I understand that once upon a time, there existed on the menu of possible reasons for closing a question one that was labelled too localized. While the label may not have been perfect, it was generally understood that the reason was supposed to be invoked when the question concerned highly idiosyncratic use of English language, which was confined to a particular person, or a particular organisation, or a particular text. The idea behind the reason was that answering such a question would not provide any general insight into English language and usage, and would thus clutter the site with something that does not serve its intended purpose.

I also understand that the reason was removed from the menu by the central Stack Exchange ‘headquarters’. I was not, however, able to find on the meta-ELU site any consensus on what was then supposed to happen, within ELU, with the questions of the kind that would have been closed as too localised before.

Several possibilities occur to me of what one might do with such a question.

(1) One may downvote the question and/or explain in a comment that it is too localised.

(2) One may answer the question, and make it a part of the answer that it is too localised.

(3) One may flag the question (or, if one has sufficient reputation, vote to close it) using whichever other reason on the menu seems the closest.

(4) One may flag the question as needing a moderator’s attention (the only available option that allows for an open-ended explanation of the reason), specifying that it is too localised.

None of these options seem satisfactory. The first one is unstable: even if some people choose it, chances are that somebody else will try to answer the question, as long as it remains open. The second one increases the clutter on the site; moreover it legitimises such questions and so encourages more of them, which in turn increases the clutter further. The third one is disingenuous, which is unfair to the person who asked the question, and may contribute to the impression that the stated reasons for closing questions are, in general, not the real ones. The fourth one creates unnecessary work for the moderators.

So what is one supposed to do with such a question?


One thing that I sometimes do is vote to close and then note, as a customized close reason, the fact that the question is too localized to be of any possible interest to anyone other than the poster. Clearly, if the goal of English Language & Usage is to build up a searchable database of well-researched, well-informed answers to questions of broad and enduring interest, it makes no sense to clog the site with questions involving unique factual scenarios that no one else will ever encounter.

The clogging becomes a problem when a site visitor tries to search for a particular word or phrase and must sift through numerous irrelevant matches to try to find a useful match. This is a tolerable burden when the extraneous matches are to questions that are of substantial value in their own right—but not when the matches are to questions that no one (including, at this point, the various individuals who originally posted them) cares about and that are just taking up digital storage space.

The removal of "too localized" as a standardized close option is, in my opinion, one of the worst decisions ever made by TPTB at Stack Exchange on the topic of closing questions. The fact that this valuable and relevant close reason was effectively replaced at EL&U by the deeply flawed "show research" close reason only makes matters worse.

  • Would you be so kind to tell me what you understand by "too localized"? I agree though that replacing it with "show research" to be somewhat questionable.
    – Lambie
    May 18 '19 at 17:36
  • @Lambie: Perhaps the most useful way to identify the sorts of questions that I think ought to be closed as “too localized” is to present examples of them from the EL&U archives. I offer five such examples below: a request for editing help; a request for a too-narrowly-defined single word; a question about the meaning of an obscure passage of text; a request for help with the wording of a marketing slogan; and a request for help in understanding the correct answers on a standardized multiple-choice test. In each case, I think, “too localized” would be a good close reason.
    – Sven Yargs
    May 18 '19 at 19:58
  • Example 1: noun/plural/single asks whether a specific sentence should use “index” or “indexes,” “was” or “were” and “that of” or “those of.” It might be possible to extract a valid general question of broad interest that would also address the heart of the poster’s question—but the poster’s question doesn’t ask it. It asks for advice on how to edit a particular question. The current close reason being promoted for closing this question is “Please include the research...”
    – Sven Yargs
    May 18 '19 at 19:58
  • 1
    Example 2: What is a word meaning “an oft used cliché or line used to defend oneself from negative criticism”? has drawn four suggested answers: rationalize, platitude, quip, and illogic/malreasoning/pretense/façade/futile remark/lame excuse. Unfortunately, each added criterion narrows the range of possible situations in which the requested word be appropriate, until finally the word would almost never be useful, which is why word doesn’t exist.
    – Sven Yargs
    May 18 '19 at 19:59
  • Example 3: The meaning of the sentence? asks for the meaning of “I think few will be found to gainsay the assertion that where he trod, his daughter need not be ashamed to follow”—a sentence in Florence Marryat, There Is No Death (1891). We might close the question “What does ‘gainsay the assertion’ mean in this sentence?” as general reference, but at least it might interest someone besides the poster. There is scant chance that anyone else will ever come to EL&U seeking an explication of the whole sentence.
    – Sven Yargs
    May 18 '19 at 19:59
  • Example 4: Make slogan sound native asks for help translating a company slogan from Russian into natural-sounding English. The post was closed (in 2011) as “too localized.” Today, we would probably close it either as “proofreading” (which it isn’t) or with a customized reason asserting that brainstorming effective marketing slogans (even within the narrow bounds of translation) is inherently off-topic at EL&U.
    – Sven Yargs
    May 18 '19 at 19:59
  • Example 5: Explanations for multiple-choice questions on a past standardized test asks for help in understanding two questions on a standardized test involving a comparison of two lengthy excerpts from texts about Linnaeus. Bizarrely (or maybe I should say, “Because no better close reason presented itself”), the question was closed for the “Please include the research...” close reason.
    – Sven Yargs
    May 18 '19 at 20:00
  • These examples would be more helpful in your answer. May 28 '19 at 10:13
  • @Araucaria: Thanks for your comment. I had wanted to keep my answer short (for once), but then Lambie asked a fair question that I thought deserved a thorough response. I may yet add the gist of my response to my Meta answer, but it takes me somewhat far afield from the posted question, since "too localized" doesn't currently exist as a defined close reason. Speaking of far afield, I ran across a reference to araucaria in one of the searches I was running a few days ago, I hadn't realized that your EL&U name invokes the monkey puzzle tree. Nice!
    – Sven Yargs
    May 28 '19 at 17:42

Your option #1 seems the best, but you need a tailor-made comment explaining why you voted to close. But first, think whether you could edit the question to make it more widely relevant, without grossly distorting the question. If this is not possible, then VTC with the tailor-made comment.

You say the downside of this is that others might answer the question, but this is true of any VTC, and it happens often.

I agree that the menu answer that seems closest is not fair to the OP, and teaches him nothing, particularly because the menu answer will probably be "opinion based" which, IMO, is overused already.

  • Did you mean #1 or #3?
    – jsw29
    May 17 '19 at 4:11
  • Oops -- I meant a hybrid of #1 and #3, VTC (#3) with a custom comment explaining the VTC (#1). Will modify the answer later -- no time now.
    – ab2
    May 17 '19 at 12:17
  • @Jsw29 In the meantime I think it should be noted that there's an "other" closure reason which allows you to write in whatever you want to be left as a comment explaining exactly why you are voting to close, that other people can see while they are voting to close. I think it's meant to be used for questions which are outside of the help center's scope though, due to the standard form "off-topic" message it leaves when it earns the most votes. That way we can close out of scope questions for more than just a finite number of reasons.
    – Tonepoet
    May 17 '19 at 15:20

If you really think a question is of such little use to anybody other than the poster that it is detrimental to the website, then I would suggest voting against it. We want to maintain the interest of our community by optimizing for pearls over sand and the voting system is balanced to accomodate for liberal votes against questions for this reason. The voting score reflects the community's aggregate interest in the question, and helps people just finding it to evaluate of how much interest the question is before they even click into it for more details. However closing questions is different, because if done without due discretion it can prevent a gem of an answer being contributed. In other words, it can optimize against both sand and pearls.

Beyond that there is not really much of an appropriate measure: Commentary is mostly reserved for constructive criticism which should lead to the improvement of the post. If the subject matter is inherently uninteresting, then no amount of feedback is going to improve the question, and if you have a suggestion to make the question more interesting then it would probably be better received if you focus on that instead of discouraging the reader by prefacing it with the narrative that the question isn't worth the trouble.

Flagging and close voting is mostly reserved for content that is problematic to the extent that it should be outright deleted from the website if it remains in its current state but too localized questions likely do not merit such extreme measures. It is not like we are a paper resource which is bereft of space to address every possible question, and it is part of our mission statement here at Stack Exchange. If a question is otherwise well-asked, and possible to answer in a constructive manner, it is effectively harmless and its removal only denies the people who are interested in the question the opportunity to address it. Something which absolutely needs to be kept in mind is that Too Localized was a network-wide closure reason which was not merely imperfect, found to be highly problematic. It was one of the closure reasons that was eliminated as a result of the war of the closes, because it covered too many vague circumstances. You could argue that if we bar questions that are "only relevant to a small geographical mass or specific point in time" that most of our questions dialectical or historic usage should be discarded, which is something I very much doubt that most of our current intended user-base wants.

The best reason we had for too localized is perhaps because some questions were of too little interest. However the problem with that is that it is too much of a judgement call to reasonably make with any degree of consistency, which is why a more restricted too narrow closure reason was rejected as a replacement. If somebody was interested enough to ask in the first place, then it isn't too much of a stretch to assume that other people may take interest in the same problem.

If other people do not take interest in the same problem, then the question will probably remain unanswered anyway, meaning that there is not even so much as a need to close the question. Moreover, these unanswered questions may serve to our benefit later if we are proven wrong and somebody comes and decides to answer it with a compelling answer, but if we close those questions prematurely, then we are denying the possibility of that even happening. This is all the worse if the answer reveals matters deeper interest.That is why we have an unanswered questions queue and possibly even the necromancer badge.

Think about it this way. Should we really have a closure reason that is simply a coded method of saying "this question bores me for no certain reason?" Honestly, I think not.

Now if we had a finite amount of space at our disposal, that could be harm us by interfering with our ability to address more noteworthy subject matter. However we are not so constrained. We are not like formal dictionaries and other early reference works, since we are not limited by how many pieces of paper can be bound together in a few volumes. A few such questions here and there doesn't really seem harmful, so long as we're not getting them so often that it noticeably detracts from the overall usability of the website, and if we are, we can probably categorize the questions more meaningfully.

Now if answering a question is effectively meaningless and risks doing more harm than good in its present state, I can understand closing it for that reason, but the too localized closure reason never really did anything to identify such a fundamental problem.

To put it another way, if I decide spend a few hours pouring through references and explaining that there is no such thing as a paraphrase bracket in the English Language of my own volition, is that really such a bad thing that you really have to take active preventative measures against it? That question went unanswered for months, and I genuinely doubt that it is a matter of broad interest that will make a meaningful impact on many people's lives since most people probably don't even think about paraphrase brackets. However, I had myself wondered about how to appropriately paraphrase before, and the few people who do take interest in the matter seem to appreciate my answer.

Besides that, if we expect to operate in a community of people who help each-other learn, then answering the boring questions some other people might have, where possible may actually inspire the people who receive constructive answers to pay the favor forward, and serve to exemplify how that can be done. It's also a very good way to potentially pay back the other members who have decided to help us with questions of ours that they may have also found to be inane.

Another concern is that there is already very little middle ground between the Primarily Opinion Based and General Reference criteria, and I don't think we really need to narrow that ground even further by taking a very severe approach against questions which can only be vaguely described as "too localized". Voting against a queston's score seems like enough of a corrective measure here.

Finally, we use the questions and answers format to suss out which questions are of interest, and need addressing, but there is a certain price to be paid for that. The unique appeal of a questions and answers website to questioners, who we certainly do need, is that they might be able to gain some personalized assistance. It seems obstructive to that purpose to just uniformly block questions with no discernable problem other than what may just be personal disinterest.

  • It is true that Stack Exchange does not suffer from a shortage of space, and that letting a localised question stand therefore does not affect how much space is available for other questions. The presence of these questions, however, still dilutes the site: they become a part of what future visitors will see when they browse the site. These questions will influence the visitors' impressions of what the site is 'all about', which will, in turn influence the questions that will be asked in the future.
    – jsw29
    May 17 '19 at 4:22
  • @Jsw29 Perhaps they might dilute the backlog, but to what extent, and would it really be by enough to be a problem? I would suppose that questions of broad interest would have a natural tendency to be asked more often, since there are more interested parties to ask, so there should be more good examples than bad. It should also be easier for people to devise the less creative search terms which bring them to questions of broad interest. Few people should click into questions of no real interest in the first place, and even fewer would vote for them, so the problem seems mostly self-correcting.
    – Tonepoet
    May 17 '19 at 6:06

You must log in to answer this question.

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