We've previously discussed the meaning of a downvote on meta, and I think that's worth reading before discussing anything further here.

I don't at all like that perfectly reasonable questions (especially from people new to MathOverflow) asking about acceptable behaviour get heavily downvoted. It's discouragin, and pretty rude --- no one is going to cleanly differentiate between a downvote meaning "the answer to your question is no", and "you shouldn't have asked this question".

There's no quick solution, as we shouldn't make rules about how people vote on meta. Here's a proposal:

On questions with clear yes and no answers, rather than downvoting the question to indicate no, give a short answer "No, you should not insult leprechauns on MathOveflow", or vote for the existing short "No" answer.

(Of course, the original poster could also provide two or more alternative answers themselves, if they are implicitly asking for a poll of opinions, rather than advice.)

  • 3
    $\begingroup$ Yep, exactly this is the typical Stack Exchange crowd behavior one can observe on the Meta part on many (in particlar the strongly by SE guidelines governed) sites in the network. Worst it gets on MSO, where no-nonsensical well thought out, and clearly formulated questions and suggestions get voted down like hell. There people simply do not discriminate between the legitimacy of the question and disagreement with what the question says or suggests ... $\endgroup$
    – Dilaton
    Commented Dec 16, 2013 at 1:31
  • 3
    $\begingroup$ Scott: Do you effectively suggest to make downvotes on meta have the same meaning as close votes? -- If yes, why not make a feature proposal to remove the then-redundant possibility to downvote questions on meta, and to make voting unidirectional (only upvotes)? $\endgroup$
    – Stefan Kohl Mod
    Commented Dec 16, 2013 at 10:23
  • 3
    $\begingroup$ While I agree with the general sentiment I would still like to point out that the question that prompted (for all I know) this post in my opinion does not qualify as perfectly reasonable and I could well imagine that somebody downvoted it (while I did not) since it was not such a good question to begin with. That being said I still agree that it would be somehow nicer to leave non-offensive questions by newcomers in the non-negative score range. $\endgroup$
    – user9072
    Commented Dec 16, 2013 at 17:23
  • 3
    $\begingroup$ @StefanKohl, no. A downvote just means you think it's a bad question or that there's something wrong with it, as usual. I don't think that's synonymous with a vote to close at all. $\endgroup$ Commented Dec 16, 2013 at 20:46

2 Answers 2


I agree with Scott. .

  • 10
    $\begingroup$ Something about voting at meta has never felt quite right to me (I sometimes miss the set-up that we have now relegated to tea); it just seems a little weird to have every single discussion subjected to a popularity contest. That said, +1 to Tom's answer. :-) $\endgroup$
    – Todd Trimble Mod
    Commented Dec 16, 2013 at 2:05
  • 3
    $\begingroup$ +1 to Todd's comment too. Issues shouldn't be reduced to a yes/no choice. But Scott's question seemed to call for it. $\endgroup$ Commented Dec 16, 2013 at 2:18

Thanks that you brought this question up again. The only answer for the question that you have mentioned was from François. And, if I understood his argument correctly, the main reason behind up-vote/down-vote system on meta is to bring the valuable questions to the top of this list. If we accept this, we can go with Stefan's suggestion and make the voting system on Meta unidirectional (only upvotes) and still the top questions would be rightly at the top of the list.


You must log in to answer this question.

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