The help center page about site scope includes the phrase

Some kinds of questions should not be asked here:

A little while ago it was suggested (and further discussed) that the use of "should" implies that the following items are to be interpreted as soft guidelines, not rules, i.e. that exceptions are not uncommon. But I believe that was not the intent of the policy. My understanding is that the types of questions listed on that page are never on topic on this site. Is that accurate?

If so, should we change the wording of that part of the help center, perhaps to something like

Some kinds of questions are inappropriate for this site:


Some kinds of questions must not be asked here:

to reduce the chance of confusion?

  • $\begingroup$ Speaking of wording on our help page, some of the example on-topic questions are pretty borderline and could be closed as too-broad. I'm thinking of the Higgs question and the Telescope question. $\endgroup$ – Brandon Enright Dec 8 '13 at 4:31
  • $\begingroup$ Yes. Oh, I need 15 characters, OK: yes, yes, yes, yes! :-) $\endgroup$ – John Rennie Dec 9 '13 at 11:32
  • 1
    $\begingroup$ @David Z: You wrote: "A little while ago it was suggested (and further discussed) that the use of "should" implies that the following items are to be interpreted as soft guidelines, not rules, i.e. that exceptions are not uncommon", and you gave links to discussions with me. I just would like to emphasize that I did not offer the wording in the quote, I specifically defined my interpretation of "should not": "As a rule, don't do that; but some exceptions may be warranted." That does not mean "exceptions are not uncommon". $\endgroup$ – akhmeteli Dec 13 '13 at 15:02

I completely agree that this is accurate. Any question that fits any of those four descriptions should rightfully be closed. There are grey areas around them - there always will be - but you have to be really far from those descriptions to be on-topic here.

I like your first suggestion, though I would actually strengthen it to the following.

Some kinds of questions are inappropriate for this site, and will likely be closed or even deleted.

  • 4
    $\begingroup$ I suggest could be replaced by will likely, but otherwise yes. $\endgroup$ – Kyle Kanos Dec 8 '13 at 3:43
  • 1
    $\begingroup$ I agree with Kyle and also would suggest actually saying "deleted" instead of "removed" - no reason not to be consistent. $\endgroup$ – David Z Dec 8 '13 at 4:35
  • 1
    $\begingroup$ Why not just "will be closed or deleted" instead of "could be" or "will likely"? Otherwise there's still a hint that you might get away with it... $\endgroup$ – Nathaniel Dec 10 '13 at 9:51

To give an opposing view, I believe there is some value in being ambiguous there. Site rules should not be fast and hard rules, but rather guidelines. I wouldn't want to close a useful and interesting question just because it might formally violate the rules. Rather, it should be decided on a case-by-case basis whether this question brings value to the site or not.

Speaking of which, I don't understand why the question mentioned above was closed. Well, OK, I understand the mechanics of what happened. It's a homework question, it was decided that homework questions are not wanted, and thus the question was closed.

What I don't understand is why the mechanics were blindly applied. Do people think the concrete question is bad, or it's harmful to answer it? Imagine the asker was trying to write a paper, and not doing homework. Would the question still be closed? Probably yes. That would mean that computational / derivational questions are no longer allowed. I'm just worried that we are narrowing the scope of the site down and down until we can only ask the most high-level vauge conceptional questions.

  • 3
    $\begingroup$ The (main) reason no-effort homework questions are not allowed, and the reason that applied to this specific case, is that we want it to be very clear this is not a site where people can come to take the easy way out of doing educational problems. So this was not a case of blindly applying a policy. I do think it's harmful to answer that question, just as much as it's harmful to answer any other no- or low-effort homework question. $\endgroup$ – David Z Dec 8 '13 at 10:52
  • $\begingroup$ Derivational questions not focussed on physical insight are indeed not allowed. Besides usually being homework, they additionally only help the OP. In the case of conceptual posts, more visitors are helped by them. Nobody will search for "height of water column required in tanker to <whatever>" and get exactly the same question they had in mind. People will, however, search "Why does pressure decrease with speed?" or similar, and get the answer tgey wanted. $\endgroup$ – Manishearth Dec 9 '13 at 1:34
  • $\begingroup$ The problem with applying case by case is that OPs argue. If you have a concrete policy to link them too, they won't. Case by case discussion is still possible on meta, but of course it would be annoying if every closed question asker did that. But if you feel a post should not be closed even if policy demands it, post on meta. $\endgroup$ – Manishearth Dec 9 '13 at 1:37
  • $\begingroup$ I must also point out that, policy aside, that question did have other problems. As I mentioned in a comment, it is ill-posed as some terms are not defined in the question. $\endgroup$ – Emilio Pisanty Dec 9 '13 at 18:16
  • 2
    $\begingroup$ I agree with jdm's position. What would we win if we, say, replace "should" with "must"? Hopefully, there will be fewer no-effort homework questions? That does not seem likely to me: there are a lot of such questions now, but it does not look like they are posted by people who read the policy at all, let alone parse it, so I don't think they care if it's "should" or "must" there. These people often have low reputation and, I suspect, don't care if they are banned. What would we lose? Reviewers and moderators will lose discretion. Is this advisable? What if we just like the question? $\endgroup$ – akhmeteli Dec 11 '13 at 12:53
  • $\begingroup$ @Manishearth made a good point about OPs arguing. However, if we have "should" in the policy, that seems enough to reject such OP's arguments: you refer to the policy and exercise your discretion. "Case by case" does not mean we need to think hard before we close any dubious question: the current policy allows us to routinely close such questions. We can allow dubious questions as an exception and close them as a rule. $\endgroup$ – akhmeteli Dec 11 '13 at 13:14
  • $\begingroup$ @akhmeteli it's possible (even common) that a lot of people like a question and yet it's inappropriate for a Stack Exchange site. Because of cases like that, I do think the policy should be spelled out so that reviewers and moderators aren't tempted to use their discretion. Remember the goal is to eliminate questions which are bad for the future development of the site, not to eliminate questions which people don't like. $\endgroup$ – David Z Dec 13 '13 at 4:36
  • 3
    $\begingroup$ @David Z: It is not quite obvious why questions people like, even if they are objectionable in some other respect, are necessarily bad for the future development of the site. If, hypothetically, reviewers and moderators don't want a question to be closed, maybe it just shouldn't be closed? If a reviewer or a moderator errs in favor of a question, others will probably correct his/her mistake, but if they don't, maybe this is a worthy question, after all? Let me emphasize that we are talking about rare exceptions. $\endgroup$ – akhmeteli Dec 13 '13 at 5:25
  • $\begingroup$ @Emilio Pisanty: I agree, the question is ill-posed. But what conclusion should we draw from that? IMHO, questions do not necessarily have to be clever (otherwise they are trick questions:-) ). If somebody asks a question, apparently (s)he has some problems with it, and ill-posedness of the question can be a powerful source of problems; then our answers can help in this respect without being detrimental to the educational purpose of homework. After all, we don't know if it's student's or professor's fault if the question is ill-posed :-) $\endgroup$ – akhmeteli Dec 13 '13 at 14:04
  • 1
    $\begingroup$ @akhmeteli I think the question was close enough to borderline that an appropriate edit could have brought it back on-topic. I was slightly disappointed that the comments discussion went the way it did instead of turning the question into a well-posed, concrete, answerable one. The OP seems to be gone, though, and it's on them to fix (at least some of) the problems. $\endgroup$ – Emilio Pisanty Dec 13 '13 at 14:16
  • $\begingroup$ @Emilio Pisanty: Thank you. And why are you slightly disappointed? It looks like you strongly favor the tightening of the policy? As for the question... Well, maybe one question is not so important in the big picture:-) $\endgroup$ – akhmeteli Dec 13 '13 at 14:46
  • 1
    $\begingroup$ @akhmeteli I was disappointed that the discussion went into why the question was closed instead of improving it so it could be reopened. But that can only happen if the OP is still there. I'm also less strongly in favour of an absolutely-no-homework policy than might appear at first sight; even my proposed policy was more an attempt to see what a working formally-no-homework policy would look like and whether we like it or not. (At least, in retrospect.) $\endgroup$ – Emilio Pisanty Dec 13 '13 at 14:50
  • 1
    $\begingroup$ @Emilio Pisanty: Thank you for the explanations. You might wish to edit this retrospect into your answer, otherwise the latter may influence the policy in the way you won't quite like:-) $\endgroup$ – akhmeteli Dec 13 '13 at 15:09
  • $\begingroup$ @akhmeteli No, I stand by my answer to this question. The four types of question described in here are not questions I want to be seeing, and we get enough of them already. If the wording can be changed to reflect that then I'm behind it. $\endgroup$ – Emilio Pisanty Dec 13 '13 at 17:08
  • $\begingroup$ @Emilio Pisanty: Thank you for the clarification. $\endgroup$ – akhmeteli Dec 13 '13 at 21:14

You must log in to answer this question.

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