What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 128 Stack Exchange communities.

For example, a user asks "I'm trying to solve this threading problem but I don't want to use lock() or interlocked.increment and I want a one line solution".

It seems like this isn't going to be helpful to anyone else and is therefore "unlikely to ever help any future visitors" and would thus be appropriate to be closed as "too localized".

Right?

More generally the constraints a user adds to a question have artificially ruled out most or perhaps all sensible answers that would work to solve the underlying need. Clearly in the limit the question becomes absurd "I want to assign 5 to 'i' without using an 'int'" but at what point does a question become worth closing vs worth closing?

share|improve this question
1  
Nah... occasionally they can produce interesting answers... stackoverflow.com/questions/4679556/… –  Mysticial Apr 1 '12 at 7:06
2  
Guess I shouldn't have linked that... I hate to see interesting things get deleted. –  Mysticial Apr 1 '12 at 8:32
4  
@Mys: In general, I agree with you. But...what exactly was "interesting" about the answers to that question? Paul demonstrated that subtraction is the inverse operation of addition, but anyone who's taken elementary mathematics knows that. Victor demonstrated that you can generate unconstructive answers by reading a question overly literally (or not reading it at all). That doesn't help anyone do...anything. –  Cody Gray Apr 1 '12 at 8:48
1  
Sounds like an interview question. Quite often these questions are invented by smarmy senior devs who have a massive ego and want to see if the potential new hire is anywhere near as clever as them. –  slugster Apr 1 '12 at 11:41

3 Answers 3

I think this is a judgment call that smart people need to make in individual cases. You can find a handy population of those smart people by looking at everyone with 3k+ reputation—in particular, those users with close vote privileges.

If 5 people agree that the question is not useful and should be closed as "too localized", they will "vote their conscience" and the question will be closed. If there is substantial disagreement about the usefulness of the question, it will probably receive an insufficient number of votes to close and remain open. Moreover, the question can always be re-opened by other voters in response to later clarification and/or mere disagreement regarding the initial closing.

Basically, I think the dilemma resolves itself using the system we already have in place.

That said, I feel strongly that in general, the answer is yes. Most of these questions with ridiculous and arbitrary constraints should be closed as "too localized" and one of the close voters should leave a comment to explain more specifically why it has been closed. Something like:

The technology/feature/mechanism to do this already exists, is well-known, and has been extensively battle-tested. In fact, you indicate that you already know about it in your question, but have artificially ruled it out as a solution for whatever reason. These constraints effectively rule out the possibility of this question generating sensible answers that will be useful to anyone else in the future, and thus it will likely be closed as "too localized". If you can add more information to your question that explains why the existing obvious solutions are inadequate for your use, then someone might be able to provide a more detailed and useful answer.

share|improve this answer

This question has become more important now that the homework tag is deprecated.

Many homework questions have artificial constraints, and it's really unclear to me what we are supposed to do about this - are we saying that all such questions are inappropriate for SO, since in the "real world" (whatever that is) one would solve the problem without such constraints?

Although this would eliminate some of the "do my homework for me" questions, it may also eliminate ones that illustrate interesting alternative techniques, and seems rather harsh.

So I guess that the test is whether the answers generate interesting solutions with applicability beyond the individual question. This might mean that one can't really judge the question in isolation, only by giving it a little time and seeing what answers show up.

A second point is that I've seen a number of questions voted as too localized because they concern an obscure/minority language/system. This is concerning, as the more obscure the language, the more you really need a resource like SO to tap into whatever experience is out there! A question addressing a few dozen people may be more valuable than a question addressing thousands of people, if it solves a thorny issue for that small community.

share|improve this answer

If the question really seems too localised but is also, once the specific, narrowing constraints are removed, not a duplicate of an existing and well-answered one and may potentially draw good, really useful answers, then, in my opinion, it could be edited so as to soften the constraints. I mean, instead of ‘I (absolutely) do not want…’ (e.g. ‘…to use this or this, and instead want the solution to be such and such’), it could be re-rendered as ‘I would really prefer...’.

This way potential answerers would be less likely to consider the question too localised and, therefore, more willing to offer answers that might be useful on a larger scale, even if not solving the problem precisely as asked. After all, it might turn out that what the author of the question is after either can't be solved at all or is inefficient compared to other solutions (which might also be explained in an answer accordingly, again, for the benefit of a wide audience).

Alternatively, instead of actually editing the question, one could simply suggest the original poster to do that themselves, possibly also explaining the reason for such a change and potential (negative) effects (like the question getting closed as "Too localised") for the asker if they insisted on sticking to their version of the question.

share|improve this answer

You must log in to answer this question.

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