As the question about the beauty of mathematics was closed, reopened and there is already a close vote again.

For sure it is possible that we close and reopen the question several times, I think for such a popular question are 5 votes not enough, maybe we should scale it, so a popular question needs much more (we should bound it perhaps) close votes to be closed.

Maybe we should make the something analog for reopening questions.

Edit: Rahul Narain did have the idea to give a button: Don't close which would decrement the vote counter for closing. I think that is it a good idea.

  • 1
    $\begingroup$ Case in point, from Mathoverflow: mathoverflow.net/questions/23478/… . That was the highest upvoted question on the site mathoverflow.net/questions?sort=votes . Open/close history: mathoverflow.net/revisions/23478/list . Meta: tea.mathoverflow.net/discussion/1165/… $\endgroup$ – zyx Mar 12 '13 at 13:55
  • $\begingroup$ Instead of scaling votes, there should be a meta thread in which the decision to close or open is to be agreed upon. As easy as that. $\endgroup$ – Asaf Karagila Mar 12 '13 at 14:02
  • $\begingroup$ @AsafKaragila then you enlarge the group of people who are deceding it, to participate in meta you need 5 reputation, to close and open questions you need 3000, and I think, with bad luck you will have two questions with a lot of discusionns (the original post and the meta discussion) $\endgroup$ – Dominic Michaelis Mar 12 '13 at 14:08
  • $\begingroup$ Last I recall most people support a democracy rather than dictatorship or anarchy. $\endgroup$ – Asaf Karagila Mar 12 '13 at 14:09
  • 1
    $\begingroup$ @Asaf: The "people" would support democracy, wouldn't they... $\endgroup$ – user642796 Mar 12 '13 at 14:28
  • $\begingroup$ If I remember correctly, when Bhutan became a democracy the "people" thought that the king should shut up and do his job. $\endgroup$ – user1729 Mar 12 '13 at 14:40
  • $\begingroup$ @AsafKaragila: perhaps you feel you're being treated... unfairly? >8( $\endgroup$ – robjohn Mar 12 '13 at 15:15
  • $\begingroup$ @robjohn: Vader was mean. You are mean squared. $\endgroup$ – Asaf Karagila Mar 12 '13 at 16:03
  • $\begingroup$ @AsafKaragila: it's been noted before. $\endgroup$ – robjohn Mar 13 '13 at 19:57
  • $\begingroup$ @robjohn: I may have memory problems, but I still remember a lot of irrelevant things. Like that reference. $\endgroup$ – Asaf Karagila Mar 13 '13 at 19:59

I can't really support having the requirements on closing/re-opening depend on the "popularity" of a question. (A popular question can be ill-suited for this site, and we have plenty of very good mathematical questions which receive few upvotes.)

I would have less qualms with increasing the number of votes needed to close/re-open a question in each cycle of a close/re-open war. But IMHO these don't occur with enough frequency (on math.SE, at least) to be a major concern.

  • 1
    $\begingroup$ Why not have it depend on popularity? This is a signal, often a strong signal, that the material is of interest to the type of people who use a mathematics site, and it may not fit on any other forum. $\endgroup$ – zyx Mar 12 '13 at 14:46
  • 2
    $\begingroup$ Like chalk.${}$ $\endgroup$ – user1729 Mar 12 '13 at 14:55
  • 2
    $\begingroup$ @zyx: One problem is that there are no standard reasons for upvoting. In this metric a question can become popular by being linked on reddit or some similar aggregator (which I believe was part of the drive for the Batman equation question, and possibly also the question that spurred this proposal). $\endgroup$ – user642796 Mar 12 '13 at 15:32
  • $\begingroup$ There are very few high-voted questions that are closed, and the closed ones seem to all be by moderator actions, not through the 5 close votes. The number of reddit linked questions whose open/close status could be affected by raising the number to close appears to be zero for the questions above 50 votes. $\endgroup$ – zyx Mar 12 '13 at 16:45
  • $\begingroup$ @zyx i guess you are right $\endgroup$ – Dominic Michaelis Mar 12 '13 at 18:13

It would be smart of Stackexchange to do this but it seems to be outside the current model. They seem to like incrementing counters to trigger events but not anything that causes multiple variables to interact or can be hacked. A reduction in the upvotes could suddenly cause a closing, and this behavior can be exploited.

Some dependence on the upvotes could help questions that are of special interest but off-topic, like the Japanese chalk question on Mathoverflow.


You must log in to answer this question.

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