I want to ask about the general appropriateness of questions about the practice of being a research mathematician, as opposed to directly about mathematics itself. To give some examples of what I mean, the page of the top 15 questions by votes includes a questions about how to referee a paper, http://mathoverflow.net/questions/36596/refereeing-a-paper-closed; a question about which journals publish expository work, http://mathoverflow.net/questions/15366/which-journals-publish-expository-work; and a question about advising, http://mathoverflow.net/questions/62972/resources-for-mathematics-advising. In the last week we've had a question about whiteboard markers, which was closed, and a question about the arXiv, which was controversial for reasons including (but not limited to) questions about whether it was on topic.

These are all questions of interest to many research mathematicians. (Of course, most things of interest are also of interest to research mathematicians, but these are of interest to research mathematicians as a particular part of being a research mathematician.) These seem to fall into a grey area in the FAQ---the FAQ says "MathOverflow's primary goal is for users to ask and answer research level math questions", which doesn't include these, but also includes various categories of explicitly forbidden questions, none of which include these.

Perhaps there's simply been a decision about how to handle these questions (though if so, I'd suggest that the FAQ be updated to include that, since the comment threads don't reflect an existing consensus on the issue). But if not, perhaps there should be.

These questions seem to have been more accepted in the earlier days of MO, and of late there seems to have been a turn against them. I admit that my motivation for bringing this up is that I'd like to see such questions explicitly allowed when they fit other criteria for being good questions. (Given the danger for some such questions devolving into discussions, or being stalking horses for starting a discussion, they might appropriately be scrutinized more closely than clearly math questions.) But I think such questions are beneficial to many users of the site (in as much as most users are research mathematicians), and (when appropriately tailored to be primarily about mathematics, rather than academia in general) good for the site (in that they may attract mathematicians in subfields underrepresented on the site, who might then start asking questions about their areas).

That said, if the conclusion is that such questions not be allowed at all, I think we'd still be better off reaching a decision on that, instead of rehashing the dispute every time someone asks such a question.