The latest example is My idea about finding the largest prime number where OP asks about a distributed processing approach to find prime numbers. There was a comment: mersenne.org – pedja which OP accepted in another comment as showing the idea was not original. There are no answers, but OP seems happy. If nothing more happens this question will continue to be bumped by the robot. Should there be a closure reason to deal with this? I think so.

  • 1
    $\begingroup$ Peripheral note on "...bumped by the robot." I have never seen a question with 0 answers bumped by the robot. It seems only questions with at least one answer but none with positive score get bumped. Am I wrong? $\endgroup$ May 5, 2012 at 4:16
  • 1
    $\begingroup$ @JonasMeyer: I believe I have seen questions with no answers bumped, but don't see an example on the front page right now. Certainly those with answers with no upvotes get bumped and I am prone to upvote an answer (even if I don't think it is very good) just to stop that. $\endgroup$ May 5, 2012 at 4:19
  • 2
  • 3
    $\begingroup$ I wish we had "No longer relevant" closing reason here... $\endgroup$
    – Asaf Karagila Mod
    May 5, 2012 at 7:16
  • $\begingroup$ @Jonas: questions with no answers can get bumped. But I think the bumping mechanism factors in also votes on the question and/or views or something like that (questions are not bumped entirely based on a uniform random distribution). So there may be some sort of bias on that. $\endgroup$ May 7, 2012 at 7:35
  • $\begingroup$ @WillieWong: Do you have evidence or a reference? I've never seen it happen. $\endgroup$ May 7, 2012 at 20:26
  • $\begingroup$ @Jonas: evidence are hard to come by, mostly because the community user bumps are not recorded in the history, so unless I take a screen shot as it happens (and as bumps happen hourly, it may take a while), it will be hard to give you evidence. But in so far as references go: meta.stackexchange.com/questions/19738/… states that the community user bumps unanswered questions, where "unanswered" is defined blog.stackoverflow.com/2008/09/ok-now-define-answered as questions with "no accepted answers and no answers with upvotes". $\endgroup$ May 8, 2012 at 8:53
  • $\begingroup$ And from Jeff's answer on this question, meta.stackexchange.com/questions/31480/… zero answer questions are definitely not blocked from being bumped. But if you mean for evidence/reference for the second part of my previous comment, I plead "I think..." as in "My experience makes me want to conjecture that..." $\endgroup$ May 8, 2012 at 8:55
  • $\begingroup$ @Willie: Thank you. I did mean for the first part. My experience has been that questions with very low (or high or medium) views/vote count, but with a single answer (or more) with $\leq 0$ vote count, can get bumped often, while questions with high views/vote count but with 0 answers exist, and I have never seen one bumped. I've never heard an official reason, but I'm not the only one to have noticed. (Although not very active lately, there were times when I spent a lot of time on MO & here & this pattern seemed to persist. Though I won't be too surprised if what I saw was a coincidence.) $\endgroup$ May 8, 2012 at 14:07
  • 1
    $\begingroup$ @JonasMeyer you are right, at least now and since a very long time (quite sure also at the time of that discussion). This question with no answer can get bumped is a myth that's hard to eradicate. The cause is I think confusingly phrased documentation and that it seems plaisible it would happen. $\endgroup$
    – quid Mod
    Aug 21, 2017 at 14:48
  • 1
    $\begingroup$ @WillieWong I don't know what was current in '09 but you might overinterpret that answer. Anyway I am quite confident in '12 it was already as it is now, that is, question with no answer are not bumped. See meta.stackexchange.com/questions/217884/… and the answer linked there. It was also like that on MO 1.0 at least as of around '10 or '11. $\endgroup$
    – quid Mod
    Aug 21, 2017 at 15:00
  • 1
    $\begingroup$ @quid: you are probably right. The earliest unambiguous documentation I can find of the behavior you described however is Shog9 from '13 (which your link links to). (Not that I bothered trying too hard on this.) $\endgroup$ Aug 23, 2017 at 12:45

1 Answer 1


It's arguable that this question should be closed as "off topic", but barring that, I think the usual approach to such situations is to encourage the commenter to post an answer, which can then be accepted or at least upvoted (and that will stop the robot). If the commenter doesn't do so, then someone else can post the comment as an answer (by convention, using community-wiki).

  • $\begingroup$ I wasn't so interested in the specific question (which I would accept as on topic, but maybe my range is broader than others) as the general problem of seeing questions that have been answered acceptably come up again and again. I have seen some resistance to posting a link as an answer. $\endgroup$ May 5, 2012 at 4:41
  • 1
    $\begingroup$ The OP of the original question did not read the related question math.stackexchange.com/q/201/17111 before he posted his one. Otherwise he would not say the largest prime number. Had he said record prime number, I agree with what @Nate said. $\endgroup$
    – Nobody
    May 5, 2012 at 6:04

You must log in to answer this question.

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