position clearer? The original wording was me being cautious and careful, but I'm okay with stating my opinion more strongly (while still being surprised that my original feelings were hard to decipher). – Todd Trimble Jul 18 at 13:57
  • @ToddTrimble Yes, it does! Thanks. I just wonder how many other such violations exist --- probably not that many, but then even a single one is already more violations than there should be. – Suvrit Jul 18 at 20:33
  • 4
    Off topic for this "thread", Todd, but have the mods looked into the suggestion that the two distinct user accounts are actually the same person? – Yemon Choi Jul 19 at 0:08
  • 10
    @YemonChoi They come from the same IP address. – Todd Trimble Jul 19 at 0:19
  • 3
    I've attempted to manually generate a trackback to this page on the arxiv page. We don't usually generate trackbacks for meta posts, but it seems appropriate in this case. (It will probably take a while to appear, as the trackback will require manual moderation on the arxiv end.) – Scott Morrison Jul 24 at 21:38

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .