I’ve reviewed a fair slug of the discussion here on the meta about so-called problem-statement questions (PSQs). I know that some of us disdain them and that others defend them. I also know from experience that on the main site there’s precious little guidance about them. And I know that others here on the meta have pointed out this situation and proposed that we remedy it. I agree that it cries out to be remedied.

Trying to put myself in the shoes of a new user, I did get to “What types of questions should I avoid asking?” where we advise, “First, make sure that your question is on-topic for this site.” Following that link leads to “How to ask a good question”, and there I find a 10-item bullet list, in which the first item reads, “Provide context, $\,$ include the source and motivation for your question, $\,$ and avoid "no-clue" questions.”

Now, some have argued here that it’s the absence of context that’s at the heart of their objection to PSQs. But even if a user follows that link they won’t find any explicit mention of PSQs themselves, whether under that moniker or any other description. And anyway, by now we’re awfully deep into a sequence of clicks. From the UX perspective, how far (I ask rhetorically) do we expect users to burrow in order to learn of a class of questions that may bring disapproval down upon their heads?

So I offer a simple suggestion, which I hope is uncontroversial and also at least somewhat effective. Let’s tweak the first item in that bullet list to read instead, “Don’t just state a problem. Instead, provide context, $\,$ include the source and motivation for your question. $\,$ Similarly, avoid "no-clue" questions.”

  • 3
    $\begingroup$ I suppose the placement of "Provide context" first in a list of items about how to improve PSQs mitigates the lack of "any explicit mention of PSQs themselves". In fact I avoid using the abbreviation PSQ because it has no consensus meaning, and I prefer to say "bare problem statement" in combination with a need to supply lacking context by way of editing the body of the Question. $\endgroup$
    – hardmath
    Commented Jun 22 at 18:36
  • 1
    $\begingroup$ See also this article on MathSE protocol $\endgroup$ Commented Jun 24 at 3:20


You must log in to answer this question.