Ok, there's always a lot of talk about closing or not closing questions that show no research effort, but are otherwise good questions (or is it just me talking to myself again?).

Skip to after the line (to "I propose ...") if you agree and are feeling lazy.

On the one hand, the asker doesn't really deserve an answer, if that's at all important. That, combined with the fact that allowing it to get answers and stick around (by not closing and deleting it), even if downvoted, sets a precedent for such questions in future, by both that user (possibly at least until he gets a question ban) and others. This precedent wouldn't really be bad if it were only for otherwise good questions, but of questions lacking visible research effort, those just lacking research effort (and aren't otherwise really, really terrible questions) is most likely in the minority - the problem is that this precedent doesn't just get set for those questions, but in fact all questions lacking visible research effort.

On the other hand, it might be useful for future visitors.

I know - official policy dictates that we should just downvote (if we feel it's appropriate to do so) and leave open, but then there's certain seemingly identical cases where we should close it, which just leaves me generally rather confused, but anyway, that's beside the point.

I did have a question in mind, but, after thinking about it a bit, I don't think it's particularly useful (others might disagree), and every time I pick a candidate for this type of question someone points out that "Oh, but you obviously could've closed it as ..." and I'm like "That's not the point!" (ok, maybe this just happened once, or, well, a few times, but in the same thread, and now I'm scarred, forever fearing the "That's not a good example" answer, I feel my heart starting to race already...), but anyway, where were we? Oh right, my feature request.

I propose that we extend the closing functionality to allow users to vote to place a question in temporary 'limbo'.

I imagine limbo working in one of a few ways:

  • Exactly as the current close functionality - obviously the easiest to implement

  • Allow users to post answers, but don't display any answers on the question while it is in limbo - this one is starting to sound like a bad idea already

  • A combination of both - hide (temporarily delete?) the current answers, and don't allow any new answers.

If a question is placed into limbo, it will remain there for ... let's say a week, after which it gets treated as a new question (i.e. it triggers "x question with new activity" on ... ummm, say this page, which, despite the naming, questions with new activity doesn't trigger, only new questions - the comments and answers can still remain), except that:

  • There's a note somewhere saying it's been in limbo.

  • It can't get placed into limbo again.

  • Anonymize the question.

    The motivation here is that the user doesn't get any reputation for the votes on the question from that point onwards. The motivation behind that is that we can upvote these questions purely on the usefulness, ignoring the fact that it doesn't show any research effort.

    I considered a semi-Community Wiki, but this isn't ideal as the user most likely won't be particularly motivated to edit the post to clear up any ambiguity, so it might be better to anonymize the question as to indicate that potential answerers shouldn't ask for clarification using comments, but rather simply state their assumption in the answer. And maybe we don't want the user to be notified of comments or answers, and possibly get badges for a popular such question, but perhaps I'm just being mean now.

Questions placed into limbo should follow similar guidelines for being reopened if edited (that is, prior to the scheduled removal from limbo).

I intentionally left the process for limbofication vague, as that's not really the point (deja vu...), but one possible way it could work - a question with a total of 5 votes, any of which being either close or limbo votes either gets closed or placed in limbo, if 3 or more close votes, it gets closed, if 3 or more limbo votes, it gets placed into limbo. Then one would presumably be able to follow a similar voting system to the current one (5 votes) to move a question from being closed to being placed in limbo (and vice versa?)

As I see it, I think there's a sufficient combination of:

  • Otherwise good questions that get closed.

  • Otherwise bad questions that stay open.

  • Just, in general, a ton of questions that don't show any research effort. This is maybe not a problem as such, but it sets the precedent for other bad questions, as mentioned above, and it certainly greatly increases the probability of duplicates, for obvious reasons.

To justify this.

Despite my earlier phrasing (or how this may come across), I don't think this has much more to do with punishment than the current system, whether that be for the common good or bad - a question getting put into limbo, then removed from limbo as described will essentially be the same (for the asking user) as a question getting closed, then automatically deleted (well, automatic deletion isn't after a week, but, again, not the point).

Given that you'll most likely On the off chance that you think this is a bad idea, consider this a "We have a problem and something need to be done about it - here's my take" post.

  • 2
    I would be concerned with the bit about unclear questions. Unclear questions with guesses as answers are never a good thing; ideally the OP should be incentivised to clarify them Mar 23 '14 at 23:21
  • @RichardTingle Unclear questions should still be closed as unclear (in favour of placing the question in limbo). And, for the slightly unclear ones, I've made an assumptions or two in my day. Sometimes correct (given that I get the accepted answer and/or a related comment). Sometimes not correct, but still useful I think (as it may get a few upvotes). And sometimes I just cover all variations in my answer. Mar 23 '14 at 23:26
  • 1
    The motivation here is that the user doesn't get any reputation for the votes on the question from that point onwards -- Any proposal for a community action that robs a user of reputation is dead on arrival.
    – user102937
    Mar 24 '14 at 0:36
  • @RobertHarvey Actually that part is to prevent otherwise good questions from accruing downvotes because of a lack of research effort. Mar 24 '14 at 2:07
  • Downvotes are an important part of the question-blocking system. Any proposal that disrupts that process is also DOA. Can I ask what problem this limbo system is supposed to solve?
    – user102937
    Mar 24 '14 at 2:08
  • @RobertHarvey It doesn't have to disrupt the process. Post score for question-blocking can be calculated on the score pre-'post-limbo-reset' (which would normally be very similar to the stage when a question could perhaps have gotten deleted in the way things currently work) The problem(s)? - "As I see it, I think there's a sufficient combination of ...". I don't think examples of these would be all that useful as I typically can't pick the best ones and this really should've been something anyone who agrees has picked up on already in their daily use of the site. Mar 24 '14 at 2:17
  • See also meta.stackexchange.com/questions/173497/… (description of current system).
    – Jason C
    Mar 24 '14 at 18:10
  • I've read your post twice, and I still no idea what problem limbo is supposed to solve, or what questions it would apply to (you absolutely need to come up with a handful of examples, otherwise why do you think your proposal is useful?), or why we need something that is different from closing and deleting. Mar 26 '14 at 12:07
  • @Gilles Perhaps it's just me trying to find some way to deal with the "Please do my homework / thinking / whatever for me" questions, which, if asked correctly (but still showing no attempt at solving the problem), are perfectly acceptable for the site. I know I should just downvote these (and perhaps comment ... for some reason), but I feel that that often serves little to no purpose. This question is typically the type of thing I'm talking about. Mar 27 '14 at 1:46
  • @Gilles I have seen countless comments saying something like "I want my 'minimal understanding' reason back" (haven't you?), presumably because people don't want questions that show no attempt at solving the problem to get answers. This is my attempt at a solution. Mar 27 '14 at 11:10
  • @Dukeling I have, and participated in that debate. The whole point was that the questions that people had been using it for were largely questions that should be answered, because they're answerable. The problem that needs solving with this question is answering them, and that's a problem we don't need a solution for because answering easy questions happens naturally on SO. Mar 27 '14 at 12:39

Executive Summary

If a question doesn't have enough research effort in it to be reasonably answerable, then vote to close it as "Too Broad." If you want them closed faster, and you are a member with Vote to Close privileges, check your caché of close votes. If it has not yet been exhausted for the day, spend six minutes on the front page or in the Close queue exercising your community-granted privileges, so that bad questions will close faster.

Oh, and downvote them. Downvotes are free on questions.

So why do we ask for research effort in the first place?

Before we revamped the "How do I ask a Good Question" topic in the Help Center on Stack Overflow, we explained why. You can still find this description in several other sites, though:

Sharing your research helps everyone. Tell us what you found and why it didn’t meet your needs. This demonstrates that you’ve taken the time to try to help yourself, it saves us from reiterating obvious answers, and above all, it helps you get a more specific and relevant answer!

All of which is true.

So what happens if the research is missing? Well, one or both of the following conditions may occur:

  1. You have to guess at the OP's level of expertise and his assumptions about the question. Answering such a question frequently leads to the reply "I tried that already, but it didn't work."

  2. You have to write a book chapter outlining every possible permutation of the asker's question. This is often the case with homework questions.

In both cases, the OP hasn't demonstrated his level of knowledge, which is why we used to have a close reason for folks who didn't "demonstrate a minimum level of knowledge" of the subject matter. But demonstrating minimal knowledge is not the same thing as "what have you tried?" In the end, SE decided that the close reason, in the balance, was causing more problems than it solved.

Vote to close these questions as "Too Broad."

That's what they are. They fulfill the second part of the close reason description, which reads: "good answers [to this question] would be too long." If a comprehensive answer that would satisfy every possible permutation of the OP's question would fill the better part of a book chapter, then it's not suitable for the Q&A format.

But they're not getting closed in time, and become broken windows

That's because you're not voting to close them.

I've seen many questions get 20 or so views, and several downvotes, but not a single close vote. Why is this? Is the community getting worn out with seeing vague, underspecified questions by newbies? Then imagine how you would feel if, instead of being a high-rep member trying to answer new questions and getting burned out closing the bad ones, that instead you are a mod, and not only have to deal with the bad questions as a community member, but also the many flags that people cast against bad questions.

It's a bad sign when questions get more vote to close flags from folks who don't have the necessary rep to vote to close, than they get actual close votes, from folks who are supposedly more engaged with the community than the newcomers.

  • But what if it is reasonably answerable (which is really the type of questions my whole request is focussed around)? Mar 24 '14 at 3:06
  • 2
    If it's reasonably answerable, then answer it. That's what we're here for.
    – user102937
    Mar 24 '14 at 3:36
  • Why the broken windows? It is due to lack of education regarding the privileges we're given. We get a notification about a new privilege but I don't recall anything requiring substantial attention on our part. I've become much quicker at voting to close questions after spending more time reading Meta. I did not understand how desirable it is to quickly vote to close. (Regarding your last paragraph, it is especially aggravating to see a multiple*10k user point out in a comment that a question suffers from a close-worthy flaw and yet not see any close votes!!)
    – Louis
    Mar 24 '14 at 11:03

Punishment vs rehabilitation

With any action of this sort there are two elements; punishment and rehabilitation. The combined efforts of the worlds governments have shown pretty conclusively that pure punishment is pretty self defeating. On the other hand rehabilitation can have long term improvements in behaviour.

The proposed limbo is by your own admission punishment based; the question is lost to you, you won't get any reputation and seemingly improvement isn't rewarded. On the other hand closure has a rehabilitation element to it; improve your question and you'll get your answer, and upvotes too!

Either we want these questions or we don't

Assuming a question has 2 states; belongs on site and doesn't belong on site; then given that either a question belongs on the site and all is well, or it doesn't and it should be closed and ultimately removed. Punishing users for something we ultimately decide we want seems confusing.

  • Consider that closing and deleting a question could essentially be the same as placing it in limbo and it then getting removed from limbo as described, except that the possible usefulness of it won't be lost. I disagree that all questions can fit nicely into one of two categories. My question already covers this, so let's add that historical locking comes to mind - it's not on topic, but it's decided that it will stick around because it's useful. Putting a note on a question essentially saying "this isn't really a good question, but it's deemed useful, so it's here" should be similar. Mar 23 '14 at 23:57
  • Consider this somewhat flawed metaphor - someone commits a crime targeting you and then pays you off (assume they'd be no better off with the money for the sake of the argument) (ignoring the obvious moral grey area for the moment). Do you accept the money and forget about it, do you reject the money and give them up to the police, or do you try to find some middle ground where you can keep the money and still give them up (keeping in mind that keeping the money is a win-no lose situation)? Assume giving them up involves the usual rehabilitation / justice / whatever. Mar 24 '14 at 0:19

In addition to Robert Harvey and Richard Tingle's points, I don't think this is a good idea for a couple of reasons.

First, we already have an existing system that allows us to put questions on hold. This system accomplishes pretty much all of the goals of "limbo".

  • Users can vote to place a question on hold.
  • Comments can still be left, which can help the OP work out issues.
  • The close reason is displayed with helpful links to question-writing tips.
  • The OP can still edit the question to improve it.
  • If questions are on hold for a certain amount of time, users can vote to delete it.
  • If a question is sufficiently improved, users can vote to reopen it.
  • These actions will place the question on top of the "active" tab (not the "new questions" tab).

I don't see much reason to add a "limbo". Being a firm believer in "no reason to do it != reason not to do it", I also see a couple of points of damage that this can do. First, it increases the amount of noise on the site; garbage sticks around when the current system already reasonably takes care of it. Second, I think having this state just confuses things; imagine, for example, having to answer a question on meta, "What is the difference between [on hold] and [limbo]?" Third, the points raised in the other two answers I mentioned.

I'm not sure if even the most modest cost is worth any benefits that could come from such a system (which I don't really see).

In my opinion, we'd be better served by using the current system as effectively as possible, rather than trying to come up with an alternative that, fundamentally, would only exist because somebody doesn't know how to use the current system effectively -- then "limbo" ends up being a sort of confusing, redundant feature that mostly sort of overlaps with "on hold" but kind of a little bit maybe not.

Personally, I can't really think of an example of any questions that should have stuck around under "limbo" rules but were removed under current rules. Like Richard Tingle said, either we want them or we don't -- and users already have the opportunity to improve "on hold" questions.

If you think an "on hold" question is answerable and its continued presence is a benefit to the site, vote to reopen it and leave a rationale in the comments. If you just want to help somebody out on a closed question, even if a question isn't that great, you can always discuss the answer with them in comments or chat then let the question fade away to delete-land (then, everybody wins; OP gets some help, and junk isn't left on the site).

You must log in to answer this question.

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