-2
$\begingroup$

I am mostly involved with the stackoverflow community of the SE sites. There are some posts which are self-answered: the question lists a bunch of issues, and the answer is going over them one by one. Here are two examples: example1, example2.

I have found many good posts here on orbits / conic sections. For example this one. I haven't found an extensive description of how to deal with Keplerian orbits, something similar to the Pandas examples from Stackoverflow.

I think I am capable of writing an extensive post containing most of the issues people encounter with when dealing with orbits. Are these Stackoverflow-like all-in-one posts welcome here?

$\endgroup$
3
  • $\begingroup$ A similar post on relation posted recently (probably not exactly the same: the post I linked is still quite specific). The reception is okay (+9/-5 for the post and +7/-3 for Xander's answer). $\endgroup$ Mar 31 at 12:40
  • $\begingroup$ That's indeed not that bad, nice example! $\endgroup$
    – zabop
    Mar 31 at 12:45
  • 1
    $\begingroup$ Maybe not exactly the same thing, but maybe loosely related: Would “organizer posts” be useful/welcome here? $\endgroup$ Mar 31 at 15:26
2
$\begingroup$

FAQ-Style Questions

The linked questions are examples of FAQ-style questions: the question post poses a number of (possibly related) questions, and then each answer seeks to address one of those questions. Generally speaking, such questions are not on-topic on Math SE. We expect questions here to be narrowly focused, and centered on a single problem or mathematical confusion.

FAQ-style questions are, as rschwieb suggests, less searchable. They are also more prone to disorganization, as users other than the original asker/answerer may have alternative answers which are likely to get lost in the shuffle. It would be better to pose each question independently (though it might be reasonable to interlink those questions).

Self-Answering

It also seems that the question of self-answering is implicit in the question, i.e. is it acceptable to self-answer a question?

In general, asking-and-answering a question is perfectly fine. Indeed, the software and design philosophy of SE encourages users to answer their own questions. However, there are some things to keep in mind:

  • One of the primary goals of Math SE (and the SE network in general) is to build a repository of questions and answers which are of interest and use to a large number of people. The goal here is not to create a solution guide to a particular text book, nor to show off esoteric knowledge. When you ask-and-answer a question, you need to make sure that both the question and answer meet the site quality standards (indeed, we might even ask for a little more, since there is an implicit claim that the question is going to be helpful to a large number of users—it certainly isn't going to be terribly helpful to the asker, since the asker already knows the answer).

    The site standards are summarized in the meta topic How to ask a good question. Note, also, that any Q&A in this style should contribute something new to the site, and should not (except under very rare circumstances) duplicate an existing question.

  • A similar, though distinct, situation in which self-answering may be desirable is in the context of a question rewrite. In this case, it is perhaps expected (though by no means required) that you will want to answer the question yourself. Again, just make sure that the work you do complies with site guidelines.

  • It is also not unreasonable for an asker to pose a question, then have answers (or hints) given in the comments. In such a case, it is entirely appropriate for the original asker to take what they have learned in the comments and write an answer. Again, please ensure that the question and answer meet site quality standards.

$\endgroup$
3
  • $\begingroup$ I'm inclined to think this is far for forgiving of users perspectives: a position that would agree fully with your second bullet, but hesitate on the first. I'm glad that you clarified that PSQ-looking questions, followed by an answer by the same asker of the PSQ is NOT appropriate. I also think the first bullet would be improved by ruling out its use when the question is a duplicate. $\endgroup$
    – amWhy
    Mar 31 at 19:17
  • $\begingroup$ @amWhy I've made some modifications to address your concerns (and suggested another use case). I felt that "don't duplicate" was implicit in the comments regarding quality, but it can't hurt to be explicit. $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 19:34
  • $\begingroup$ More mispellings and typos I made in first sentence (per a brief discussion elsewhere!). I meant only to say that your position seemed far to lenient wrt such question/self answer questions, compared to a great number of users. But no matter now, because I think your adjustments make it clear when and when not, it is appropriate to ask a question, and then answer it. Thanks $\endgroup$
    – amWhy
    Mar 31 at 19:43
4
$\begingroup$

Posts like the two you linked (this and this) would not be appropriate for this site.

The problem is that asking multiple questions in a post is going to be less searchable and the format makes it awkward for people other than the self-answerer to answer.

Note that I do not say self-answering is of any issue. Indeed, we encourage self-answering, but usually expect that you originally faced some problem finding the answer. It's not recommended to make question-answer posts when you are really not facing any problems with the question. It is a better use-case for people who figure out their own problem after being stuck initially. Exceptions to this general rule would be made according to the quality and value of the post.

$\endgroup$
27
  • 2
    $\begingroup$ I'm not sure that I agree with the statement that a question should be asked before an answer is known (to the asker). If someone has a good Q&A pair, i.e. a question of general interest and a good answer, and that Q&A pair does not yet exist on the site, I don't see any reason why one should not post the question and answer. Of course, every question I've asked is of that kind, so maybe I'm biased? $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 13:25
  • 1
    $\begingroup$ @XanderHenderson To me, that strays too far into the depths of treating the site like a repository. The perverse incentives fueled by the ability to post any number of question-answer pairs do not seem palatable. I guess I will emphasize what I'm saying to be that it should not be absolutely prohibited but it should also not be encouraged unless the post has some exceptional quality. $\endgroup$
    – rschwieb
    Mar 31 at 13:35
  • 1
    $\begingroup$ @XanderHenderson I believe I have, at one point in the past, named a user who had a hundred or more posts that amounted to creating a solutions manual for a book, in self-answered style, on this site. At least when the user faces an actual problem, they are naturally limited by their working attention. That is, one would not be likely to be actively engaged with a dozen or more problems at a time. $\endgroup$
    – rschwieb
    Mar 31 at 13:37
  • 2
    $\begingroup$ I do not mean to suggest that any ol' question should be asked-and-answered, but I find the idea of restricting self-answers to those questions which are answered in the process of asking, commenting, etc to be far too constraining. There are other circumstances in which questions are reasonably self-answered, e.g. when the question is of broad interest. $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 14:08
  • $\begingroup$ And, as with anything on this site, quality matters. A user who posts hundreds of low-quality Q&A pairs ought be discouraged (by the community, or via moderation tools). $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 14:10
  • 1
    $\begingroup$ @XanderHenderson Like I said, I am not suggesting a prohibition, just a reasonable quality bar. I don't know which one you mean by "restriction." I gather you do agree with restriction to prevent the misuse I'm worried about. I don't think "not recommended" can really be labeled a restriction. $\endgroup$
    – rschwieb
    Mar 31 at 14:29
  • $\begingroup$ @XanderHenderson reviewing again, I can see how the "stipulate" sentence clashed with the next sentence and could have led to this interpretation. I've edited to rectify that. If there's any disagreement left I don't see that it's with what is written currently. $\endgroup$
    – rschwieb
    Mar 31 at 14:35
  • $\begingroup$ Downvoter, you. may want to explain if you disagreed with the latter half of the post (new edition or old) or if you are a fan of multi-part questions and answers. $\endgroup$
    – rschwieb
    Mar 31 at 14:38
  • $\begingroup$ It reads better, but it still seems like you are implying that the only (or best?) use-case for self-answering is if an asker discovers the answer in the process of getting feedback on the question. However, anyone motivated to read the comments here will likely figure it out. $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 14:51
  • 2
    $\begingroup$ @XanderHenderson I've tried my best to eliminate the interpretation using only. I am not sure there is a better use-case. But certainly it is the main use-case, is it not? $\endgroup$
    – rschwieb
    Mar 31 at 14:58
  • $\begingroup$ I have no idea what the main use-case is. Do you have any data to support a conclusion? Regarding good/better/best, my favorite use case is in the context of rewrites. $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 19:31
  • $\begingroup$ I agree with you, @rschwieb. I'm not fond of folks posting a question they solve two years ago, and thinking "I'm going to show just how clever I am and maybe earn some good, quick rep for it". Certainly there are rare altruists who see a question that has no canonical answer, and they are committed to providing one. But from my experience over ten years, most of the times I've seen someone post a question and instantaneously post an answer, they already were, or soon became, "show boaters", folks who wanted to be seen as clever, or folks who were seeking rep. $\endgroup$
    – amWhy
    Mar 31 at 22:19
  • 2
    $\begingroup$ @rschwieb To get to my point about agreement, I am fine with self answered posts as long as they are good questions (context etc.) and were, initially, a problem the OP struggled with. Even if posted seeking help, and they later, maybe after some comments, figure it out, and then post as an answer. $\endgroup$
    – amWhy
    Mar 31 at 22:21
  • $\begingroup$ And @Xander you failed to address my point that many folks who think they have a clever answer to a question, post the question, and an answer .... which turns out to be a dupe, due to their failure to search in the first place. One's own perception of competence and cleverness does not justify subjecting the entire site to a show-off. $\endgroup$
    – amWhy
    Mar 31 at 22:25
  • $\begingroup$ @amWhy In my answer, I wrote: "The site standards are summarized in the meta topic How to ask a good question. Note, also, that any Q&A in this style should contribute something new to the site, and should not (except under very rare circumstances) duplicate an existing question." This seems to address the issue of duplicates... I also addressed the issue showing off, i.e. the point is not to show off esoteric knowledge. $\endgroup$
    – Xander Henderson Mod
    Mar 31 at 23:44

You must log in to answer this question.

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