I am adding this clarification just to be sure that my posting is not being misinterpreted. The reason that I am adding this clarification is that the negative reaction to this meta posting seems to be inconsistent with how I have perceived MathSE reviewers reacting to the corresponding MathSE questions.

The focus of this meta posting is narrow, and only applies to those MathSE postings that (for example) have otherwise met the requirements in this article. For example, an otherwise high quality posting that is a solution verification posting that brings a new approach to a previously discussed problem on MathSE.

The original meta post article, unedited, is below.

This question is a follow-up posting to this meta-MathSE article.

It seems as if there are two sides to the issue of how to deal with solution diagnosis/verification postings that are not of general interest to the MathSE community. Here, I am specifically focusing on those postings where the original poster shows work and where the question would otherwise be considered high quality. Within this subset of postings, while I am not specifically focusing on postings that provide an alternate approach to questions whose answers have already been posted (i.e. the underlying question is a duplicate), that is certainly one of the frequent characteristics of such questions.

From what I can gather, in the previous post, there seems to be merit on both sides of the issue:

  • On the one hand, if the OP shows work, I feel that the offered solution, that is typically asking either "Is this alternative approach valid?" or "Where did I go wrong?" deserves an answer.

  • On the other hand, with rare exceptions, I feel that the posting is typically only of interest to the original poster. This implies that the posting does not belong in a repository of general interest questions.

So, my question in this meta posting is:
Can an elegant compromise be reached that satisfies both of the concerns shown above?

In the previous posting I suggested creating a $<$not-of-general-interest$>$ tag. My idea was that any MathSE reviewer with a sufficient rep, who reviewed a posting that fit the above scenario, would simply add the $<$not-of-general-interest$>$ tag to the posting. I envisioned a process that would automatically wait $7$ days from the date of the first posting, and then schedule the posting for closure or deletion.

The impression that I got was that the programming involved in creating the process would be significant, perhaps very significant. This is why, in this posting, I am using the phrase elegant compromise.

So, now I have another idea, that may or may not allow an elegant implementation. What if a quarterly site maintenance process was run, that automatically looked for any posting with the $<$not-of-general-interest$>$ tag that was more than $7$ days old? This process would then automatically delete the posting.

I am totally ignorant of the site maintenance procedures at MathSE. I have no idea how difficult it would be to implement such a process. This is the major question that I am asking, in this meta posting.

How difficult would it be to implement such a procedure?

While all responses are welcome, I am hoping to receive responses from both experienced MathSE reviewers who are very knowledgeable in site maintenance procedures, as well as MathSE moderators.

The remainder of this posting discusses peripheral issues that may become irrelevant, if no reasonable way can be found to implement the suggested procedure.

For what it's worth, I can't really see anyone objecting to high rep reviewers or moderators deciding, that a posting's long-term disposition should be deletion. The original poster shouldn't care because they received their answer. Anyone who disagreed with respect to a specific posting's pending disposition could discuss the issue in the Cured chatroom.

If the proposed compromise idea in this meta post is judged to be worth pursuing, then the issue of (for example) whether the site maintenance process should immediately delete the posting or merely close it, can be discussed. The issue of what authorization/voting is needed to assign the $<$not-of-general-interest$>$ tag can also be discussed.

I suspect that the general MathSE view would be that such issues are less critical than the corresponding issues that relate to (for example) problem-statement-questions that have just been posted. This is because I think that after the original poster gets their question answered, they won't be that concerned about the long-term disposition of their question.

If feasible, one idea is to attach a mouse-hover property to the tag, so that when the original poster mouse hovers the tag, they receive an explanation of the significance of the tag. Then, if they wish, they have at least $7$ days to save the question and any answers received, privately on their own pc.

  • 3
    $\begingroup$ If a post should be deleted, it should be deleted immediately, not just after the poster has got their answer. SE is not a homework assistance or personal tutorial website; this idea is just asking for it to become one. No. $\endgroup$
    – Nij
    Jul 17 at 8:12
  • 1
    $\begingroup$ @Nij "Here, I am specifically focusing on those postings where the original poster shows work and where the question would otherwise be considered high quality." If the original poster (for example) otherwise satisfies the requirements in this article, then does the posting (still) deserve to be deleted simply because the original poster is bringing an alternate approach to a previously discussed Math problem? Was my posting unclear on its focus? $\endgroup$ Jul 17 at 14:47
  • 2
    $\begingroup$ Personally, I find "not-of-general-interest" to be inherently subjective (for example, do niche topics fall in this category; if so, does that mean they should be closed/deleted). Further, imo, inevitable deletion disregards the effort put into answering (and asking) the question, especially if the underlying problem was nuanced or required considerable effort to resolve. $\endgroup$
    – user1082389
    Jul 17 at 15:50
  • $\begingroup$ @BeKind I am attempting to roll with the punches. The previous meta-post, which this meta-post is following up, focuses on a MathSE posting that was closed, and would have been deleted,and only because the MathSE posting referred to an underlying question that was itself a duplication. Rather than trying to overturn the entire culture of removing such questions from the repository I am attempting a compromise, that would allow the repository to be kept pristine, and still allow those posters who deserve an answer to receive one. ...see next comment $\endgroup$ Jul 17 at 16:42
  • $\begingroup$ @BeKind I successfully lobbied to re-open the underlying MathSE posting. At the same time, in discussion with a moderator at the Cured chatroom, I was unable to convince the moderator that the question should be re-opened. People aren't chosen to be moderators at random. I can't ignore the perspective represented by the moderator, regardless of my feelings. So, I am attempting to find a compromise that would in my opinion, be an improvement on the current state of affairs. $\endgroup$ Jul 17 at 16:46
  • $\begingroup$ @user2661923 Evidently, my experience on this site is limited, so I'll speak from a personal point of view. In my opinion, all solution verification questions without an easily discernable mistake that subjects it to duplication should remain open and undeleted on this site (this seems to be contrary to the general site consensus). Nonetheless, as a compromise, it may be valuable to initiate a chatroom for such kinds of questions, which would be easier to implement and can satisfy all parties involved. $\endgroup$
    – user1082389
    Jul 17 at 16:49
  • $\begingroup$ @BeKind The Cured chatroom already exists, and for just that purpose. The difficulty is that any discussion of a specific MathSE posting rests on what premise is prevailing as to how a MathSE posting should be managed, when it provides a different analysis on an underlying MathSE question that is a duplicate. I am attempting to change the underlying premise of how, in general such questions should be managed. This is distinct from debating a specific posting with respect to whatever premise is prevailing. $\endgroup$ Jul 17 at 16:55
  • $\begingroup$ @user2661923 I meant initiating a chatroom where users can help solve the questioner's SV problem, not a chatroom to discuss it's closure/deletion/etc. $\endgroup$
    – user1082389
    Jul 17 at 16:58
  • $\begingroup$ @BeKind My mistake for misinterpreting your intent. Your idea is a good one that is often used. However, as a general procedure, it will probably not be effective with original posters. For one thing, there may be a minimum rep requirement for a new poster to join a chatroom that someone else has initiated. For another, discussing the problem in a chatroom is not typically as effective as letting the original poster see a MathJax composed, line-by-line analysis/dissection of the original poster's analysis and the corresponding underlying question. ...see next comment $\endgroup$ Jul 17 at 17:03
  • $\begingroup$ @BeKind However, you bring up an interesting point, which is the exact same point that this meta-posting is attempting to focus on. You have suggested that the easier compromise may be to resort to such an answer-oriented chatroom. The major question of this meta-post, which no one has yet responded to, is how difficult would it be to implement a refinement to a quarterly site maintenance procedure? $\endgroup$ Jul 17 at 17:06
  • $\begingroup$ @user2661923 I stand by my point that leaving SV questions (that are not obviously duplicates) open and undeleted is the best course of action. Unfortunately, I find that other solutions are victim to some or the other problems. As expressed before, I'm not in support of an automated deletion system. $\endgroup$
    – user1082389
    Jul 17 at 17:10
  • $\begingroup$ @user2661923 I remember talking about the "follow" tool in an earlier discussion. Follow questions, and once their job is apparently done (perhaps in $7 $ days, say) then vote to close (it should be closable, that isn't much of an issue) and finally vote to delete eventually. I put it this way : we know the window is "broken", but the broken window is still serving a purpose to some user, so we can keep the window broken for some time and then fix it once the user's purpose is done. The problem is two-fold (1) Monitoring broken windows, the bigger problem however is ... $\endgroup$ Jul 18 at 13:04
  • $\begingroup$ ... that people start breaking windows because it's too difficult for them to enter the building (ask a "good question") otherwise, and we start getting many more such questions because of our part-time leniency, than good ones. We should still ensure that good ones are the ones to strive for. That balance is important. Otherwise, I can use the follow tool on questions you suggest (ping me on this thread if you want me to follow questions and I'll try it) and we can see how the theory goes with a one-person (i.e. me) cleaning up the broken windows. I can try this. $\endgroup$ Jul 18 at 13:05
  • 2
    $\begingroup$ @BeKind Actually , for SV questions, we should put our effort into making sure that they are searchable. What I'm saying is, someone searching for "prove that xyz is algebraic" should not land upon someone else's SV question "is my proof that xyz is algebraic correct?" (unless they are mature enough to reconstruct the proof from scratch, which is great!), but should land upon an answer to a question "prove xyz is algebraic" (ideally the best of not too many duplicates), with a canonical answer. SV questions can cloud searches otherwise. With good editing of SV questions, we can succeed. $\endgroup$ Jul 18 at 13:11
  • $\begingroup$ For starters, too many questions which are actually SV aren't tagged like that (and vice-versa). So tagging will ensure that SE's search functionality can avoid (or restrict to) SV questions. I have actually changed my entire perception regarding attempts on a question recently, but that's for a different thread. That, however, basically captures one of the big problems with SV questions, that I think we can avoid if we all wish to work towards it. SV questions needn't be tagged duplicate and closed if they are easily searchable and not obfuscating searches otherwise. $\endgroup$ Jul 18 at 13:13


You must log in to answer this question.

Browse other questions tagged .