27
$\begingroup$

Recently, I have noticed there is an increasing number of closings (or at least, close votes) on questions like this one. That is, archaic questions with accepted answers, which do not match the "effort" criterion practically imposed on most new questions nowadays.

I stress new, because this criterion is used mainly to discourage homework-grabbing practices, and to make it easier to write helpful answers for the OP (which is not by definition identical to great answers in my book, although they overlap naturally).

But for the type of question I describe, these motives do not apply. I am curious to hear about reasons for these close-votes.

I can think of one good reason to vote-to-close these questions: to indicate that they are not indicative of what constitutes a good question on MSE today. However, this reason is IMO mitigated by the other main purpose of MSE, namely that of a mathematical knowledge repository. Moreover, if it is for this reason that these questions ought to go, then the close reason should indicate this; I don't think that the close-votes really envisage reopening the question after "context" has been added.

N.B. I read this thread, but I contend that the scope is different enough for this not to be a duplicate.

$\endgroup$
10
  • 3
    $\begingroup$ I don't see how the two purposes are in contradiction. Closing an old question does not make the knowledge go away; it's still visible to absolutely everybody. $\endgroup$
    – user856
    Commented Jan 23, 2015 at 22:14
  • 21
    $\begingroup$ @Rahul, unless closing is preparation for deletion. $\endgroup$
    – Gerry Myerson
    Commented Jan 23, 2015 at 22:17
  • 2
    $\begingroup$ Is that a self-answered question? As you state yourself, the only point of keeping old questions and answers is to build a math knowledge repository. If a question is of no value to such a repository, it should be closed. There can be multiple reasons for that to happen, and the people who closed that question thought one of them applied. You might disagree with that assessment, but that's a different matter. The distinction between old and new questions, or questions with or without accepted answers, is irrelevant. $\endgroup$
    – Najib Idrissi
    Commented Jan 24, 2015 at 9:03
  • 4
    $\begingroup$ @NajibIdrissi I see your point, but in that case the close-votes should state "of no future value", instead of "no context". That would be a different story altogether. But now, there is no amount of context that would improve e.g. the question I linked, therefore the close reason is wrong (and the purpose of "no context" is very time-dependent). $\endgroup$
    – Lord_Farin
    Commented Jan 24, 2015 at 9:07
  • 3
    $\begingroup$ So your only concern here is the wrong close reason? As an aside, there are plenty of ways to add context. And the purpose is not necessarily time dependent: context can improve searchability (by adding new keywords), help users understand what is difficult about the problem (the answer can just sweep difficulties under the rug by using clever techniques, and then you wonder how the answerer thought of that...), why such a question is interesting at all, or simply understand the problem at all by including relevant definitions, etc. $\endgroup$
    – Najib Idrissi
    Commented Jan 24, 2015 at 9:15
  • 4
    $\begingroup$ @Najib No, my concern is in the why; I can't see why we would like to close this type of question with this reason. I agree that my perception of "context" was a bit narrower than what you see (thanks!). But if this is all, the effort put in closing these should IMO be accompanied by creating stellar abstract duplicates that go above and beyond any specific example. The closing should then be as a duplicate. $\endgroup$
    – Lord_Farin
    Commented Jan 24, 2015 at 10:48
  • 2
    $\begingroup$ @Lord_Farin closing as duplicate also creates push-back; sometimes to un-reasoanble extents. Though also reasonable. In particular, many an asker here, especially of these questions, does have perfectly fine "duplicates" at their disposal anyway, typically some abstract some concrete (I mean, their lecture notes, their textbooks, similar examples seen in class, etc.). Those do not help them, whence it might be somewhat pointless to point them at ours. (ed ajf) $\endgroup$
    – quid
    Commented Jan 24, 2015 at 11:49
  • 1
    $\begingroup$ @quid That's true for new questions, I agree to an extent. But it does not apply to old questions, which my point was about. $\endgroup$
    – Lord_Farin
    Commented Jan 24, 2015 at 11:56
  • 8
    $\begingroup$ Why doesn't it apply to old question? Just that nobody did empty some forgotten trash-bin for a year does not mean the garbage needs to stay with us forever or transformed into some artefact. $\endgroup$
    – quid
    Commented Jan 24, 2015 at 11:59
  • $\begingroup$ @quid A question needn't remain open for OP's understanding's sake if it's old. So it should be closed as a duplicate. More so since otherwise the garbage will simply be replaced by new instances of the same garbage; it's just not a "solution" for the long run. $\endgroup$
    – Lord_Farin
    Commented Jan 24, 2015 at 12:03

2 Answers 2

Reset to default
14
$\begingroup$

The only reason old questions are kept is to serve as repository of mathematical knowledge. Very poor questions, regardless of the quality of the answer, do not deserve to be thus enshrined, and reflect poorly on the site.

I wish the excellent teachers on this site had spent their time on worthy questions, rather than rewarding very poor ones. But, if your old answer is at risk of deletion due to being associated with a question of suspect quality, there is an easy fix: edit the question until it is as good as the answers accompanying it. Finally, if appropriate, add your revised question and answer to the list of abstract duplicates, so that your answer will maximally benefit future students with similar questions.

$\endgroup$
27
  • 10
    $\begingroup$ This makes little sense for the "no context" reason. The extent to which a question is useful to people visiting the site in future has little to do with what thoughts the OP may have had. In fact, it is better for that purpose to keep the question free of that kind of clutter. $\endgroup$
    – user208259
    Commented Jan 25, 2015 at 2:15
  • 2
    $\begingroup$ @user208259 please see various forms of context Even a humble user like me feels able to do improvements along these lines in various cases. It should be quite trivial for the "best teachers." But even if only excessively poor typesetting and alike would be fixed it were already an improvement. (ed ajf) $\endgroup$
    – quid
    Commented Jan 25, 2015 at 2:20
  • 2
    $\begingroup$ @quid If you would prefer to see the questions edited, then do that instead of deleting them. $\endgroup$
    – user208259
    Commented Jan 25, 2015 at 2:27
  • 2
    $\begingroup$ @user208259 I did not express any personal preference but commented on feasibility and potential usefulness that you denied based on, or so it seems to me, incomplete understanding of what "context" means. Further, I cannot even vote to delete, so your comment seems somewhat out of place for this reason too. $\endgroup$
    – quid
    Commented Jan 25, 2015 at 2:35
  • 1
    $\begingroup$ @quid It's not about you personally. The point is, that if for a particular question, editing is feasible, that's a good reason not to delete it. But if the objection is that the OP didn't include his inexpert attempts at a solution, that certainly isn't something that should be edited in by anyone if the purpose is to have a useful question and answer for future visitors to the site. On the contrary, it's better not to have that in most cases. $\endgroup$
    – user208259
    Commented Jan 25, 2015 at 2:38
  • 2
    $\begingroup$ I have had to downvote. Unlike other SE sites, on this site we tend to reserve editing of questions for typos, formatting, and similar things. We don't generally rewrite the OP's question to meet out own standards - that is up to the OP. $\endgroup$
    – Carl Mummert
    Commented Jan 25, 2015 at 3:11
  • 3
    $\begingroup$ @Carl This happens often on old questions. Example: closed, deleted, edited by someone else than the OP, undeleted, reopened. $\endgroup$
    – Najib Idrissi
    Commented Jan 25, 2015 at 8:39
  • 2
    $\begingroup$ @user208259 Everything but the OP's background can be supplied in a later edit (the origin of the question can be tricky to find but not impossible). Once again, see this for an example. "Include your work" can be "this question is difficult because usual techniques to solve similar problems don't apply", for example, something that someone else can add (and there are other ways still). "Motivation" isn't hard to find for questions at this level. References and definitions are obvious ones. $\endgroup$
    – Najib Idrissi
    Commented Jan 25, 2015 at 12:32
  • 1
    $\begingroup$ People are free to disregard whatever they want. As Willie says these are suggestions. If someone who is able to ask clear mathematical questions can find another way to add context that's good too. The list is meant to be a guide for new users who don't know how to ask suitable questions. It's also a bit weird to see you assert what a policy was meant for when it was written more than a year before you created your account (10 days ago). $\endgroup$
    – Najib Idrissi
    Commented Jan 25, 2015 at 12:42
  • 5
    $\begingroup$ Do you want to talk about "useful"? That question is unfindable. Users who seek the answer to that question and somehow manage to find it will not understand why it should be difficult to answer. A user who stumble upon the question will not understand why it should be interesting, where it comes from. It's unlinked to any similar problem that one could encounter; there's no general strategy, just a ad-hoc solution. If a student came to you and asked that question, would you simply hand them a paper with the answer on it and expect every new student to refer to that sheet of paper? $\endgroup$
    – Najib Idrissi
    Commented Jan 25, 2015 at 12:52
  • 3
    $\begingroup$ @user208259 the analogy with problem books is flawed as those books, at least those I know, typically provide some context. They being a collection, possibly not for each problem individually, but there typically is some context; perhaps in some case it is very reduced, yet even the title of the book normally is such that it provides some context alone. $\endgroup$
    – quid
    Commented Jan 25, 2015 at 14:11
  • 4
    $\begingroup$ @user208259 Yes, my position is that it's the responsibility of the people who don't want a question to be deleted to improve the question so that it meets the website's standards. $\endgroup$
    – Najib Idrissi
    Commented Jan 25, 2015 at 14:43
  • 1
    $\begingroup$ In response to the comments above about what value certain isolated questions carry, I contend that abstract duplicates are the way to go. I will make a separate thread about this as one of the digestions of this discussion. The current options advocated are all considered lacking by a significant portion of the user base, so we ought to contemplate alternatives - and these are most likely going to require more effort. Quality comes at a price, so we shouldn't really expect the current, quite lazy practice to stand the test of time. Something must be done. $\endgroup$
    – Lord_Farin
    Commented Jan 25, 2015 at 15:01
  • 2
    $\begingroup$ @NajibIdrissi You are obviously not able to perceive that enforcing these "standards" is counterproductive when they result in bad content being saved (unreadable because of an OP's rambling) and good content being deleted (clearly stated problem and solution). The standards are doubly illogical when they are applied retroactively, placing an unrealistic burden on answerers to save good content. $\endgroup$
    – user208259
    Commented Jan 25, 2015 at 15:04
  • 3
    $\begingroup$ @user208259 I explained this above already; typically lack of context is not the only reason for deletion. The presence of context is one thing to consider. A general assessement of the overall quality is made. If it is not good enough it is deleted; added context can improve the overall quality and thus prevent deletion. $\endgroup$
    – quid
    Commented Jan 25, 2015 at 16:19
5
$\begingroup$

I understand your point view Lord_Farin (but maybe not in the same manner), and I see this as double jeopardy. Older questions had to pass different standards to make it past the court of public opinion when they were first posed and the ones that are still around passed. Now, we are prescribing new standards to question that were already tried and found not guilty, i.e. no closure and delete. Some or many of OP may not be around any more so they aren't here to put down what they attempted. Also, for the really old post, I doubt the OP will even remember what their working was when they original asked it.

I noticed that quid pointed out that many of the educators are still answering the same questions (I am not calling you at here quid); there has always been a solution to that problem though, close as duplicate of an older post with acceptable answers (not necessarily accepted answers since some people don't accept them).

$\endgroup$
10
  • 3
    $\begingroup$ The question is though how many versions of a proof that $5 \mid n^4 - 1$ do we want? Shall we dupe-merge until we have thirty, fifty, hundred answers? I think at some point the marginal benefit is not only negligible but then negative. This is as documented also not a question of visibility only. (ed ajf) $\endgroup$
    – quid
    Commented Jan 25, 2015 at 3:18
  • $\begingroup$ @quid the question should be closed before it recieves additionally answers or merge non dupe answers to the old post. $\endgroup$
    – dustin
    Commented Jan 25, 2015 at 3:22
  • 1
    $\begingroup$ Yes I know it should; but in practice it is quite tricky to get it close so quickly. $\endgroup$
    – quid
    Commented Jan 25, 2015 at 3:39
  • $\begingroup$ @quid that is why the non dupe answers should be merged to original post and then close the newer post as a dupe. $\endgroup$
    – dustin
    Commented Jan 25, 2015 at 3:40
  • 3
    $\begingroup$ The SE dupe mechanism is completely brain-dead. The worst part is not possible duplication but, rather, inhibition of new answers by new experts that have recently joined the site. Rarely do old questions receive newer, better answers - they are stagnant. Big design flaw. $\endgroup$
    – Bill Dubuque
    Commented Jan 25, 2015 at 3:42
  • $\begingroup$ @BillDubuque maybe a meta request needs to be proposed then so SE can considering doing something if the post receives enough hype. $\endgroup$
    – dustin
    Commented Jan 25, 2015 at 3:44
  • 9
    $\begingroup$ @Bill Please explain how new experts are prohibited from answering old questions. $\endgroup$
    – Najib Idrissi
    Commented Jan 25, 2015 at 8:12
  • 1
    $\begingroup$ @Najib This has been explained at length elsewhere. Anyone who has been here long enough knows that only rarely do older questions receive new better answers. The quality of answers rarely improves over time because SE platform does a very poor job of re-exposing older questions and re-motivating potential new answerers. Many questions from the early days of the site are frozen in time with low-quality answers. They will probably never receive better answers. The recent rapid close and delete campaign greatly exacerbates these stagnancy problems. $\endgroup$
    – Bill Dubuque
    Commented Jan 25, 2015 at 15:23
  • 4
    $\begingroup$ The key to the stagnancy issue is to ignore the claims by SE.com that this site is primarily an archive. A few questions may be useful as an archive, but it seems clear from experience that the main benefit of this site (math.SE) is to provide answers to the OP of each question. Not only are old questions not re-exposed, the search feature is nearly useless for mathematics. @Bill Dubuque $\endgroup$
    – Carl Mummert
    Commented Mar 2, 2015 at 12:31
  • 1
    $\begingroup$ @CarlMummert I would agree the search isn't great but I have used with a good amount of success so it can be done. $\endgroup$
    – dustin
    Commented Mar 2, 2015 at 16:06

You must log in to answer this question.

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