Someone has recently been flagging lots of very old questions for closure as "primarily opinion-based" or "off topic". In my opinion, there is no good reason to close very old questions, especially if they have many upvotes, or answers with many upvotes, so I have voted against almost all of these flags. But perhaps this should be discussed.

  • 2
    Looking at highly upvoted closed questions it seems that in the past it happened quite frequently that question was closed several months after being asked. (I assume that somebody a bit of spare time and sufficient SQL skills would be able to find more detailed stats from data explorer, but even by simply checking the linked search results one can find examples where this period was longer than one year.) – Martin Sleziak Oct 13 '17 at 12:13
  • 4
    I don't think it normally makes much sense to close old, answered questions -- unless as "no longer relevant", but that close reason doesn't exist any more. In my opinion, "no longer relevant" should be the only possible close reason for old questions. – Stefan Kohl Oct 13 '17 at 12:23
  • 2
    @MartinSleziak: Many of these were closed in the MO 1.0 era when historical locks were not available. – François G. Dorais Oct 13 '17 at 16:50
  • 2
    As @FrançoisG.Dorais points out, closing might have changed a bit after joining SE. So here is the same search query restricted to posts made after joining SE. (I.e., list of closed questions posted since July 2013.) I do not know whether it is possible to search for questions closed after a certain date. (I think that this can probably done only via SEDE.) Randomly checking a few posts from the linked list, the interval between posting and closure seems to be much much shorter. – Martin Sleziak Oct 13 '17 at 17:08
  • 2
    I have seen the same phenomenon and FWIW I agree (in most of these cases) with Neil. Quite a few of these might, if asked now, be more appropriate for academia.SE, but they were asked long before that site got off the ground – Yemon Choi Oct 14 '17 at 19:32
  • This older discussion seems (at least to some extent) related: Should there be a “statute of limitations” for closable questions? – Martin Sleziak Oct 19 '17 at 7:36
  • 2
    I note that some people now seem to be flagging old questions, which would be off-topic if they were asked now, for deletion. This seems to me even more drastic, and I'd welcome people's thoughts or justifications (one way or the other). Is this a case for reviving tea.mathoverflow? – Yemon Choi Oct 22 '17 at 17:26
  • 2
  • 1
    And what about these delete votes? – 35093731895230467514051 Oct 22 '17 at 18:22
  • 1
    @JosephVanName I can't vote to cancel delete votes cast by other people. In any case, this meta.MO question is about action on "very old questions" to use Neil's words, even if my recent comments may concern some newer questions – Yemon Choi Oct 22 '17 at 21:17
  • @Yemon Choi. I really do not get what point if any you are attempting to articulate. – 35093731895230467514051 Oct 22 '17 at 21:24
  • 1
    @JosephVanName My apologies for any confusion on my part. I brought up the issue of people voting to delete old questions that would now be probably off-topic, since it seemed relevant to Neil Strickland's original question. Your comment asks about some other unfair/unjustified delete votes on your own answers. It's not clear to me why you brought this up as a comment on this particular meta.MO question – Yemon Choi Oct 22 '17 at 21:29
  • My point is that people upvote, downvote, close vote, and delete vote on MO for nonsensical and emotionally charged reasons. – 35093731895230467514051 Oct 22 '17 at 21:35
  • Yes, emotions are often nonsensical :) Still we have built-in emotions in our "software", so we should accept it. Regards, – Marek Mitros Oct 24 '17 at 11:37
  • 1
    Sorry, I do not understand the topic here. Just want to add, that we cannot neglect psychology when doing portal. The rules should be for people. Exactly, people do decisions based on looking at the others. Why do you condemn this ? We should define such rules on MO which will be good for people. For example we should remove down-votes, and allow only up-votes. Similarly, on facebook, you can only add "likes", not "dislikes". No one is happy when seeing down votes on his post. See example below. – Marek Mitros Oct 25 '17 at 12:17

First, let me stress that I am not trying to suggest any specific course of action here, just argue why those flags make some sense. The whole point of this answer is trying to explain a phenomenon, not defend it. Take me as a devil's advocate if you will. I don't have a strong opinion on whether such questions merit closure, and in most cases I personally prefer to keep those old questions open.

If there are old questions which, if asked now, would be off-topic, it does make some sense to close them. The scope of the site has evolved — especially in connection to other SE sites like Mathematics and Academia — and closing old questions that would be better at those sites sends a signal to current and new users. If we want to discourage users from posting new questions of some kind, it can be (in part) signaled by marking old questions of that kind as off-topic.

If a question is months or years old, it typically has received all the answers it will ever get, and closure does not effect visibility or voting. Therefore, while closure might not gain much, it certainly doesn't hurt much, either. Having questions closed doesn't cost any reputation; it only hurts one's question score which isn't all that relevant in most cases.

Moderators can also lock a question and this is one of the offered reasons:

This question exists because it has historical significance, but it is not considered a good, on-topic question for this site, so please do not use it as evidence that you can ask similar questions here. This question and its answers are frozen and cannot be changed. More info: help center.

Locking makes it impossible to comment or vote, but the question is still visible. I think closing does less harm, but locking is a relevant option too.

I'm not advocating that we should go and close all (or any) old questions that don't fit the current scope. I'm just saying that it's not utterly nonsensical to close at least some of such questions. I assume that sending clear signals and having a consistent scope is what the flags were trying to achieve. But I'm not the flagger in question, so I can only guess.

  • 1
    Certainly, sometimes new posters cite an old question as evidence that their new question is on-topic. Isn't there some sort of banner that exists stating "this is a legacy question" or something ... I seem to remember seeing these banners. – Gerald Edgar Oct 18 '17 at 14:28
  • @GeraldEdgar I looked into that at a site where I moderate and added some details. The wordings have changed over time, and the one I posted is the current one. – Joonas Ilmavirta Oct 18 '17 at 14:40
  • 2
    Great, thanks. I think our over-eager closers should consider whether the old question they want to close should instead be flagged so the moderators can lock it with this banner. – Gerald Edgar Oct 18 '17 at 14:44
  • 1
    I would be curious to know what triggered down votes. I'm not trying to argue that we should agree with the flags and close such questions, but trying to see where the flagger might be coming from. – Joonas Ilmavirta Oct 18 '17 at 18:50
  • 5
    @JoonasIlmavirta: I simply find it odd to close a question as off-topic months or years after it has been asked and answered. After it has been well-received for a long time, and has got a lot of positive feedback in the form of answers and upvotes. If a question is off-topic on this site, I think this should be decided within a reasonable timespan, and ideally before it has been answered. And locks which also prevent edits, even trivial ones e.g. to update links, and adding comments, are a rather drastic measure. – Stefan Kohl Oct 18 '17 at 19:21
  • 1
    @StefanKohl Thanks! I agree that the lock is a very drastic measure. Closure does less harm, but there is little to be gained with it either and it would take too much work to try to close all such old questions. I would let the old questions be, but I have no strong opinion. – Joonas Ilmavirta Oct 18 '17 at 19:25
  • 1
    @JoonasIlmavirta I do not understand that "passion" for closing questions. What is the reason for it. I have done down-voting, because I am strongly against closing good questions. I will add answer to this thread with more explanation of my point of view. I hope it will not be removed :( – Marek Mitros Oct 24 '17 at 9:38
  • 1
    @MarekMitros I don't understand a passion for closing either. I understand why some may vote to close old questions, but I would only do that myself very rarely. The whole point of my answer was trying to explain a phenomenon, not defend it. Your answer is of course a welcome addition to this meta discussion! :) – Joonas Ilmavirta Oct 24 '17 at 11:27
  • @JoonasIlmavirta Thank you for your explanation. Sorry for doing down vote. This way you can understand that no one is happy when experience down vote or closure of his own post on MO or other portal. That's why I suggest to remove down-voting from MO policy, just allow upvoting. On facebook you can only put "likes" not "dislikes". People prefer to hear good things, not bad things about theirself. This is psychology. If you have to say something negative to your wife (or other person), you should think twice, how to do it in order to maintain relationship. – Marek Mitros Oct 25 '17 at 12:25
  • 1
    @StefanKohl General guidance for the entire SE network is that if what is "on-topic" changes, than closing old questions that no longer meet the requirements is fine (but there's no reason to seek them out). I think people interpret closure too negatively, everyone involved keeps earned reputation, links still work, it just prevents activity on something no longer appropriate. IMO, if something is off-topic and is currently attracting attention, than closure makes sense, otherwise its basically a pointless (no benefit, no harm) activity. – mbrig Oct 25 '17 at 17:33
  • 4
    @mbrig the professional mathematicians on MO have their own culture and customs, so even if the rest of the SE network likes to close and delete at their time of asking well received questions this does not necessarily be true for this community too. Also, note that MO was founded outside the SE network and joined only relatively recently and is by contract not obliged to follow and hold up SE dogma if it clashes with the interests and tradition of research mathematicians. If most people here consider the content of old questions still valuable, closure or even deletion makes no sense... – Dilaton Oct 25 '17 at 17:54
  • 2
    @Dilaton I specifically did not say deletion because closure and deletion have very different meanings. Deleting content means its valueless or harmful, but closure (and keeping it around forever) implies that it has value but the community no longer wishes to see activity around it. Which is exactly the intent with valuable but no-longer on-topic content. Conflating the two is quite possibly why people get unnecessarily angered by having their questions closed. – mbrig Oct 26 '17 at 1:16
  • 1
    Re: it certainly doesn't hurt much. One thing which can be influenced by this is that user with too many closed questions can get into question ban. Still, I vaguely recall reading somewhere that downvotes on posts contribute to post ban much more than closures, so in this regard it hurts only a little. (And this might be relevant only if the user in question had many of their questions closed.) – Martin Sleziak Oct 27 '17 at 5:42
  • @mbrig I have gained the impression, since the activity on this post, that some of the flagging for closing is done with the express intention of subsequently voting for deletion. I may be mistaken, though – Yemon Choi Dec 4 '17 at 22:03


I am strongly against of closing old question, just because they are old. Imagine questions asked by David Hilbert or Friedrich Gauss. Why would you close them ? There is saying "no such thing as a bad question" in many languages. I have looked "highly upvoted closed questions" link offered by Martin Sleziak. There are many interesting questions with answers. Unfortunately I have no time to read them. Either concerning mathematical career, publishing process or mathematics itself. All of them are valid. We are human beings, so we ask questions and we look for the answers.

Few comments about MO

MO portal is organized as "questions and answers" forum. I am not sure whether it works in a good way. It is not possible to avoid discussions between people on certain subjects. Therefore - even moderators try hard - there are and there will be discussions on MO.

I have one suggestion - to remove down-votes and keep only up-votes in MO policy. No one is happy receiving down votes. This is psychology. What positive do we achieve by putting down votes on someone answer or question?

It is funny, because I received down-vote after I added suggestion to remove down-votes :)

Few comments about closing questions

What is the reason for closing question on MO? The standard explanation says that the reason is "not research level" question. My experience is that sometimes the reasons are:

  • person asking is not known
  • language of the question is not good - not using mathematical slang currently used
  • the subject is not hot topic - currently not on the agenda of working mathematicians, not popular enough to attach interest

For example philosophical questions asked by known mathematicians are accepted. The same questions asked by layman are rejected. Say that I ask "how to imagine 3-manifold ?". That question would be closed as "not research level question". If the same question would be asked by Michael Freedman, then it will be accepted.

There are two questions on this forum with divagations what Sir Atiyah thinks about octonions. This appears interesting and no one dares to close such a question. When I ask about octonions, my questions are being closed. I have to fight to defend my questions against closure. Is it fair?

I agree that questions of type "can someone help me to do homework?" should be removed or moved to stack exchange.

You must log in to answer this question.

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