Here is the longer version of the title:

Which is more appropriate: to "risk" asking your question directly on MO, or to check the appropriateness of your question on meta in advance?

From time to time, there is a meta-question evaluating the appropriateness of a particular question before asking the question on MO. Personally, I don't think it is a good thing to do. However, reading some previous post on meta, it seems that there are different opinions on this issue. Is there any general policy?

PS. Depend on the answer(s), I'll decide about my recent question on MSE :)

PPS. Upon Carnahan's suggestion (below), I am going to elaborate on my opinion. First of all, here is what meta is for (according to help center)

asking questions about how the websites work,asking questions of the community,posting bugs,suggesting improvements,proposing new features

To be honest, I cannot realize which one of these supports pre-meta-checking of a question. The help center aside, my personal opinion is: it is better to avoid having a mini-Mo inside meta. For example, Just imagine the questions with no answers (or "bad" answers) on MSE were to be migrated to meta for pre-appropriateness checking for MO, what would happen to meta?

  • 1
    $\begingroup$ Instead of making an isolated statement of opinion, could you elaborate with a reason why you don't think it is a good thing to do? In the absence of qualifiers, it sounds like you are saying that it is never a good thing to do. Is that the generality you intend? $\endgroup$ – S. Carnahan Oct 28 '13 at 13:48
  • $\begingroup$ @S.Carnahan I am going to add something to the body of the post since my answer to you is too long for a comment. $\endgroup$ – Amir Asghari Oct 28 '13 at 14:42
  • $\begingroup$ Amir, to answer one of your questions, I think pre-asking falls under asking questions of the community. As for the imagined scenario, the fact is that questions aren't migrated from MSE to MO meta. And I don't imagine they ever will be. $\endgroup$ – Todd Trimble Oct 29 '13 at 10:53
  • $\begingroup$ @ToddTrimble Of course, the imagined scenario wasn't about a formal, automatic migration. It is about considering that choice for the asker of the question. $\endgroup$ – Amir Asghari Oct 29 '13 at 15:07
  • 6
    $\begingroup$ I am note sure this question is appropriate for this site. It should be posted on meta-meta-mathoverflow. $\endgroup$ – Joël Oct 29 '13 at 23:55
  • $\begingroup$ @AmirAsghari Ordinary users cannot migrate questions from MSE to MO, only moderators at MSE can. And we are quite conservative in that we only do this if we deem the question to be well fit for MO (and the user requests it). I can't remember any soft question having been sent over and I don't think anyone of us would allow it. $\endgroup$ – Michael Greinecker Nov 6 '13 at 16:11

I think asking here is a very appropriate use of meta: one purpose of meta is to clarify or build a consensus regarding what is on-topic, how questions should be posed, etc.

Generally speaking, we really need to work on building a public understanding of this, because we are getting an awful lot of questions which are well below the customary standard of acceptable questions, and frankly this concerns me a lot. (It pains me to have to be so active in closing questions, as I feel this should be left mostly to the community, but it's getting to be a bit much these days.)

  • $\begingroup$ Of course, if there is a general consensus about using meta in this direction, I would be the first person who uses it. But, to be honest, I cannot see why the features of MO alone should not be considered enough for keeping MO "clean". $\endgroup$ – Amir Asghari Oct 28 '13 at 16:04
  • 2
    $\begingroup$ @Amir: There is a long history of usage of meta for this purpose dating back through MO.1 days. Of course, anyone who wants to just post at MO without pre-asking is welcome to do so, but for those people who would not rest content with having their questions closed without further discussion: the general idea is to keep MO for mathematics, and not to get into discussions about why a question was closed in MO comments, but rather to address them here. $\endgroup$ – Todd Trimble Oct 28 '13 at 16:45
  • 1
    $\begingroup$ By the way, I don't mind at all if someone wants to take a risk without pre-asking. I wouldn't pick between just asking versus pre-asking here as the more appropriate course. It's just the post-mortems that don't belong at MO (according to long-standing custom). $\endgroup$ – Todd Trimble Oct 28 '13 at 16:48
  • $\begingroup$ Convinced, and happy about this custom. Of course, I am going to use it asap :) $\endgroup$ – Amir Asghari Oct 28 '13 at 17:09

Apologies if this is repeating something historical. I am willing to give proper credit if someone else finds out where that credit goes. I'm not going to reread tea.mathoverflow anytime soon.

I think the true option is MathOverflow-checking. It seems reasonable to make use of the full database of questions by doing a match on questions that were risked and then rejected. Hopefully the platform can be tweaked to compare an asker's potential question to a "reject" database and raise flags if there is something similar in the potential question. (One could weight the rejects by age or some other criteria to keep from getting stuck in the past.) I think having the system raise an automatic flag before submission will be most effective in the bulk of cases; people who bother to read can then try meta-checking (at their leisure) or MathOverflow-risking (at their peril).

Of course, this means building a reject database with reject reasons, and using it carefully (displaying only the reasons as well as a similar text snippet), but it means that only the asker sees the reasons: no one else sees anything until the asker takes the risk (or the check).

Gerhard "One Of Many Different Options" Paseman, 2013.10.28

  • $\begingroup$ This is a really fantastic idea and I think it would be very effective. It doesn't even need to be anything too smart: If a flag is simply raised when the title contains phrases like "oldest" or "open question" or things like that I think that would drastically cut down on the amount of closing that needs to be done by a human being. $\endgroup$ – stankewicz Oct 29 '13 at 11:38
  • $\begingroup$ @stankewicz There is already such a thing. For example, if you start writing as the title: "what is your opinion ...", something pops up saying "the question you're asking appears subjective and is likely to be closed". However, I don't know how "smart" it is, and if there is any way to make it smarter. $\endgroup$ – Amir Asghari Oct 29 '13 at 15:40
  • 1
    $\begingroup$ Indeed. My suggestion includes populating the database with some fake failed questions which will trigger more flags, and hopefully do some automatic advice giving before the new question hits the database. Gerhard "Ask Me About System Design" Paseman, 2013.10.29 $\endgroup$ – Gerhard Paseman Oct 29 '13 at 16:02

As one who has tried posting a revised post first on meta (suggested by a user on the earlier post), I found that most comments were "why are you asking here instead of main?" So, for others, I suggest posting on main and discussing here. I will do the same if I ever have another question.

I ask few questions. When I ask them here it is because the answer isn't something from a book (I have already searched for those answers) but it might be something known to those who have experience in the subtle nuances in the patterns.

Whatever the outcomes of my questions, I can always find some positives to guide my mathmatical explorations.


@Amir, develop your judgment, have guts, and ask directly on the main Q&A MO--that's my view (not an advice). Asking first at meta may cause doubling the time and effort devoted to your question, at the expense of questions by other participants.


You must log in to answer this question.

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