Today I came across a question that wasn't actually a question. Rather, the author read an interesting factoid and turned it into a question, adding the factoid as an answer straight away. The reason I say it isn't a question is because there was never a problem or something that the author didn't know. In other words, there was an answer first and then a question was posted to give that answer a place on the site. At least, that's how I think things must have gone.
It's unclear to me what the policy on this is. On the one hand, the help center clearly states that self-answering is explicitly encouraged. However, I don't think this should be interpreted to mean that self-answering negate flaws in the question.
On the other hand, the help center also states that questions should be based on actual problems that you face. However, that's also a pretty slippery slope because while it seems clear to me that this question wasn't an actual problem, the author can always claim that it was, and there is no way to disprove that definitively. On top of that, it's the answer that makes it seem clear (to me) that this is the way things went, but as we have previously assessed, the answer itself isn't a problem.
I vaguely remember reading something about such situations in the past, but my searches on the topic have come up empty. Therefore, my question is: are such questions considered to be on topic on Stack Exchange?
x
?" is bad even if the self-answer is "It's 4, sincex = 2 + 2
", then the question was bad.