-15

Summary

I think it would be beneficial to mark self answered posts, where people only "ask" the question in order to share their knowledge/findings, in a way so that it is obvious to everyone checking the question out that this is not really a question that is seeking an answer. And most importantly: that the answer given is indeed the one that solves the problem for the person that asked the question.

Why do I think this would be beneficial?

  • This would prevent someone from investing time, trying to help OP to find a solution (note that we can only mark our own answer as the correct one after two days (on Stack Overflow) so it is not immediately obvious that the provided answer is sufficient already)
  • It could prevent some people from downvoting either the answer or the question itself

Especially the last bit tends to be very frustrating.

The story behind this question

Let's consider an example of mine: Message "error: use of undeclared identifier 'assert'"

At first glance this problem looked like dozens of other questions already posted on Stack Overflow (I know, because I did indeed try to find an answer there before). In my particular case however my situation was slightly different. Therefore after having failed to find a solution with existing Stack Overflow questions, I decided to share my findings after I had figured it out.

The solution turned out to be very specific to my problem, but it seemed reasonable enough that someone else might run into the same issue, so I decided to share it nonetheless.

Me posting this self answered "question" resulted in the following

  • My question getting downvoted (as there was no comment made as to why, I have to guess: I think someone saw the question and though "This has been asked here before" and thus downvoted)
  • My answer getting downvoted (again without explanation, so here's my guess: The answer doesn't seem to be very general and without taking into account who the question actually asked, it could create the impression as if this was just a pseudo answer trying to catch a few accidental upvotes or something like that)
  • Another answer was created (that contained the standard answer for the standard problem, that I however did not have). Given that I wasn't looking for another answer in the first place, this just means that someone invested some time trying to help, where there was nobody actually seeking help.

I think chances are that all of this could have potentially been prevented if the question was marked as Q&A-style immediately after posting.

How could this "marking" look like?

I am not a good UI/UX designer, so there are probably better ideas for this, but here is my suggestion:

  • Change the background color of the question and the original author's question to something that is different from the standard background. This should create a connection between the question and the answer that is only present if the question was "asked" in a self answered manner
  • Add a label somewhere next to the question stating that this is self answerd-style (optionally including a short explanation of what this means)
  • Accept the written answer immediately, so it is clear that the problem has been resolved
30
  • 13
    Other answers are to be encouraged always. Just because you weren't looking for that alternative answer doesn't mean nobody else is. The audience for Q&A is everyone. Mar 2, 2021 at 9:11
  • 6
    But that's my point we really shouldn't answer questions to help the OP, we should answer to help everyone. The needs of the many outweigh the needs of the one. Even having an accepted answer shouldn't put anyone off for that same reason. Mar 2, 2021 at 9:16
  • 1
    That seems like a very poor explanation to me. Being "useful" in terms of a Q&A question is basically out-of question as the answer given is useful. More or less by definition. You might argue about downvoting the question, but doing so for the answer seems like someone anticipating that the solution that OP has found, is not useful for OP. I mean I also won't downvote someone's answer to a question just because it didn't help me. I only do so, if I think this is unhelpful to OP and me and everyone else that might visit the question
    – Raven
    Mar 2, 2021 at 9:26
  • 8
    Please note that the quality standard for a "normal" Q/A and a self-answered Q/A is the same. Your example question could benefit from an MCVE. The issue I see is, that the question can most likely only be answered correctly by guessing.
    – honk
    Mar 2, 2021 at 9:36
  • 3
    Please read the first sentence of my last comment again :)
    – honk
    Mar 2, 2021 at 9:39
  • 3
    Well, I'm not sure whether this a matter of opinion. I have seen complaints similar to yours on Meta (especially MSO) and the answers basically said that you can't apply a different standard to your question just because you know the answer. Well, you can, but then you should not wonder about downvotes ;)
    – honk
    Mar 2, 2021 at 9:49
  • 2
    It is probably not the first time this has happened. What turned up in your research? Mar 2, 2021 at 9:50
  • 2
    Original blog post (though it doesn't contain much guidance): It’s OK to Ask and Answer Your Own Questions Mar 2, 2021 at 9:50
  • 3
  • 2
    @Raven how can you share your findings if you can't give a good and comprehensive explanation of your problem? This is really key to the concept of sharing information. No-one will benefit from your solution if it is unclear what problem it is solving. Also if a question lacks details it will, and should, be closed for that. Closure is often a route to deletion.
    – Luuklag
    Mar 2, 2021 at 9:55
  • 1
    Well this example of yours is indeed a bit different, and it would suffice as a question, albeit a very mediocre one. Also I can't help but feel that your answer has plenty of potential homes in the network. There are a lot of questions about this error message, and your answer would have multiple questions to which it could be added. Simply state there that the presented solution didn't work for you, and that the thing you did solved the problem. There really was no need for a new question.
    – Luuklag
    Mar 2, 2021 at 10:02
  • 4
    "It is not intended to be answered by others." If that's really the case, then perhaps it doesn't belong on Stack Exchange. The Help pages on every site state that (in theory) self-answered questions are welcome, but such Q&A pairs are supposed to be evaluated like regular questions and answers, they do not get special privileges.
    – PM 2Ring
    Mar 2, 2021 at 10:02
  • 1
    Yours seems to be that a Q&A question should be asked in a way that allows others to answer it. Well, to my understanding this is a core feature of SE. Mine is that a Q&A question should be seen as sharing my findings and thereby documenting the solution. The most frequent reply I have seen to that view is: "Maybe a personal blog would be more suitable for you." Anyway, I think you would have a better time playing by the rules than insisting on your opinion ;)
    – honk
    Mar 2, 2021 at 10:05
  • 1
    There is no guarantee to avoid downvotes. Even excellent questions get them. Perhaps someone mis-clicked, or doesn't like you or the topic. There is only one way to prevent them, and that is not posting.
    – Luuklag
    Mar 2, 2021 at 10:07
  • 1
    That's why I said to post that info in a comment: they can only be edited or deleted by mods. ;) You don't need to post all your research, but you should post links to a few relevant related questions (& say why they aren't adequate dupes), so people know that you've done due diligence on checking that it's not a dupe. And to reduce the likelihood that someone will attempt to dupe-close your question with those targets. ;)
    – PM 2Ring
    Mar 2, 2021 at 10:13

1 Answer 1

5

I think it would be beneficial to mark self answered posts, where people only "ask" the question in order to share their knowledge/findings, in a way so that it is obvious to everyone checking the question out that this is not really a question that is seeking an answer. And most importantly: that the answer given is indeed the one that solves the problem for the person that asked the question.

There's already a mechanism for this: You can see how many answers a question has, and if you accept your own answer to show that it solved your problem, it will show that your question has an accepted answer. Here are two examples of the MSE questions page: The first one is self-answered, and as such showed up from the beginning with the '1 answer' box. The second one shows what that box looks like when the answer is accepted: it's very green.

enter image description here enter image description here

In other words: If you don't want to waste your time on already answered questions, you don't have to! You can just click on the ones that have zero answers and aren't closed.

This would prevent someone from investing time, trying to help OP to find a solution (note that we can only mark our own answer as the correct one after two days (on Stack Overflow) so it is not immediately obvious that the provided answer is sufficient already)

Even if you have accepted your answer, this doesn't stop people. This question of mine on Arqade still got another answer after I already wrote AND accepted my own answer.

Like people said in the comments already, your question isn't just there to help you. It's also there to help future visitors. If you have hacked together your own solution that's fine, but if someone else knows a different way of doing the same thing, having a second answer is good. Especially if that answer can explain why it does things differently from your own.

Having people invest time on already answered questions can be good, and should not necessarily be prevented: The first answer isn't always the best one.

It could prevent some people from downvoting either the answer or the question itself

I doubt that. Self-answered questions are the same as regular questions in that regard: Your question and answer should meet the site's standards. Just because you posted it together with an answer doesn't automatically mean your question is a good question for the site.

The kind of self-answered question I see the most is spam, then the next often seen category is low-quality, lacking details or off-topic. Again: Those don't meet the site guidelines, so they get downvotes, close votes, and flags. If your idea is going to prevent that quality control from happening, I'd rather take a pass on having it.

2
  • I know of this mechanism and I do not want to stop people adding answers per se. I just want to be able to signal that the question already has an answer that solved the problem. And as I can't accept my own answer before 2 days have passed, I wanted to suggest a different kind of highlight for this.
    – Raven
    Mar 2, 2021 at 11:33
  • 3
    @Raven The reason you can't accept your own answer within 48 hours of asking is so that everyone has a fair shot at answering as given here. A different kind of highlight would only complicate matters, in that case, find the arguments for the 48 hour limit and argue against those for removal of the 48 hour limit. Still though, the only way to avoid anyone ever writing a different answer to your question is by having it closed, locked or deleted, none of which I can recommend. If people want to spend time writing better answers, let them!
    – Tinkeringbell Mod
    Mar 2, 2021 at 11:44

You must log in to answer this question.

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