I have a question regarding "trivial" questions which sometimes appear on this site. By "trivial", I mean they don't require more insight than to "look in the manual". These usually aren't good candidates for a Q/A site, because they can easily create a flood of almost-duplicate questions, none of these helping any future user.

A good solution in some cases is to create a generic question, making all those real duplicates of that generic question. However, it is not always possible, and even when it is, I understand this site doesn't wish to become a parallel manual for common packages.

Before I mention the question I have in mind, disclaimer : "it's in the manual" isn't always an useful answer, because some manuals are more helpful than others, provide more examples, etc. But in that case, we should expect the asker to show that he has browsed the manual, and even if possible, to show the exact part he doesn't understand.

That's the same idea as on forums, "google is your friend": sometimes finding appropriate keywords for a google query is a problem by itself.

But once we took care of these problems, some "really trivial" questions remain : the example I have in mind is this recent question. As you can see, the needed work from the question's code snippet to a real answer is... very small.

I voted to close this question as "too localized", but I was really unsure:

• in some sense it could be a duplicate of a more general question about tikz styles (however, I couldn't find any such duplicate, so that settles it)
• in some sense it's not a real question... because it's hard to know what is asked, what the OP misses to transform his code to an actual answer
• in some sense it's indeed too localized, because the answer won't probably help anybody (but the question could!)
• and finally, don't forget that maybe the asker really wants to learn something, probably that "trivial" point he is missing in his code (maybe the fact that "rectangle" is the default shape?)... in which case the question should be kept open.

More specifically, about the "too localized" flag, since it is the one which "won", when I read its description:

This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet.

I can't say how it applies to the question: it is neither specific to a localization or a time-frame, and the problem isn't particularly "extraordinary narrow". However, once again, we can't really make the "Duplicate" or "Not a real question" descriptions fit there either, in my opinion.

So my question is : should such questions be left open by default? Or what is the appropriate closing reason to choose?

## Related questions on meta

In my opinion, as long as a question is searchable in its current form, and there is a chance that another user searches for the question, it is not too localised. Localised are questions that end up being typos, missed braces or dollar signs, that are solved in the comments.

If a question is about a "rectangle", but the "rectangle" being irrelevant, it might make a valid duplicate of a question about "ellipse" if "ellipse" is irrelevant, the relevant parts are the same and the answers to the other question are written in general enough way. It that case, the "ellipse" question can be linked, and it might be a good idea to add a comment:

The solutions there apply to rectangles as well, the issue is completely the same.

Then the "rectangle" one can be closed as duplicate. This minimises the noise in the sense of identical answers, yet keeps all the relevant information in place.

Remark: This is exactly the opposite thing to "same disease, different symptoms" case, where I'm strongly against duping since it can be very confusing.

• Also, about your interpretation of the "localised" flag, it is indeed valid, but wouldn't it then be a good idea to mention it more clearly in the description of the closing reason? – T. Verron Feb 21 '13 at 7:18
• @T.Verron The TL description starts with: "This question is unlikely to help any future visitors." I guess that it is quite precise and in agreement with what I whote here. The question can be helpful to another visitor, because he might want to change the style of rectangle. – yo' Feb 24 '13 at 14:12

My preference is for creating questions that tackle the same issue but are asked in a cleaner and slightly more general form and then closing the old trivial question in favour of the better trivial question. I don't think "triviality" is such a big issue, if we can pin down questions of the same sort.

I'm not a Tikz person, but I guess a question along the lines of "Is there a simple way to make a scheme for this kind of basic shape?" might be the kind of slightly more general question that fits here.

In my opinion, no question is too trivial or too complicated. This is a site for newcomers as well as experts.

You may ask the opposite question: Shall we vote for closing questions as too localised, when an answer obviously is useful only for a handful of very skilled users? Aren't questions regarding LaTeX output routine or advance TeX programming very localised?

In my opinion, the main problem is that some of the users here are too ‘trigger happy’; they vote for closing a genuine question because it has the same answers as other genuine questions. This trigger happiness hinder answers evolving into better and more useful answers, because few people answer old questions that already have accepted answers.

No question is too simple, no question is too complicated!

• I would tend to agree with your point in general, but there are definitely cases in which the question is so "simple" [simple isn't really the word I'm looking for here] it becomes unhelpful : the answerer won't add anything to the global knowledge of the site, he will only copy/paste elementary parts of the manual, or do some trivial (maybe long though) work the asker was too lazy to do by himself. We could keep these questions, but wouldn't they then hide the "good" questions and answers in a mass of hardly helpful questions? – T. Verron Feb 25 '13 at 12:39