39
$\begingroup$

When I ask questions, it's always for the same reason: because I'm curious about the answer. It's not like my life depends on it, it's just a random question I asked myself while thinking about a certain topic and it doesn't usually have a definite origin. It just pops up into my brain, so I think about it for a while. When I can't come up with a satisfying answer, I ask a question here. It happened that the question was closed because I didn't write where it came from and why it was relevant to me. Well, it came from my mind, completely randomly, and it wasn't really relevantto me, it's just curiosity.

So my question is, what should I write in these cases? But most importantly, why are we always required to write why the question is relevant to us or where it came from, when in some cases this is completely irrelevant to the answerers?

$\endgroup$
6
  • 14
    $\begingroup$ It's part of giving context. Knowing the context of a problem helps answerers know what level the asker is coming from, what sort of answer will work best with them, and often whether or not the question would be expected to even have an answer. $\endgroup$
    – JonathanZ
    Commented Jul 5 at 13:55
  • 23
    $\begingroup$ You're not required to write why the question is relevant to you. You are required to post more than a bare statement of the problem, to give context in one way or another. Writing why the question is relevant to you is just one way to give context. $\endgroup$ Commented Jul 5 at 21:54
  • 6
    $\begingroup$ I have seen statements like "I think this is a natural/interesting question which is of general interest to the math community." or "I came across this question while working on ___, and am asking because of curiosity." be successful in place of absolutely nothing. If you're asking out of curiosity, or because you simply think the question is interesting, you can often just write that, though it is best to expand somewhat on these sorts of statements if you can. $\endgroup$ Commented Jul 6 at 17:38
  • 6
    $\begingroup$ As an example for @JonathanZ's comment, I've essentially only asked one question in MSE, which was something that arose in my random reading of old math literature. The arithmetic result I came across seemed interesting to me and I played around with it a little before posting my question. To make sure I sufficiently captured the interest of someone who could possibly help (something I actually wanted to do), I included a fair amount of context. That is, you can also think of context as advertisement for your question. $\endgroup$ Commented Jul 6 at 20:19
  • 1
    $\begingroup$ It's frequently the people who have little experience actually asking questions that have the strongest ideas about how one should ask questions. I check profiles frequently so now I can usually tell ahead of time which ones come from a "theoretical" approach to question asking and which come from folks actually asking the questions. $\endgroup$
    – uhoh
    Commented Jul 12 at 1:03
  • 1
    $\begingroup$ Whatever the answer is or isn't, I can certainly say as a casual reader that if such a policy truly exists, its application is absurdly inconsistent. The evidence: just read all the upvoted comments on the front page, and see how many of them say why the question is relevant. $\endgroup$
    – AakashM
    Commented Jul 12 at 12:41

2 Answers 2

28
$\begingroup$

This isn't about why the question is relevant to the asker; it is about why the question is relevant to the community.

This site isn't a tutoring site, nor is it a site for obtaining answers to any random question which pops into the head of any given person. This site is a repository of questions and answers—note the emphasis on the word "repository".

The goal of the StackExchange network is to build a searchable database of questions and answers which, in the vast majority of cases, solve user's problems via search. The ideal interaction with an SE site is that a user runs a search through some search engine (e.g. Google, Bing, etc.) and are served up a page from SE somewhere in the top results. In the best of all possible worlds, that user never feels compelled to create an SE account, nor to post a question.

This means that questions posted on SE sites are, in principle, supposed to be of interest to a wider audience. A question that someone just comes up with, off the top of their head, is unlikely to be terribly interesting to anyone other than the asker. So if one wants to ask a question, it is necessary to explain why anyone other than the asker should care about the answer to that question. This helps search engines to list the question correctly when others ask, and it provides details to potential answerers who might be able to further narrow down (or expand) the context.

Questions asked here should be of interest to a wider audience, and it is the job of the asker to explain that relevance.

$\endgroup$
22
  • 8
    $\begingroup$ Yeah, thanks, I get it now. I didn't think of SE that way, my bad, you actually changed my view completely, thank you. The strict moderation on duplicates also makes much more sense now. But to be honest, I can't think of a mathematical question that is inherently irrelevant or useless. A person might be wondering the same thing in ten years and will find the answer. So another question that I have now is how can a question be irrelevant? $\endgroup$
    – Elvis
    Commented Jul 5 at 14:07
  • 12
    $\begingroup$ @Elvis No one really cares about evaluating a definite integral using integration by parts for the umpteenth time time (homework problems are generally not relevant). No one wants to see your incorrect proof of the Riemann hypothesis or the twin prime conjecture. No one is going to be interested in something that came to you in a dream last night. And so on. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 5 at 14:09
  • 1
    $\begingroup$ Great answer. But I'd like to add as a content that " In the best of all possible worlds, that user never feels compelled to create an SE account, nor to post a question." is not the view of the web site owners - they measure their value by things like number of new accounts or posted questions. So new users often don't get this more accurate understanding until they interact with the community for a bit. I think it can lead to some unnecessary friction sometimes. $\endgroup$
    – JonathanZ
    Commented Jul 5 at 17:49
  • 4
    $\begingroup$ @JonathanZ I don't speak for the current owners of the site---they are more than welcome to send in a CM or some other staff if they feel the need. The founder felt that the best possible outcome is a passive search. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 5 at 17:52
  • $\begingroup$ @XanderHenderson - Yes, I assume you, as a mod, have to consider what role you post under. I should have been clearer that my comment was my personal observation, and not something that you should add to your answer. (I tried to edit my comment in time to reflect that, but got trapped by the edit window :-(. $\endgroup$
    – JonathanZ
    Commented Jul 5 at 17:57
  • 1
    $\begingroup$ So does this mean I can't post an interesting but random and quite specific question here like this? $\endgroup$
    – pie
    Commented Jul 6 at 17:40
  • 5
    $\begingroup$ @pie What about my answer suggests that you cannot? $\endgroup$
    – Xander Henderson Mod
    Commented Jul 6 at 17:43
  • 2
    $\begingroup$ @XanderHenderson "nor is it a site for obtaining answers to any random question which pops into the head of any given person. " and "This means that questions posted on SE sites are, in principle, supposed to be of broad interest to a wider audience. A question that someone just comes up with, off the top of their head, is unlikely to be terribly interesting to anyone other than the asker." Because I can't see people look for question like these in a google search or even came across question like these at all. $\endgroup$
    – pie
    Commented Jul 6 at 17:45
  • 8
    $\begingroup$ @pie You are, I think, ignoring the rest of the answer. It is the job of the asker to explain why their question is interesting. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 6 at 17:47
  • 9
    $\begingroup$ @pie Your question did not remain a random one popping into your head, once you made an attempt to make it interesting to everyone else. You wrote code, found something interesting happening that you could not explain, and posted it. It was then accessible and more interesting to a wider audience, and so it was received well. $\endgroup$ Commented Jul 7 at 5:10
  • 4
    $\begingroup$ I would also add that the requirement is very easy to fulfill. 20-40 words of explanation generally suffices, so the requirement isn't very onerous. $\endgroup$ Commented Jul 11 at 7:28
  • $\begingroup$ I think a more precise statement may be that questions are better ceteris paribus if they are of interest to a wider audience. For example, a question is not worse just because it's in a less mainstream branch of math. It's just that if you can make a question have broader appeal via minor changes in presentation, it becomes better as a result. $\endgroup$
    – user10478
    Commented Jul 11 at 17:37
  • 3
    $\begingroup$ @N.Virgo (1) The profanity is unproductive, and is borderline with respect to a moderator taking action (since I am engaged in this discussion, I will leave it to other moderators to make a final decision). (2) A "wider audience" might still be a small audience, but if the only person who might ever have any interest in the question is the person asking it, then it doesn't belong here. The job of the asker is to explain why anyone other than the asker should care. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 11 at 17:41
  • 2
    $\begingroup$ @XanderHenderson Anyone who feels they can define and constrain Stack Exchange in a few simple words doesn't really appreciate the extent of what Stack Exchange is. Yes it is a repository; that is indeed one thing that it is. It is also a (wonderful, valuable) community. It is also resource for getting answers personally. It is other things as well. cf. uhoh's lemmas The benefits of participating in SE to thousands or perhaps millions of people last a lifetime - even if the repository were erased. $\endgroup$
    – uhoh
    Commented Jul 12 at 0:47
  • 2
    $\begingroup$ @uhoh When I write meta answers, I try to be relatively unsubtle. While I am willing to hedge a bit, I would prefer that answers read in a somewhat more absolute way, as this prevents a lot of rules lawyering. The kinds of questions being discussed here are already a tiny minority of the questions asked on the site, and we are debating edge cases within that minority. I am willing to hedge a bit, but I also don't want to leave too much room for ambiguity. For the purposes of a meta answer, the primary goal of the site is to create a repository. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 12 at 2:20
15
$\begingroup$

I strongly disagree with the accepted answer (at least as it was formulated when I wrote this).

On a pragmatic level, it's certainly true that questions that appeal to a broader audience tend do be better received in general, for reasons that I think should be fairly obvious. But I very strongly disagree that this notion of "broad interest" should become a criterion that questions must obey.

The reason is that such a criterion is in direct contradiction with the idea that the site should serve as a repository of knowledge, expressed in the same answer. The value of a repository of knowledge does not lie in questions of "broad interest". It lies in questions of specific, niche interest that are phrased precisely and organised well. The whole point of being a repository of knowledge is that when some other person comes along in the future who happens to have the same specific, niche question, the answer is there and can be found. If questions are closed on the basis of not being of "broad interest", then those answers will not be there and the entire point of being a repository of knowledge is lost.

For the site to be useful as a repository of knowledge, a question should be phrased in such a way that if someone has the same question in the future, they will be able to find it. Mathematical context can be important for that, e.g. stating definitions etc. But personal context, i.e. talking about where the question came from, what the user needs it for, places they've already tried to look it up, etc. is often irrelevant noise. It can be relevant sometimes, but requiring it in every question is a mistake.

In other words, I think it should be completely and utterly fine to post questions that just came into your head, and the only thing they should need to include is the mathematical content of the question. This always used to be fine. The recent tendency to move in the opposite direction is counterproductive and should be reversed.

$\endgroup$
10
  • 3
    $\begingroup$ You're right, the only thing that helps make a question searchable are its statement/tags and the presence of keywords/key equations, and if there are none then in principle a question could be a bare problem statement and still have optimum searchability. Why I think that other content is added is (1) sometimes the user writes something they think can help answer the question, but could easily be superfluous, or (2) bare problem statement questions seem to indicate users trying to get homework done etc. which is a moral issue for answerers. $\endgroup$ Commented Jul 11 at 8:54
  • 3
    $\begingroup$ I think that you have focused too much on a single word (which I have now removed from my answer). A repository of thousands of questions which are of no interested to anyone is useless. A question on the SE network should be of interest to more than just the one person who answers is. If the only person who is interested in the answer to the question is the asker, then the question probably isn't a good fit for this site. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 12 at 2:22
  • 3
    $\begingroup$ I also note that, while you start by saying that you strongly disagree with me, there is almost nothing in you answer which actually contradicts what I wrote (other than your objection to a single word). I never said that a question must be of interest---I said that "in principle, [questions are] supposed to be of interest". I never said anything about mathematical or personal context (indeed, I largely agree with you that "personal context" is generally not very useful). $\endgroup$
    – Xander Henderson Mod
    Commented Jul 12 at 2:26
  • $\begingroup$ Also, keep in mind that my answer is a response to a question about why context is important. Context takes a random question that just popped into someone's head and is of no interest to anyone except the asker, and transforms it into a question which very well might be interesting to another person. Again, despite you telling me that I am full of shit, you don't actually seem to be disagreeing with much of what I wrote... $\endgroup$
    – Xander Henderson Mod
    Commented Jul 12 at 2:28
  • 3
    $\begingroup$ @XanderHenderson I'm sorry you took my answer differently from how it was intended. This seems to be partly due to my use of language in another comment, which I didn't put much thought into and I really do apologise for the impression it gave. I'm reacting not so much agains your answer specifically, as against a general recent tendency to VTC questions that are expressed concisely, even if they are mathematically interesting and contain all relevant mathematical context. $\endgroup$
    – N. Virgo
    Commented Jul 12 at 2:36
  • $\begingroup$ It is quite a subtle thing. I would say that a question like "how do I solve $x^2 + 3x - 4 = 0$?" is lacking context (and more importantly is probably a homework question), mostly because we don't know what the asker does or doesn't know about solving quadratic equations. But a question like "is it true that every continuous map preserves open sets?" already contains all necessary context in that sentence alone and should be left open (assuming it's not a duplicate), since there is no real ambiguity about what the asker means or what would count as an answer for them. $\endgroup$
    – N. Virgo
    Commented Jul 12 at 2:41
  • 4
    $\begingroup$ Moreover, for the latter question it's very likely that someone in the future would wonder the exact same thing, whereas for the former it's rather unlikely that someone would want to solve the exact same quadratic, unless they happened to be assigned it as homework. My point is that all these facts are obvious just from the question statement alone, assuming you know some basic algebra and point-set topology. In the case of the topology question it doesn't make sense to force the asker to "explain why it's relevant", but all too often recently I see such questions closed for that reason. $\endgroup$
    – N. Virgo
    Commented Jul 12 at 2:45
  • 2
    $\begingroup$ @N.Virgo I actually don't agree that all of the context is provided there. In a topology class, a student is expected to start from the axioms of topology and work forward, so "open sets" are a very basic and fundamental object. However, the same question might come up on a real analysis class, where "open sets" are defined in terms of a metric. Someone in a real analysis class is not going to be well served by the answer provided to someone in a topology class. $\endgroup$
    – Xander Henderson Mod
    Commented Jul 12 at 2:54
  • 1
    $\begingroup$ @XanderHenderson you're right, I was just thinking the same thing. But the example can be massaged a bit to get rid of that, e.g. "is it true that every continuous map between topological spaces preserves open sets?" It was a bad example because it lacked an important piece of context, namely that I meant it as a point-set topology question rather than a real analysis one. I guess all I'm saying is that this is the kind of thing "context" should mean, as opposed to other things people often ask for, such as "which textbook did you get this from?", "what do you need this for?", etc. $\endgroup$
    – N. Virgo
    Commented Jul 12 at 3:03
  • $\begingroup$ Thanks for your answer ! Really needed. I share the exact same opinion. "Repository"+ "searchable" does not require to include all level of motivation/context ! $\endgroup$
    – Olivier
    Commented Jul 12 at 7:43

You must log in to answer this question.

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