I seem to take the contrarian viewpoint here (by here, I mean the general sense of closing questions because they are particularly theoretic or advanced).
When I think of closing things as "too localized," three things come to mind.
I think of MathOverflow's "too localized" blurb in their faq. It says:
If your question was closed as "too localized," it was probably an explicit homework problem (or looked like one). MathOverflow is not the right place for such questions; they send the message that you want somebody else to do your work for you, and you're not even willing to ask it in a general form.
Needless to say, we do not close explicit homework problems, even when they are exactly requests that other people do your work for you. But the nice bit is the last phrase: "... and you're not even willing to ask it in a general form." This resonates within me.
There is a thread at SO about addressing What questions should be closed with reason "too localized"? (from 2009), with an awesome answer (written in 2011, much more recently) that gives a perfect (IMO) feel for "too localized questions in the form of an example. They said the 'cannonical "too localized" question is:'
Why is there a green Honda Civic parked out in front of my house?
This is too localized because:
- Who cares?
- Is it even still there? Go check.
- What are the chances that this question could ever be answered in a way that would benefit anyone else?
- Now is it there?
The answer was written two years later because suddenly people were applying "too localized" far too often (this is mentioned in the answer, and in other answers, and in the comments there)
Also from that question was another answer that said:
I always thought that meant that the question does not have appeal to a broad enough base of developers. It's like that guy at a company meeting with 500 people in it who will raise his hand and ask a really specific question about his project that nobody else in the room gives a damn about and won't get off of it. :) Something like ...
"Here at my company we use a library called AwesomeLib. But when I call DoGreatStuff on CoolObject in AwesomeLib, I get the following exception ..."
This answer got no upvotes, although there was a $+5$ comment that said "So basically, people are only to answer questions that that appeal to as many as possible, rather than to help someone solve their problem? I thought these sites were exactly for that: a place where a person can get help solving their problem."
This is sort of how I perceive this occasion: MSE is a community open to almost any question, and one of my guiding principles here is that people should be able to ask questions where they will get the best answer.
Earlier this year, the question Consider changing the wording of the "too localized" close reason was raised. Were I to give a concrete alternative to the "too localized" guidelines, I might recommend listening to what Shog9, the SE Community Coordinator, recommends:
This question is unlikely to help any future visitors; it is so specific to you, right here, right now that no one else will ever have the same problem or be able to benefit from its solution.
But to end, I should note that one cannot argue with Mariano's answer: people should vote to close whatever they think should be closed (although what they think should be closed is a somewhat fluid concept). But I emphasize temperance. If I were to try to close everything that doesn't interest me, or even that I can't imagine anyone actually being interested in, then set theory would be gone (I'm sorry Asaf) - but that's silly. Just because it doesn't interest me doesn't mean others aren't interested. As long as there is valid mathematical content that isn't a green Honda Civic on the front yard, then I see no reason to get rid of the question.