This post was reworked by me to more accurately convey my point. This may have rendered some comments obsolete.

This question has a follow-up here.

The problem
There are quite a few questions wich could be answered by putting the question (or its title) into a google query and taking the first result. [just-google-it questions]

The main characteristics of these are

  • The answer to the question is farily simple to obtain using google using the same words that appear in the question (So "How do I solve $\min \|Ax-b\|_2$?" doesn't meet this criteria because google will only find the answer given "ols" or "least squares" as additional input)
  • The question may tend to be very localized (there used to be a close reason for this)
  • The OP shows no effort at all (else he would have googled and found out the solution himself)
  • Canonical answers (those not adding additional aspects to the question) will be short and localized as well, so they are unlikely to be of use for future visitors or other users reading the question
  • Since google can find the answer, the question has a simple and concise formulation without the need of additional details

Questions I consider of the just-google-it type:

    The wikipedia article on eigenvalues clearly deals (even with both kinds of) multiplicity.

Such questions should be closed to increase the overall quality of Math.SE, however none of the current close reasons seems to be to the point about the problem:

  • Unclear what you're asking: Not really
  • Not about Mathematics (within the scope): The on-topic list doesn't exclude such questions.
  • Missing context: It's not missing context. The OP couldn't improve the question to meet the quality standards without asking a different (more general) question. See also What do we mean by "context"?

I see multiple options to improve the situation

Adding a new custom close reason
A possible close reason more applicable to such questions would be one discussed here or possibly another all-new custom close reason along the lines of

This question appears to be too local to your specific problem to be of future use. Please consider asking a more general question, providing your specific problem as a motivational context.

Changing what is on-topic
This would require a careful rewrite of our on-topic guidelines to exclude such questions. This however needs a precise and concise definition of a just-google-it question.

What do you think? Is the "Missing context" reason sufficient for such questions? Do we actually welcome such questions?

  • 1
    I don't understand how this is not covered already by "unclear what you're asking" and "missing context". Could you give some particular examples of questions you would like to apply such a reason to? – Milo Brandt Mar 29 '15 at 16:19
  • 3
    @Meelo I think the "missing context" is just overused and some questions like this simply aren't missing context, they are either missing insight or missing effort or missing quality, but not context. – AlexR Mar 29 '15 at 16:22
  • 1
    @Meelo Also see your own meta post regarding the problematic name "missing context". A valid answer to my question would also be to suggest renaming the "missing context" reason to "lacking input / low quality" or whatever. Another choice would be to differentiate "low quality" (format / localisation etc.) and "lacking effort" (do my homework, prove this, solve that) – AlexR Mar 29 '15 at 16:27
  • @AlexR Ah; I'm inclined to think the reason we would close it would be because it is missing context. This doesn't mean the question is otherwise fine - the other issues seem addressed better by edits, comments, and downvotes. To me, at least, unless poor quality gets to the point of "I have no idea what you're asking", it's not a good reason to close. – Milo Brandt Mar 29 '15 at 16:31
  • @Meelo Low quality as in "What is $\frac12 + \frac13$?" isn't really missing any context. The question is very clear and so is the answer. But still it's not a question worth keeping on the site, because it's too localized. – AlexR Mar 29 '15 at 16:35
  • 1
    One could argue it is off-topic. In the same way that MathOverflow is reserved for research-level inquiries, one could argue that this site is reserved for questions that Google cannot readily answer. – Michael Grant Mar 29 '15 at 18:29
  • 3
    @MichaelGrant This would require a change in the on-topic section of the site. – AlexR Mar 29 '15 at 18:38
  • We are discussing a change here, aren't we, either way? ;-) – Michael Grant Mar 29 '15 at 18:39
  • @AlexR Sure, tact definitely matters. I would not want someone to think their question is stupid. Even if the question is readily addressed by a dictionary-level resource, nobody is born knowing the dictionary ;-) I think I would simply offer that Math.SE is not attempting to supplant more static online references. – Michael Grant Mar 29 '15 at 18:49
  • 1
    Note that what is just Google it for you may not be for someone else, especially for someone who is not fluent in English. – Brian M. Scott Mar 31 '15 at 9:37
  • 1
    @BrianM.Scott I am aware of this, wich is why I made the definition so that it requires google to find the answer using precisely the same formulation as the OP. ("copy the question (or it's title)") – AlexR Mar 31 '15 at 14:37
  • 1
    @Alex: I really don't think that someone who struggles with English should be expected to think of trying that; I don't use search engines that way, though I suppose that this could just be my age showing. And asking here instead of hoping for a lucky hit has the advantage that one can expect an interactive response if one struggles with the answer that one gets. Yes, I see the occasional question that makes me roll my eyes a bit, but I can't see getting all hot and bothered about them, especially at the cost of making MSE even more unwelcoming to newcomers. – Brian M. Scott Mar 31 '15 at 16:27
  • 2
    @BrianM.Scott I fail to see the connection between struggling with English and thinking about using a search engine. Furthermore, it is not at all clear that somebody that struggles with English is served well in the interactive, multi-agent and sometimes fast-paced environment of SE. (Where the served well is meant in comparison to consulting a more static source.) – quid Apr 1 '15 at 16:32
  • 2
    @BrianM.Scott the point is that putting a question into a search-engine to get an answer is not that language-dependent. To which kind of questions you will get an answer might well be language-dependent though. So to then think of typing the question in English into the search-engine is about as "unusual" as asking it in English on some internet-site. On the furthemore: for one thing, it can happen that some native-speaker will berate you for an idiosyncratic idiolect and related things. – quid Apr 1 '15 at 16:58
  • 2
    Sorry, can't resist: – conjectures Apr 1 '15 at 17:43
up vote 16 down vote accepted

If a question includes a precise statement of the question they wish answered, includes enough context for us to judge where the OP's misunderstandings may lie and what sort of facts may be obvious to them, and is actually about mathematics, then, excepting extraordinary cases (to which we have custom close reasons and moderators), it should stay. The worst case is that no one will want to answer it. The best case is that someone will provide an insightful or at least different way of looking at it.

The Q&A format has significant advantages over the longer forms you're likely to find by googling (e.g. Wikipedia articles). An article addresses a topic and may or may not answer someone's question - and, even if it does, it might not be a major point; it might just be a passing remark which does not satisfy someone unfamiliar with the material. In any case, it is likely to be buried within the article, meaning someone seeking an answer will have to dig before they find it (if it is indeed there). On this site, an answer can be pure useful information - the answer can be focused and thorough, and altogether, a much better resource for someone with the question. Moreover, it is not constrained by the overall purpose of an article - for instance, Wikipedia loves being formal to the point where it's less useful for people just learning. You'll find a lot more intuition and elementary explanations on this site - which can help the learner far more. Given the potential of our format, it makes little sense to restrict our scope based on what other places cover.

The situation you describe really seems adequately handled at the moment. In your example, it's clear that the question lacks context. You seem to suggest that we should close it, since if the OP made any effort, they would probably answer it - i.e. the effort is "trivial" in some sense. However, as it stands, what'll happen is that the question is (rightly) closed for lack of context and will remain so until it is deleted. If the OP had some legitimate confusion and later came back and put in the effort to improve the question, then we can reopen and answer it. There's no point in this process where antagonizing the OP by closing it for being too easy to answer via other tools would make a positive difference.

So, I suggest, for questions which can easily be answered by Google, but are otherwise fine*, the appropriate course of action is:

Leave a comment suggesting the OP google it, or pointing to some particular resources. Ask if the OP finds these resources unclear or what - i.e. ask the OP to add more context based on the fact that explanations are readily available.

If this does answer their question - well, that's great (and hopefully they'll do something like answer their own question). If it doesn't, then we get more insight into what's tripping them up. This way, we can lever external resources for exactly what they're worth - and minimize the effort we spend explaining things that the OP (after they do some research) already understands.

  • 2
    I have the feeling that "missing context" is used as a one-size-fits-all for "missing something". Could you tell me what context is missing for the question I linked, for example? – AlexR Mar 30 '15 at 1:11
  • 3
    @AlexR The measure I use is: "Is there anything in the post which helps me understand why the OP cannot answer their question?". The answer here is "No." - meaning that, were I to write an answer, I couldn't help them understand anything they didn't already - I'd just be doing some work for them. I'd feel differently if they'd provided some attempts (e.g. matrices they looked at). Closure under this reason makes sense because, without more, we cannot answer well. – Milo Brandt Mar 30 '15 at 1:16
  • What you describe isn't really the context of the question, it's the effort put into it before dumping it here. I think there should be this distinction between "put more effort into it" and "make it a question worth keeping". – AlexR Mar 30 '15 at 1:19
  • 2
    @AlexR The word "context" is unenlightening (as from my previous meta post), but I think "the effort put into it before dumping it here" is decidedly under our use of the word "context" -it helps us understand their difficulty. And if they added more effort, I think it could very well become a question worth keeping - even if I'm not sure what sort of effort they could do that wouldn't immediately solve their question (i.e. I think we should keep open the possibility that, if they had some particular confusion, the addition of that as context would improve the question, even if it is unlikely) – Milo Brandt Mar 30 '15 at 1:29
  • The policy is fine with me, but then we should re-think the wording of the particular close reason. I remember there has been a discussion about it, but I didn't find anything quickly. – AlexR Mar 30 '15 at 1:31
  • @AlexR I completely agree with you there. The only related post to that I could find was this (and I suppose mine, tangentially). You might as well start a new meta thread for it - I think it would be highly productive to have that discussion. – Milo Brandt Mar 30 '15 at 1:40
  • Ah thanks, the other linked one was the question I had in mind, but it turns out I thought it was a more elaborate discussion ^^ I'm too tired right now (3:42 AM here) but we should definitely raise that point. – AlexR Mar 30 '15 at 1:42
  • @AlexR "Could you tell me what context is missing for the question I linked, for example?": from this, for example the OP could have showed some attempt at solving the question, provide some motivation, tell us where the question comes from, indicate their own background, or give definitions; or also link to similar questions, for example. In fact it's not clear to me what context you believe is present in the question you linked. – Najib Idrissi Mar 30 '15 at 8:24
  • @NajibIdrissi So that "missing" is used as "necessary, but not essential"? What I mean by this is that the question at hand can be answered with any further information and the answer is likely to still help the OP gain some insight (although the specific user seems to have a general misconception about miltiplicities of eigenvalues, but what self-diagnosis could give us this piece of context?). – AlexR Mar 30 '15 at 12:48
  • @AlexR The point is that missing context makes the question of little value to a math knowledge repository, which is one of the goals SE is trying to achieve. – Najib Idrissi Mar 30 '15 at 16:43
  • @NajibIdrissi I think we both agree that such questions should be closed - it's the wording that stumps me. "Missing context" is an extremely broad reason, as can be seen from the elaboration you linked: Six different bullet points fall under the same category, whereas other close reasons have only a small and very clear scope. – AlexR Mar 30 '15 at 16:47
  • @AlexR This has been discussed ad nauseam: there isn't much room for site-specific closing reasons. It can only be listed in the "Off-topic" section, there is a character count limit, and AFAIK there can only be one. You could create a custom close reason every time, but that would be terribly annoying, hence the current situation. – Najib Idrissi Mar 30 '15 at 16:53
  • @NajibIdrissi There can be multiple site-specific reasons nested under "off-topic": Stack Overflow has five. – user147263 Mar 30 '15 at 16:54
  • @Woodface Oh, okay, I didn't know. Since it comes up rather frequently, maybe there could be a discussion on whether that list needs to evolve on math.SE. – Najib Idrissi Mar 30 '15 at 16:56
  • @NajibIdrissi See my suggestions above: "Adding a custom close reason"... – AlexR Mar 30 '15 at 17:49

Actually I had one of those questions sometime ago. I wanted to know how to prove the volume of a cone using triple integrals. Someone was nice enough to point me to a site I had not considered. I learned on that site and was able to help someone else in a similar thread.... my answer to that is now part of the community wiki.

I say, if you can't say anything nice, or the question is beneath you to answer then don't answer it. Someone will feel they have something to offer the OP and they will take the time to teach him/her.

Downvotes, and flags just make the OP feel like their question is invalidated and do not contribute to the OP's understanding.

  • Do you have some references to those posts? – Peter Mortensen Apr 6 '15 at 10:25
  • They are in my profile. You are welcome to surf through any of my questions/answers. – Chris Apr 7 '15 at 13:15

If the first Google result happens to be a Math.SE question, then obviously the proper thing to do is close the new question as a duplicate.

But more likely the first Google result is a Wikipedia article, and though Wikipedia's math content is probably better than its history and politics content, it's still a good idea to be wary and skeptical.

Scrolling just a little bit down on the Google results one might find a more credible source, like the blog of a renowned mathematician or course material for a university math course. Either of these, especially the latter, is likely to be unhelpful to someone who has jumped into the middle of a subject, bypassing vital background. That's the impression I get from Omar with his eigenvalues question.

Plus I don't think that's a "just Google it" question. Here's what I think is a true "just Google it" question:

What is the definition of a prime number?

Google will even go so far as single out one result to present as the answer to such a question.

If such a question does get asked on Math.SE and it's not a duplicate, then is it really that big an effort to write a short answer off the top of your head or copy and paste from the first Google result? If you want to be an abrasive jackass about it, you can include a link to, e.g.,

  • Note that neither do I want to be a "jackass" nor was Omar's situation visible at that point, because that was his first such question. – AlexR Apr 2 '15 at 5:58
  • 1
    Related to your suggestion about LMGTFY: What is our policy for “(let me Google that for you) links” ? – Martin Sleziak Apr 2 '15 at 9:42
  • @MartinSleziak In his answer to that question, Douglass calls such answers "disrespectful." That seems consistent whit how an abrasive jackass would behave. It doesn't sound to me like Alonzo is endorsing that sort of behavior. – user155234 Apr 2 '15 at 21:48

There already exist much better options to improve the situation.

Flag as duplicate

If Google answers it, then it's probably neither the first nor the last time the question has been asked here. And in the rare event there really is no duplicate of it, just answer the question and now you have something to flag as duplicate of when the question is inevitably asked again. (I kind of got the idea for this one from Meelo's answer to the question about the deluge of IEEE-754 questions).

Downvote it

Not that you actually need any encouragement to downvote, of course. Correct me if I'm wrong, but isn't there some mechanism where if a question gets enough downvotes it gets automatically deleted without needing a close reason?

Leave it alone

If I've learned anything from the Mathematics category of Yahoo! Answers is that there are a lot of people on the Internet besides me, people with more time and greater inclination than I to answer very basic math questions. Math.SE won't grind to a halt just because I decide not to aggravate myself with lazy askers one particular night.

  • 4
    My problem is not that such questions get answered, but more that they flush more interesting and higher quality questions off the front page, i.e. they constitute much noise wich makes finding good questions harder. As for downvote / close - the discussion was started because I'm not satisfied with the current close options. I want to give an asker enough feedback via the close reason that he / she can improve his / her questions in the future. – AlexR Apr 2 '15 at 6:01
  • I don't know about that, the front page seems to change awfully fast. I don't know about you or Bob, but it seems to me the front page is more filled up with questions on topics that don't interest me than questions that can be answered with a simple Google search of a basic term. I suppose I ought to be more interested in eigenvalues, but I would have skipped over Omar's question not because it seemed too easy but because it didn't interest me. – user153918 Apr 2 '15 at 14:03

You must log in to answer this question.

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