5
$\begingroup$

I was a bit surprised (annoyed?) by the migration of this question from MO to MSE. Yes, the question could have been asked on MSE, but it was asked on MO, received much attention there, with a high-level discussion in the comments. The answer was not high-level, but was apparently well-received by MO users.

The stated reason for the migration is "because it can be answered on Mathematics Stack Exchange", but in this case it was answered on MathOverflow, so that reason does not seem to apply.

There are many questions which have been answered on MO but could be answered on MSE or on HSM, but if they are well-received on MO, why transfer?

On a personal note, I find migration to be disruptive for the community feeling that at least for me is the most attractive aspect of MO.
$\endgroup$
13
  • 2
    $\begingroup$ I am not sure whether this is intended to be a discussion of a general issue (and the linked question serves merely as an example), or a discussion about this particular case. If it is the later, the tag (specific-question) might be suitable here. (And maybe also (math-stackexchange) might be a suitable tag.) $\endgroup$ Jul 2 at 9:25
  • 7
    $\begingroup$ well, this specific question triggered me, but it's not just about MSE, I'm hoping this will not become a new trend in moderation. In particular, there are many well-received questions on MO that could also have been asked on HSM, I would find it quite demotivating if moderators would start migrating them. $\endgroup$ Jul 2 at 9:46
  • 23
    $\begingroup$ My concern with the referred question is that it may give a false impression of what is on-topic on this site, and encourage people who don't know the site to ask similar-level homework questions -- and then complain that these are closed, pointing to that question which got lots of upvotes. (As to number of upvotes -- the question went to the HNQ list and thus got notable exposure to people outside MO.) In contrast, for example questions on the history on mathematics, on teaching mathematics or on aspects of academia relevant to mathematicians are not likely to encourage homework questions. $\endgroup$
    – Stefan Kohl Mod
    Jul 2 at 10:48
  • 3
    $\begingroup$ @StefanKohl I think there's an important difference between asking homework questions and asking about an unexpected solution that a student produced for a homework question. I've also had the experience of asking a homework question and getting a better solution than the one I intended. If MO had existed at the time, I might well have wanted to ask about it there. $\endgroup$ Jul 3 at 17:11
  • 2
    $\begingroup$ @CarloBeenakker: What I find surprising (annoying?) is that, according to what you say (I didn't see the original), the question was well received on MO. I have seen absolutely diluted and light questions receiving plenty of views and upvotes, and deep, difficult questions disappearing into obscurity (to the chagrin of their authors). $\endgroup$
    – Alex M.
    Jul 4 at 19:52
  • 14
    $\begingroup$ Migrating questions on the basis not of whether they are appropriate for this site, but on the basis of whether they are well received, would turn MO into (more?) of a popularity contest. It has often been observed that high-reputation or well known users can get away with asking questions that wouldn't have been accepted if asked literatim by a new user. This may be unavoidable, but enshrining it as policy would not, I think, be a clear win. $\endgroup$
    – LSpice
    Jul 4 at 23:40
  • 16
    $\begingroup$ I voted to migrate this because of the simple reason that the question is far from being about "research level mathematics". In retrospect, there would have been another reason to close this question, namely that the poster did not do any prior research on this (just google "sqrt(3) irrational" and you'll find what you are looking for). $\endgroup$
    – Dirk
    Jul 5 at 9:14
  • 5
    $\begingroup$ For some context, the poster is a reasonably well-known professional mathematician with papers in top journals and who advised REU projects for many future mathematicians. I’m not a big fan of sending professional mathematicians to math.SE and would still prefer the MO definition that it’s a site for professional mathematicians rather than “research level.” $\endgroup$ Jul 5 at 16:04
  • 12
    $\begingroup$ @NoahSnyder I would rather that posts be judged by their content rather than according to the social status of the author. By your definition, we would have to kick out independent researchers who happen not to be paid for doing mathematics, or even someone like Aubrey de Grey. $\endgroup$ Jul 5 at 20:00
  • 2
    $\begingroup$ @TimothyChow It doesn't follow that including professional researchers means we need to kick out non-professional researchers. A better standard is that MO is open to questions from anyone that a professional mathematician (at the graduate student level and above) would be likely to ask. $\endgroup$
    – Will Sawin
    Jul 6 at 1:55
  • 3
    $\begingroup$ @WillSawin I am fine with your standard because it is based on the content of the question rather than the social status of the questioner. I was reacting to Noah Snyder's comment, which began by noting the social status of the questioner and then affirming a desire to make MO a site for people belong to a particular social group. $\endgroup$ Jul 6 at 3:21
  • 4
    $\begingroup$ In a subtstantial sense, every question that can be asked and answered on MathOverflow can be asked and answered on m.s.e. $\qquad$ $\endgroup$ Jul 7 at 19:53
  • $\begingroup$ @StefanKohl Since you mentioned as one concern that the question was advertised in the HNQ, I will add a reminder that mods can manually remove a question from then HNQ. (And regular users can simply add MathJax to the title, if suitable.) Of course, this is only tangential here - the discussion is mainly about migration. $\endgroup$ Jul 8 at 5:58
8
$\begingroup$

The phrasing of the reason is unfortunate, IMO. It makes sense to migrate if a question is better suited to MSE; it doesn't necessarily make sense to migrate just because the question can be answered on MSE. In what follows, let me pretend that the wording "can be answered" has been changed to "better suited" since that may have been what was intended anyway.

Suppose that a question that is better suited to MSE gets asked on MO and is well-received. Does that mean it shouldn't be migrated? I'm not sure. Whether the question is well-received is an a posteriori criterion and I think that in general, we should be using a priori criteria to judge which site is more appropriate. Otherwise it tempts people to try to "game the system" and try their luck posting to a site that they kind of know is the wrong site.

For the particular question under discussion, I think the point is that it's not clear that MSE is actually the better site. The question does have some flaws; as someone has pointed out, the asker should probably have done some Googling first, but these flaws just mean that the question should be improved (regardless of which site it is posted to), and don't really address the issue of which site is better. Andrés E. Caicedo's comment indicates that some version of the question could have been better suited to MO, but I'm not sure that the question as asked is better suited to MO.

I agree that migration can be disruptive and annoying. So I'd propose the following policy: if there is doubt about whether MO or MSE is more suitable, then leave it on MO; but if it's fairly clear that MSE is more suitable, then go ahead and migrate it even if the question is "well-received." (Of course, if the question has flaws, then those should be addressed, probably before any decision about migration is made.)

$\endgroup$
7
  • 8
    $\begingroup$ IMO a necessary condition for migration away should be that a question is not suitable for MO (regardless of whether it might be more suitable elsewhere). $\endgroup$ Jul 5 at 16:19
  • 8
    $\begingroup$ I agree with Julian Rosen's comment, migration is definitely some kind of closure, which supposes that the question is deemed not suitable. And this is the reason I voted for migration in this very case. That is was well-received (by a number of users) was for me a reason to vote for migration rather than bare closure. $\endgroup$
    – YCor
    Jul 5 at 20:48
  • 2
    $\begingroup$ @YCor That's interesting; generally speaking, I don't think of migration as a form of closure. Do you also think of migration from MSE to MO as a form of closure? Or is it that MO is "high status" and MSE is "low status," so that migration from MO to MSE is a form of closure but migration from MSE to MO is not? $\endgroup$ Jul 5 at 23:47
  • 3
    $\begingroup$ @TimothyChow to vote migration from MO to MSE requires 4 clicks, the first is to click "Close", the second is choose "A community-specific reason. This question doesn’t meet a MathOverflow guideline." I don't think the inverse migration is based on similar steps (I never did it, I just heard that it has to be asked by OP to moderators.) $\endgroup$
    – YCor
    Jul 6 at 7:13
  • 6
    $\begingroup$ @YCor (and Timothy Chow): SE software allows communities to have some predefined destination for migration votes, but does not allow arbitrary migrations by user votes. Way back when the predefined destinations was asked for (of SE) by the moderators based on the presence of a large number of successful migrations initiated by the mods. (Successful meaning that the migrated question is not rejected/closed by the target community.) Currently MO is not a predefined migration destination on MSE. So any migration from MSE to MO must be done by moderator. $\endgroup$ Jul 6 at 13:30
  • $\begingroup$ I usually edit in links to comments, but in this case perhaps you intentionally did not name the someone who pointed out that Googling would have been a good start. Would you mind if I edited in a link? (I can't find the comment by @AndrésE.Caicedo that you mention, so can't link to it; I am referring to the other comment you mention.) $\endgroup$
    – LSpice
    Jul 7 at 19:54
  • 1
    $\begingroup$ @LSpice Links added. $\endgroup$ Jul 7 at 20:14

You must log in to answer this question.

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