3
$\begingroup$

Update: I gave up trying to convince other people to use “too localized” instead of “off topic” when closing too elementary questions. It seems that most people disagree, and I started to feel that it is a waste of time to try to convince people otherwise. For now, I still hesitate to vote for off topic (or to be listed as one of the people who closed the question as off topic) when I see too elementary a question, so probably I will not vote to close it. Hopefully there are enough people to close it anyway.


When we vote to close a question, we choose the reason from several choices. It seems that if a question is too elementary, it is often closed for the reason “off topic.”

I understand that this choice of reason is based on the consensus about the expected level of questions on cstheory.stackexchange.com. However, if a question is closed for the reason “off topic,” the following explanation is added to the question:

closed as off topic by [Someone], [Someone], [Someone], [Someone], [Someone] [Date]

Questions on Theoretical Computer Science are expected to generally relate to Theoretical Computer Science, within the scope defined in the faq.

I have trouble even parsing this sentence because of the tentative name of the website, but more importantly this does not mention the expected level of questions on cstheory.stackexchange.com. It plainly states that the question is not related to the theoretical computer science, which is often false. This gives the false impression that we take the theoretical computer science very narrowly. I think that this is problematic.

(Added in revision 5: The link in the explanation quoted above links to the built-in FAQ, not to the FAQ on Meta. The built-in FAQ is not clear about the expected level of questions, either, except that the sentence “Theoretical Computer Science - Stack Exchange is for theoretical computer scientists and researchers in related fields” may suggest that research-level questions are expected. A clearer statement can be reached by the following chain of links: the explanation of “off topic” → built-in FAQFAQ on MetaWhat kind of questions are too basic? But I doubt that users can find this path unless they are already familiar with the website.)

What should we do?

Here are a few suggestions:

  1. If the consensus is really that cstheory.stackexchange.com is for questions of at least certain level, the explanation of the reason “off topic” should be rewritten to reflect it. Is this possible?
  2. On MathOverflow, too elementary questions are often closed for the reason “too localized.” Obviously this is an abuse and it certainly confuses some users, but it is possible to interpret it as “This question is so elementary that it is useful only for the asker,” and it is understandable (or at least less prone to misunderstanding). How about adopting this convention, at least until the explanation of “off topic” is modified to a better description?

Related:

  • See a related FAQ entry and the links given there for discussions about what we mean by “too elementary.”

Added in revision 4: Whereas the description of the close reason is not shown on the question page on MathOverflow, it is shown on cstheory.stackexchange.com, which may make abusing harder. Here is the description of the close reason “too localized”:

This question would only be relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet.

The point of the suggestion 2 above is that it is at least possible to argue that too elementary questions fall into the third group (and possibly also the second group) of this description.

| |
$\endgroup$
  • 3
    $\begingroup$ To me, using "too localised" seems a bit strange. Usually elementary questions are exactly the opposite of questions relevant to "a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet". They are questions that are of interest to virtually all TCS students, they are covered worldwide at all universities on basic CS courses, etc. $\endgroup$ – Jukka Suomela Oct 9 '10 at 18:01
  • $\begingroup$ @Jukka: I agree that using “too localized” is not ideal either, but I still think that it is far better than claiming that a question is not about theoretical computer science. If there is a better alternative, I want to know it. $\endgroup$ – Tsuyoshi Ito Oct 9 '10 at 18:35
  • 1
    $\begingroup$ Well, the "off-topic" reason at least refers to the FAQ, which then defines that too elementary questions are indeed off-topic. But I agree that the description could be more clear. $\endgroup$ – Jukka Suomela Oct 9 '10 at 19:36
  • $\begingroup$ @Jukka: If I am not mistaken, the shortest path from the explanation of “off topic” to the description stating “too elementary questions are off-topic” is: the explanation of “off topic” → built-in FAQFAQ in metaWhat kind of questions are too basic?. In my opinion, this is too long to actually follow, especially for new users. $\endgroup$ – Tsuyoshi Ito Oct 9 '10 at 19:44
  • $\begingroup$ Revision 5: I updated the post to explain that why the link in the explanation of “off topic” to the FAQ does not help much. $\endgroup$ – Tsuyoshi Ito Oct 10 '10 at 23:16
  • 1
    $\begingroup$ My suggestion is that when we are voting to close the question, we should add a comment to the question that (a) it's too basic and (b) there's no "too basic" button in StackExchange, so we're using Reason X instead. I agree; I don't like any of the choices for Reason X. Surely one of the closers should be able to do this. $\endgroup$ – Peter Shor Nov 10 '10 at 16:38
  • 1
    $\begingroup$ Tsuyoshi, why not just choose whichever reason you prefer and vote on that ? Even if there's no consensus, it's the NUMBER of votes that decides closing. $\endgroup$ – Suresh Venkat Nov 10 '10 at 18:02
  • $\begingroup$ @Suresh: The close reason shown when the question is actually closed is chosen by majority (see “What if multiple close reasons are used on a question?” in meta.stackexchange.com/questions/10582/…). Therefore, even if I choose “too localized”, if three or four choose “off topic”, it is shown as if I voted for off topic. At this moment, I cannot tolerate this. $\endgroup$ – Tsuyoshi Ito Nov 10 '10 at 18:14
  • $\begingroup$ I see. but your update implies that this outcome is SO HORRIBLE you can't even countenance voting to close a question. Is that the case ? $\endgroup$ – Suresh Venkat Nov 10 '10 at 18:16
  • $\begingroup$ @Suresh: For now, yes. (I thought I stated “for now” in the update, but it seems that I had erased it by mistake while I was editing the update. Now I inserted the words “for now”.) I guess that it will take some time for me to get used to seeing the close reason “off topic” for too easy TCS questions. $\endgroup$ – Tsuyoshi Ito Nov 10 '10 at 18:22
  • 3
    $\begingroup$ ok. It's your choice for sure, but a 'grit my teeth, post a clarifying comment and vote to close' might be more useful to all of us than 'I can't stand to vote on this'. Of course you've been very helpful and community-minded already in so many ways ! $\endgroup$ – Suresh Venkat Nov 10 '10 at 18:27
  • 1
    $\begingroup$ I epsilon-prefer "off topic," but I think neither one is sufficiently explanatory. I think the real fix here is to agree on some comment-wording like what Kaveh proposed elsewhere in meta, and also put the homework/motivated-question thing back into the FAQ somehow, or into an easily-linkable place like a separate Meta thread. However, am I rushing to do this myself? Nope. Maybe this weekend, unless someone beats me to it. $\endgroup$ – Aaron Sterling Nov 10 '10 at 22:16
  • 1
    $\begingroup$ Why not add a question to the FAQ asking "Why was my question closed as 'off topic'?" Then you could simply add a comment pointing out that elementary questions are considered off topic for this site, and include a link to the relevant section of the FAQ. $\endgroup$ – Joe Fitzsimons Nov 11 '10 at 5:49
6
$\begingroup$

I have been voting "too localized" in place of "off-topic" as Tsuyoshi Ito suggested for sometime. But now that I reread the statement, I disagree. The text says:

Questions on Theoretical Computer Science are expected to generally relate to Theoretical Computer Science, within the scope defined in the faq.

It is not saying that it is not related to TCS, it says it should be in TCS within the scope defined in the faq. I prefer to vote the homework/elementary question as off-topic, and add a comment that "the question seems to be a homework/elementary". It is better suited for the situation and gives a feedback that they are off-topic for the site. I also don't think any confusion can arise when there is a comment like that.

| |
$\endgroup$
  • 1
    $\begingroup$ Thank you for your reply. I already explained (1) why the link to the FAQ does not avoid the confusion and (2) why a comment is difficult to override what is said in a bigger font, but I accept the fact that your view is different from mine. $\endgroup$ – Tsuyoshi Ito Oct 15 '10 at 20:16
  • $\begingroup$ @Tsuyoshi: Maybe you are right. It am not really sure which one is better. I don't that the OP will be confused, but other people might get confused. I wish we could add "research level" to the text. $\endgroup$ – Kaveh Oct 15 '10 at 20:45
  • $\begingroup$ @tsu correct, also, if you want to wear "kid gloves" and be nice to new but clueless users you should be adding comments to their closed posts anyway. $\endgroup$ – Jeff Atwood Nov 7 '10 at 21:49
3
$\begingroup$

Given that we don't have direct control over the messages that get listed when choosing to close, I think it would be more useful to explain in comments what the problem is

| |
$\endgroup$
  • $\begingroup$ Do you think that it works? The closing description is much more prominent than comments, and I do not see how denying the description in comments is effective. $\endgroup$ – Tsuyoshi Ito Sep 22 '10 at 10:41
  • 1
    $\begingroup$ Suresh: that's a reason for not changing the message (although we could lobby the SX team), but not a reason against Tsuyoshi's other suggestion, changing our close reason to "Too localized". Consider how this would work if comments start "Too localised or elementary: this is a blah-de-blah question". $\endgroup$ – Charles Stewart Sep 22 '10 at 11:12
  • 1
    $\begingroup$ fair enough. so should we put something in the FAQ regarding what option to use for closing ? $\endgroup$ – Suresh Venkat Sep 22 '10 at 17:05
0
$\begingroup$

When voting for closing, please give a pointer to a page where the asking person might get help. I think it is bad style to close without knowing/pointing to an alternative.

| |
$\endgroup$
  • $\begingroup$ I have made some effort to keep this discussion focused on which close reason to choose. Please open a new thread to discuss your own proposal, which is not about the close reason for too elementary questions. $\endgroup$ – Tsuyoshi Ito Oct 16 '10 at 13:26
  • $\begingroup$ Since Suresh's answer also tackled the process/style of closing, not the reason for closing, I figured my answer would be in scope, sorry. $\endgroup$ – Raphael Oct 25 '10 at 7:16

You must log in to answer this question.

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