I'm curious as to why the moderators on this site are so zealous about closing topics that they perceive as off-topic, "not real questions," etc. There are many questions, and I count one or two of my own among them, that the average informed and reasonable person would probably consider fairly relevant to discussions of theoretical computer science...even if they might be 0.5% off-track as far as adhering to the guidelines of the site for optimally phrased questions. These questions will be asked by someone with only the best intentions at heart, and will be immediately "voted to be closed" and commented upon by moderators, typically in a fashion that might be characterized as abrasive, if not rude.

I recognize the importance of keeping the site clean of obvious spam and questions that truly detract from the experience of using the site or the general quality of the content...but do we really need to be so harsh on people who post at least semi-reasonable questions? Rather than immediately write a comment to the effect of, "This is the worst question I've ever seen, I think it should be closed immediately," why not work with posters of sub-optimal questions to try to help them improve the quality? You could give people twelve hours of grace (or something) before the topic is closed. This way, I think you might do a better job at avoiding driving newbies away.

I'm not asking this question in the spirit of being accusatory; I just want to make a gentle suggestion that someone rethink the topic closure policy. I think that given that one of the goals of the site is outreach to non-experts in the field (and I'm certainly one of these non-experts!), I think it would make a lot of sense to think seriously about doing a quick cost-benefit analysis of being so strict on topic closure. What do you gain, and what might you lose by being so strict/abrasive?

In order that this question might not get closed as "not a real question" :), let me ask one: What is the justification for the strictness on question closure? I am not sure that there isn't one, so perhaps I should ask if there isn't some deep-seated StackOverflow cultural mandate that I wasn't aware of that would make my comments above appear foolish. If there's a perfectly good reason for handling topic closure the way it's being handled now, I'd love to know what it is.

Thanks for any thoughts on or answers to this.

  • 1
    $\begingroup$ This was hashed out in meta.cstheory.stackexchange.com/questions/636/… -- we want to avoid this site becoming a homework question answering site. If a question isn't a homework question, and the answer cannot immediately found by googling or Wikipedia, then I think we should let the question stand. But maybe you're right. If a question doesn't make any sense, maybe we should let the OP try to reformulate it. $\endgroup$ Nov 27, 2010 at 20:18
  • $\begingroup$ Please stop thinking that cstheory.stackexchange.com is a website for discussions. It is not. $\endgroup$ Nov 27, 2010 at 20:49
  • 2
    $\begingroup$ I didn't mean to use the word "discussions." I meant, questions-and-answers related to theoretical computer science. I didn't mean that it is a discussion site. $\endgroup$ Nov 27, 2010 at 21:00
  • 2
    $\begingroup$ @Peter Shor, I looked at the discussion you referenced. I agree with what you say: we should let the question stand if it is non-trivial and non-homework. Some of the "elitism" discussion on that thread is a little disturbing. Is this really an "elitist" site? If so, that's pretty poorly thought out. Why make a site only for experts? Don't they all know each other anyway? $\endgroup$ Nov 27, 2010 at 21:03
  • 1
    $\begingroup$ (And by "discussion" I meant "question.") $\endgroup$ Nov 27, 2010 at 21:07
  • 2
    $\begingroup$ @Philip: Please make your question more concrete by giving specific examples of questions that were in your opinion closed too quickly. Currently, it is impossible to give any reasonable answer. I haven't noticed that people are closing anything too quickly. $\endgroup$ Nov 27, 2010 at 21:27
  • $\begingroup$ @Jukka: I agree. Lately, we've been doing a good job of only closing the questions that should be closed. $\endgroup$ Nov 27, 2010 at 21:43
  • $\begingroup$ @Jukka: Here's an example of a question that was not closed, but that some considered to be irrelevant. cstheory.stackexchange.com/questions/1593/… . Note, by the way, that the poster is Lance Fortnow, and thus unlikely to be way out in left-field in terms of asking the question. Maybe I should have said, "why the extreme criticism of questions." My point isn't a technical critique of the act of closing a question; it's more of a, "why is the community occasionally hostile to reasonable questions?" $\endgroup$ Nov 27, 2010 at 23:35
  • 3
    $\begingroup$ Yes, this is an elitist site (in the same sense that MO is), the motivation for starting it was a MO style site for researchers in TCS and as has been stated in the post linked by Peter there are other sites for less elementary questions like Math.SE. $\endgroup$
    – Kaveh
    Nov 28, 2010 at 14:46

2 Answers 2


A bit of "historical" perspective: one of the concerns that we've had (because of close proximity to mathoverflow and comp.theory) is to make sure that the quality of questions is high, and that questions are clear and well-stated. The rough maxim here was: "a site for experts will draw non-experts, but not the converse".

Your question is really two parts: why are we closing questions quickly, and why are we being unpleasant in doing so ?

For the first part, I think there was a natural desire to "clean the front page" in the beginning so as to set the tone for what questions are considered desirable. while I don't think we're sufficiently mature that we can let go of this responsibility, I think it's ok to spend more time working with the OP to fix a question, and maybe we can be more constructive in our suggestions.

On the second part, I think that occasionally some folks do get a little stroppy, and this is uncalled for. We should all remember that things written in comments can be misconstrued easily, and so we have to bend over backwards to be constructive and civil in our responses.

Having said all of this, I will admit to frustration when a number of iterations through the comments fails to improve the question, because the OP, rather than fixing the question, merely answers the comments with more comments. While I personally don't vote to close easily (because my vote is binding), I probably would do so more often if only as a signal to the OP to work on improving their question.

  • 2
    $\begingroup$ I've seen several cases where the OP tries to fix the question, but does not actually make it any less impenetrable, while getting annoyed at our inability to understand his question. This is a good way to get a question closed. $\endgroup$ Nov 28, 2010 at 13:41
  • $\begingroup$ OK, that is very reasonable. I appreciate your perspective and understand what you are saying. $\endgroup$ Nov 28, 2010 at 14:52

I just see only one of your questions has been closed and that was for being exact duplicate:

What techniques are used for proving algorithms optimal?

There are different reasons for closing questions and it would be more useful if you gave examples of questions you think should not have been closed.

The interpretation of comments depends on the person, I don't think anyone wants to be rude here, they simply state the reason that they are voting to close the question and this is actually being nice (one can vote to close without stating any reason.) I have not seen any comment even close to "This is the worst question I've ever seen, I think it should be closed immediately," so I think this is more about how you read them than how they are written.

I think the appropriate time before voting to close a question depends on the reason for closing, e.g. if a question is being closed for being "exact duplicate" I don't see any point in waiting before voting.

I think the discussion would be more constructive if you restrict the discussion to one specific reason for closing.

And as stated in the FAQ, when a question is closed it does not mean that much. It can be edited and reopened. (I guess the trouble is that most new users don't read the FAQ but the point of FAQ is exactly answering those Frequently Asked Questions so the answers for those questions are not repeated.)

I haven't checked the numbers but I don't think we are closing question with higher frequency than MO (even though I would have expected otherwise, one reason is: the site is still in beta and most of our users are new to the site and are not familiar with the policies).

  • $\begingroup$ What we should do is ask the OP to reformulate the question so it makes sense, and if it does, consider reopening it. $\endgroup$ Nov 27, 2010 at 20:22
  • $\begingroup$ @Peter: I do try to mention ways of improving the question so there is no need to close it if I see them, after all we really want to have more question (but without decreasing the level of site). But I think we have already stated this in our FAQ, what is the point of having a FAQ if no one reads it and we have to state the same thing over and over? $\endgroup$
    – Kaveh
    Nov 27, 2010 at 20:27
  • 3
    $\begingroup$ It's not all about me and my closed/deleted questions (though there was one other that I deleted that was going to be closed, I was annoyed at the discussion taking place). You're also right that the one comment I suggested was exaggerated. I think my broader point is that from my own perspective, I can tell you that it can be unpleasant to have people trash your question. I'm largely just suggesting that people be constructive and civil as possible, especially to new users. $\endgroup$ Nov 27, 2010 at 20:59
  • 3
    $\begingroup$ @Philip: You're right: everyone should try to be more civil. $\endgroup$ Nov 27, 2010 at 21:44
  • 2
    $\begingroup$ @Peter, I couldn't agree more. $\endgroup$ Nov 28, 2010 at 1:19
  • $\begingroup$ @Philip: Agreed, we should be more constructive and civil but we expect new users to be respectful and nice as well, e.g. they should understand that close/down votes are not personal, and everyone is just trying make the site better and more useful. $\endgroup$
    – Kaveh
    Nov 28, 2010 at 15:40

You must log in to answer this question.

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