9
$\begingroup$

This question How fundamental is undecidability? was closed within 25 minutes of being asked. The OP was rude to a moderator in the comments, and it isn't a great question, so I do think closing it was a fair call.

However, for me this has highlighted that we seem to be closing questions rather quickly, and certainly before the OP has a chance to change the question in light of comments, for instance to add pointers to literature, or to explain their motivation.

This may be a reasonable rule, but I don't see it in the FAQ. Some poor questions in the earlier days of the site went on to be revised (by the OP and by others) to become interesting and useful, generating significant numbers of page views and votes. There were even instances where for initially uninteresting questions some great answers were provided (e.g. Highest lower bound on NP problems (TSP), True Bit Complexity of matrix multiplication is $O(n^{4})$, and Permanents - Approximation and connection to integer factorization come to mind). I don't have easy access to the vote histories for pages, but I seem to recall that some fairly anodyne questions even ended up being heavily upvoted after a stellar answer was provided, sort of basking in the reflected glow.

Previous discussions mentioned wanting to keep a clean front page during the beta, see On closing, and the awesome power of moderation, How Much Time before Closing a Question?, and Why the extreme strictness on topic closing?.

I would even hesitate asking a question like Rigour leading to insight today. This question started out quite vague but through feedback turned into a reasonable question, and more importantly led to insightful responses by several people.

Is there an unwritten rule that no-one (not even Ryan) shall ever receive the reversal badge?

$\endgroup$
4
  • 5
    $\begingroup$ I think it is fine (and good) to close quickly, if it really is a clear case (like the question that you mentioned). After all, closed questions can be re-opened if needed. $\endgroup$ Apr 7, 2011 at 11:57
  • 3
    $\begingroup$ When the case is as clear as this one, I don't see the problem with closing quickly. The questions you mention above are hardly comparable to the ones that were recently closer. $\endgroup$ Apr 7, 2011 at 15:57
  • 2
    $\begingroup$ I have to agree with the others. This example is not the best test case for the proposition that we are closing questions too quickly. Also, it's april, and all the homework-seeking hounds are coming out of the closet, based on the kinds of questions that are popping up :) $\endgroup$ Apr 7, 2011 at 22:11
  • $\begingroup$ This question seemed to me to contain a kernel that might have been worth polishing. For better or worse, the StackExchange mechanism is that questions are the fields in which answers can be planted, so a bit of leeway might allow serendipity to flower. Note that I believe it is rude to ask someone's badly phrased question in a better form -- yet this is what was suggested by Kaveh at cstheory.stackexchange.com/questions/5872/2on-vs-o2n (closed within five minutes), and it seems this opinion is shared by others. $\endgroup$ Apr 8, 2011 at 7:26

3 Answers 3

6
$\begingroup$

As a relatively new user, I am surprised by how quickly some questions seem to be closed. Because of that, sometimes I've been afraid of asking a question because it might not be "research-ish enough".

In some cases it is absolutely clear that a question is just homework, but I find it annoying that just because the question looks like homework it is almost immediately downvoted and/or closed, without much insight as to why it is being downvoted/closed.

This is particularly true for an area I'm interested in --Automata Theory and Formal Languages--, because many homework questions seem to come from undergrad students taking this course ("Is $L$ context free?"), and also because it doesn't seem to be a mainstream research area (it was 50 years ago, but not now). However, some questions are interesting enough, but are welcomed at once with the following message:

Welcome to cstheory, a Q&A site for research-level questions in theoretical computer science (TCS). Your question does not appear to be a research-level question in TCS. (...)

See, for example, this one. Another example is the question about $a^ib^jc^k$ with $i\neq j\neq k\neq i$, which ended up with a blog post a year after it was asked.

I'm sure that much (if not all) of what I'm saying now has been said before, but I felt the need to express my opinion, because it looks like not much has changed, unless these practices were really worse ten months ago.

$\endgroup$
2
  • $\begingroup$ the stackexchange system seems to require 5 close votes with the voters listed to close the question, but here it seems different where many questions are closed after a single vote by a moderator [usually kaveh]. not sure why. there is an idea in some of the threads that the site will attract a higher caliber of researchers/audience if simple questions are closed quickly, but I dont know if this is necessarily true because Ive seen several very high calibre researchers complain about questions being closed too quickly. on the other hand there is broad consensus on trying to keep out homework. $\endgroup$
    – vzn
    Feb 1, 2012 at 16:39
  • $\begingroup$ @vzn, FYI, I close questions not more often than other mods, and I normally let the community close it so I don't use the mod powers and only cast the 5th vote. I would prefer if I could vote as a normal user but currently the system doesn't have such an option. $\endgroup$
    – Kaveh
    Feb 2, 2012 at 23:57
2
$\begingroup$

"Note that I believe it is rude to ask someone's badly phrased question in a better form". No, it is not, and we have discussed it before on meta and AFAIR we decided that should be the way if you have an interesting question based on a bad question asked by someone else.

Regarding quick closing I think it is good when the question is clearly off-topic. Regarding this particular question, I think closing it as subject/argumentative might have been better.

As a side issue, I start to feel (I haven't checked it, we can search for question with high number of opposite votes) that we are getting a higher level of protest votes (people voting based on votes of others) and I think that is not good. The votes should not be dependent on other people's votes (I saw it in somewhere I don't remember now but I think we should add it to the FAQ). I think that is part of the reason that we don't have highly negative questions (and therefore no new reversal badges) because some users are using their votes as protest to other user's votes. If we vote down questions which are not good in our opinion but are not off-topic, then there will be open question with highly negative total votes, but that is not happening right now. My impression is that a negative vote is interpreted as a really bad thing (which IMHO shouldn't be the case).

$\endgroup$
5
  • $\begingroup$ Kaveh: Regarding the side issue, I agree it would be worth stating in the FAQ what it means to vote up and to vote down. However, with many of our active users below 500 reputation points, I think it is understandable that downvoting is reserved for very special cases. $\endgroup$ Apr 8, 2011 at 8:32
  • 1
    $\begingroup$ @András Salamon, hmmm, I guess you are right. I feel that some users interpret a negative vote as a big deal (kind of an insult to the OP) and protest by voting in the opposite direction since they think the question does not deserve those negative votes. We may want to do something to make it clear that a negative vote is not a big deal and users should vote based on the question and independently of the total amount of votes a question gets. $\endgroup$
    – Kaveh
    Apr 8, 2011 at 8:48
  • $\begingroup$ It is an intriguing possibility, though I'm not sure how to find out if people really do behave this way. It might be worth highlighting that answers (and questions) can be deleted, which not only removes the negative effect of the downvotes, but also removes the negative effect for the downvoters: meta.stackexchange.com/questions/5221/… $\endgroup$ Apr 8, 2011 at 10:38
  • 1
    $\begingroup$ Personally, I upvote if I see a particularly well-phrased and/or interesting question. I downvote if I see a particularly illformed or pointless question. "Normal" and off topic questions do not receive any votes. $\endgroup$
    – Raphael
    Apr 8, 2011 at 20:55
  • 1
    $\begingroup$ @Raphael: An upvote means "this is useful". A downvote means "this in not useful". Based on the explicit text provided by the SE software, it could be argued that you are being too restrictive in the way you interpret a vote. You seem to be requiring something like two sigmas away from the mean. $\endgroup$ Apr 13, 2011 at 8:36
2
$\begingroup$

@Janoma,

  1. The issue of down voting without commenting has been discussed extensively and the current view is that commenting is not required but encouraged.

  2. Closing is not a big deal as is explained in the comment template says. A question can be reopened if it is deemed suitable for the site by 5 high rep members of the community.

  3. It is not easy to see whether a question is research-level or not unless the author of the question explains the motivation and background. It is normal to close a question which looks like non-research-level when there is not any motivation or when it seems that the author does not understand the elementary stuff.

  4. The question you have linked to is by student working on a problem set. IIRC, the question already had a -1 vote when I posted the comment. And again as the full comment says:

    Welcome to cstheory, a Q&A site for research-level questions in theoretical computer science (TCS). Your question does not appear to be a research-level question in TCS. Please see the [FAQ] for more information on what is meant by this and suggestions for sites that might welcome your question. Finally, if your question is closed for being out of scope, and you believe you can edit the question to make it a research-level question, please feel free to do so. Closing is not permanent and questions can be reopened, check the [FAQ] for more information.

    The FAQ also contains an item My (wonderful!) question got closed! What do I do now?.

    I think we have emphasized it enough that closing is not a big deal.

  5. Further more, neither of the questions that you have linked to have got closed in the first place, so I am not sure if they are examples of what you are claiming.

  6. If you think there is nice question hidden inside a question, then you can make suggestions to the OP to improve the question, or ask a new question yourself.

  7. The opinion at the time was that we are not closing the question too quickly, there is no problem with closing clearly off-topic questions as soon as possible. If later the question is improved to an on-topic question it can be reopened. As you could see from the state of this discussion before your new answer, it was accepted that quickly closing these questions is fine so I am not sure why you expected a change regrading in closing them.

ps: next time, please start a new discussion by posting a question in place of posting an answer to an old discussion.

$\endgroup$
7
  • $\begingroup$ I'm not going to reply to everything, but here are four thoughts. First (in response to No. 5), I was not claiming anything in my reply, just expressing my personal opinion. Second (in response to the PS), I thought if I posted a new question, I would have been told to read previous questions (like this one), because the issue had already been discussed so my question was repeated. $\endgroup$
    – Janoma
    Feb 3, 2012 at 2:59
  • $\begingroup$ Third, even though some issues might have been widely discussed already, it is perfectly normal (and expectable) for a new user not no know about them. There might be consensus in those discussions about closing not being a big deal, but a new user might not even know it, and they might feel that closing a question is a rather violent reaction to a newcomer. $\endgroup$
    – Janoma
    Feb 3, 2012 at 3:00
  • $\begingroup$ (continues from number 3) Also, neither the name nor the URL of the website hint that it is supposed to be for research-level questions only. If I have a homework question for, say, my automata theory class, and I'm new to SE, it is likely that I'll go to a search engine and find cstheory.stackexchange.com, a Q&A for TCS, which in principle is exactly what I'm looking for. $\endgroup$
    – Janoma
    Feb 3, 2012 at 3:00
  • $\begingroup$ Finally, by your previous comment and your lengthy answer, I seem to have touched a nerve. I will say no more then, because I do not wish to star a sacred war on opinions, authorities and newbies asking why $\{a^nb^n:n\geq 0\}$ is not regular. $\endgroup$
    – Janoma
    Feb 3, 2012 at 3:01
  • $\begingroup$ @Janoma, you have not touched any nerves. :) I was not criticizing, I was just try to help explain. I suggest that you read the answer a few hours later to see that. ps: it is fine to start a new discussion about previously discussed issues if after reading the previous discussions you think there is a need for revisiting it, just post it as a new question and link to the previous ones. The point of the comment we post and you have quoted the first part is making it clear to the new users that closing is not a big deal. $\endgroup$
    – Kaveh
    Feb 3, 2012 at 5:02
  • $\begingroup$ by the way, when someone visits the site the following message is shown on the top: "Welcome to Q&A for theoretical computer scientists and researchers in related fields - check out the FAQ!" and the FAQ explains the scope of the site more carefully. $\endgroup$
    – Kaveh
    Feb 3, 2012 at 5:05
  • $\begingroup$ Fair enough then :) BTW, I know the message is shown, but my guess is that the average internet user does not bother to read the FAQ, or, for that matter, a welcoming message (they skip to the content). Also, it might be the case that people confuse a Q&A network with an all-topics-are-welcome forum. $\endgroup$
    – Janoma
    Feb 3, 2012 at 11:47

You must log in to answer this question.

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