0
$\begingroup$

I'm reading this post about enforcement of quality standards, and it puzzles me.

I feel like most of the time, it a user were able to ask a question that follows the quality standards, they would actually have enough knowledge to answer their own questions. A lot of the activity in the first phases following the asking of a question are actually about rephrasing the question and better understanding what the asker means.

A typical example :

OP : help me prove X, i tried this and that.

Comments : several people try to understand the question better

Answ 1 : X does not hold, here is a counterexample

OP => Edit question, realizing there was a caveat, rendering Answ1 moot

Answ1 gets deleted by original poser (no longer relevant), finds a new caveat and posts it ans Answ2

OP does not reply. End of day, Answ2 does not get upvoted or downvoted in the following ~ 2 hours

Next morning : Answer2 was heavily downvoted by community and closed, probably because most of those who voted to close did not read the discussion in the comments (especially those of the already deleted Answ1), and could not keep up with the edit timeline.

Final results : the people who took time to help OP are left disappointed as they feel MSE does not reward their efforts appropriately.

I personally think that this process is most valuable to the OP (since the process itself helps the OP, not the actual anwser), even though it is completely against the new policy, so here are my questions:

  • Is my analysis "correct", in the sense that I am not alone in feeling this way ? Am I way off, and if so, why and how ?

  • Isn't "helping the OP" just as important a concept as "keeping long term good quality standard" ? Especially knowing that if the OP could frame the question in a quality standard compliant way, then they could probably answer it themselves.

  • Should this kind of low quality posts be "translated" afterwards into a clear "good question / good answer" type interaction, somehow "rewriting history" ? If so, who would do it (especially when OP has had its answer and is not interested in the question anymore).

$\endgroup$
19
  • 8
    $\begingroup$ I may misunderstand you but this post seems to describe the asking of a question as an interactive process with discussion, which is a forum, not a Q&A site. That would make this inappropriate for the site, though perhaps suitable for chat (I think there's a chatroom specifically for requesting feedback on questions that should work like this). Could you clarify, please? $\endgroup$
    – postmortes
    Jun 10 at 10:11
  • $\begingroup$ Indeed, this is my point. I am not trying to say that this iterative process should be the norm, rather only that it is a type of interaction that I often encounter and it is helpful to the OP. If I understand your comment, you hint at the fact MSE is not the right place for this type of interaction ? $\endgroup$ Jun 10 at 10:24
  • 13
    $\begingroup$ If only all the askers would participate in this kind of interaction. I think that a typical scenario where EoQS bites is that somebody copy/pastes their homework here, and somebody else answers. We have a long history of kicking those who only dump their homework here. This did not stop them from asking (giving the site a bad reputation), because somebody else still did their homework for them. EoQS chiefly seeks to stop those answerers who just answer anything they can without going through the steps you describe. $\endgroup$ Jun 10 at 10:29
  • 10
    $\begingroup$ The question "Help me prove X" when it is not clear that X is true is terrible as a question. There is no justification for this except poor writing and communication. To avoid this is precisely on of the goal of the standards. I do not see why it would be impossible to follow. It's basically common sense to be clear about the context. $\endgroup$
    – quid Mod
    Jun 10 at 10:31
  • $\begingroup$ I think quid has expressed it quite well in his comment, but yes, I think that the chat is more suited to interaction, while the questions are a repository of (good) questions and answers by design. $\endgroup$
    – postmortes
    Jun 10 at 10:33
  • $\begingroup$ @JyrkiLahtonen : Thanks, this is very helpful. $\endgroup$ Jun 10 at 10:34
  • 2
    $\begingroup$ @quid : I kindly, though wholeheartedly disagree. Trying to proove things that are actually wrong, and going through rephrasing steps seems to me like the soul of the research process, and I've gotten very valuable help from this kind of questions in the past. $\endgroup$ Jun 10 at 10:34
  • 1
    $\begingroup$ The iterative process is in general good, but I think the process should be recorded in the question itself (comments are disposible - they can be deleted - while questions and answers are never truly deleted), e.g. they could edit the question with the results of the discussion. Also, the action "OP => Edit question, realizing there was a caveat, rendering Answ1 moot" should not have happened. Instead, they should have posted a new question, linking back to the original with a brief explanation of how they are connected (for caveats, see here). $\endgroup$
    – user1729
    Jun 10 at 10:34
  • 2
    $\begingroup$ I agree with many things user1729 said. But actually I think that when this process renders an answer or three moot, this is a win, and the moot answers can be deleted. I know that I am probably in the minority here, and historically many have "protected" the answerers. As if the answerers were kids sitting in an exam simply answering whatever the teached asked them on the occasion - even when an omission or oversight made the question trivial. This has given me pimples for many years now. I don't think the answerers are entitled to such protection, when the goal is to produce good math. $\endgroup$ Jun 10 at 10:40
  • 10
    $\begingroup$ @G.Fougeron it's not clear with what you disagree precisely. The context should be laid out clearly to avoid misunderstandings. If the context is "I hope to prove X. But I do not know if it is true. I would also be interested in a slight weaker statement along these lines." then it simply should be presented as such. There is no reason whatsoever to condense this to "help me prove X." $\endgroup$
    – quid Mod
    Jun 10 at 10:45
  • 3
    $\begingroup$ In my opinion the EoQS is somewhat statistical in nature. Again something where I may cherish a minority opinion, but I am not overly concerned about a new user accidentally answering a low quality question or four. On the other hand, the users who have answered thousands of low quality questions won't get much sympathy from me. The reason I am probably in a minority here is that we have zero history of different rules applying to different users. Anyway, I think it is fair to expect more from the veterans of the site. $\endgroup$ Jun 10 at 10:45
  • 1
    $\begingroup$ @JyrkiLahtonen I think it depends on the counter-example(s). In particular, if the omission "made the question trivial" then these answers should be deleted (the link in my comment talks about "trite" counter-examples, which is pretty much just the same thing). $\endgroup$
    – user1729
    Jun 10 at 10:54
  • 3
    $\begingroup$ No it is not nit-picking it is the core of the problem. That's no a subtle point about language. What you sketch out shows precisely why what I said is the way to proceed. It seems to avoid all or at least most of the problems you point out with very little downside. $\endgroup$
    – quid Mod
    Jun 10 at 11:14
  • 4
    $\begingroup$ It strikes me as borderline absurd to argue that for reasons of language we cannot expect a poster to make clear if they know for a fact (at least for all practical purposes) that what they want to prove is true and merely search for a proof (as in an assigned work for example) that what they want to prove is true or if instead they suspect or hope it to be true and try to establish what is true. Especially if at the same time one assume that a productive exchange with the poster in comments is possible. $\endgroup$
    – quid Mod
    Jun 10 at 11:17
  • 3
    $\begingroup$ @user1729 You said "Also, the action "OP => Edit question, realizing there was a caveat, rendering Answ1 moot" should not have happened." I agree, but I don't entirely agree with your solution. The answerer is, in part, responsible for creating the problem, as they answered a question which was not clear. Once the situation has gotten to the point that there is a mismatch between the question and answer(s), then they only appropriate course of action may be to delete the mess and start over with a new question. But it would be far better if we never got into that mess in the first place. $\endgroup$
    – Xander Henderson Mod
    Jun 10 at 20:52
4
$\begingroup$

As stated in the question here, the interaction would not be appropriate for Math SE. If a question is unclear, community members should use the comments to ask for clarification, and to help the asker improve their question. The ideal interaction is

Asker: Please help me to prove X.

Commenter(s): I don't understand the question. Can you please explain?

Commenter: X does not hold, here is a counterexample.

Asker: Oh, I mean to exclude those kinds of cases. [Edits post]

This same process should iterate until either

  1. The question is improved to the point where it can be meaningfully and authoritatively answered, at which point folk should feel free to point an answer.

  2. The asker abandons the question in a less-than-high-quality state, at which point the question should be closed and deleted without being answered (note that a potential reason for an asker abandoning a question is that they get an answer in the comments, which then makes them feel it is unnecessary for them to pursue the issue any further—this is a perfectly fine outcome, followed by closure and deletion).

The point is not that an asker should come to the site with the mathematical maturity and knowledge to ask a good question. Rather, they should be engaged, and implement feedback. Even if they don't arrive here knowing how to ask a good question, the community can help them pound a poor question into shape.

Addressing a couple of your questions more precisely:

Isn't "helping the OP" just as important a concept as "keeping long term good quality standard"?

Bluntly, no. The primary goal of Math SE is to connect askers to answers. The ideal use-case is that an asker arrives at Math SE, searches the existing database, and finds something which helps them answer their question. Every new question asked here should, in principle, fill some "hole" in the database. As such, a new question should (1) actually be new (and not a duplicate) and (2) should actually fill a hole in the database of knowledge, in the sense that it is likely to be of use to others in the future.

"Helping the OP" is one goal, but it is not the primary goal of this format. If the only person helped by a question is the original asker, then the question isn't a good fit for this site.

Should this kind of low quality posts be "translated" afterwards into a clear "good question / good answer" type interaction, somehow "rewriting history"?

Yes, absolutely. Ideally, the original asker would take the time to improve their question so that it meets community standards and can be answered. If the asker abandons the post, then we have guidelines for posting a new question.

I don't think that "rewriting history" is a problem at all. The goal is to produce a polished artifact at the end. The process by which we produce that artifact is far less important. If the end result is that the original asker improves their question, great! But deletion of the original question and the inclusion of a higher quality rewrite as a new question is just as good with respect to the overall mission of Math SE.

$\endgroup$
1
  • 2
    $\begingroup$ Thanks a lot for this detailed answer. It truely helps my understand the "goals" and "philisophy" MSE sets for itself. I think part of my misunderstanding stems from the fact that the point system in the SE communities has always felt (to me at least) like a kind of currency that one could exchange (though bounties) in order to attract higher visibility. I guess the preconceived idea that I need to revisit is that "good deeds" are rewarded with points. Mostly, high quilaty answers (with high enough engagement) are rewarded by points. The reward for the rest is a well groomed site I guess ... $\endgroup$ Jun 11 at 12:18

You must log in to answer this question.

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