Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 157 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

The officially up-voted stance on meta-SO is that it is good to self-answer questions. It is promoted in the FAQ and we even have a badge for it (Self-Learner):

Should I not answer my own questions?

Yet, most SO users still seem to have a bias against self-answers. I have, many times, posted a difficult question, discovered that no one on SO had a good answer for it, done the hard work required to answer it, wrote out an answer, and gotten close to no up-votes. Meanwhile, people who gave very preliminary answers to my question before I solved it have many up-votes as if they were the ones who answered the question.

This happened to me, once again, in the following question. I was trying to find the area of an intersection between two circles. Before asking this question, I googled, of course, and found an article on Wolfram that wasn't specific enough.

I posted to SO, hoping for a more specific answer, using the variables I had available in my program. In response, I got someone else's googling, listing the exact same Wolfram article that wasn't specific enough. This made me realize no one on SO had an answer and that I was going to have to do the work of answering my own question by translating the information in the Wolfram article to a more specific solution. Which I did.

As expected, the person who gave me a link to the general Wolfram article I already knew about before posting has more up-votes than my detailed algorithmic answer:

Area of Intersection between Two Circles

So, is there a better way to translate meta-SO's consensus encouragement of self-answering to the broader community?

share|improve this question
I don't necessarily have a bias against self answering unless it's obvious that the asker asked the question solely to answer it themselves (something that apparently happens a lot in the SO c++-faq tag) – Powerlord Nov 23 '10 at 15:38
@R. Bemrose: the official stance is that posting a question solely to answer it allowed – Andreas Bonini Nov 23 '10 at 15:41
@Kop Right. Not to pick on you, R. Bemrose, but this demonstrates to me the kind of disconnect we have in the community between what we have collectively agreed to on meta-SO through reflection and discussion and what many peoples' gut reactions are on SO. Sometimes the gut reaction is the exact opposite conclusion of what we determine by reflection and discussion. I don't know yet exactly how we can bridge that gap and make the policies we determine here reflected in the policies people follow there. – Chris Redford Nov 23 '10 at 17:14
@credford only a fraction of Stack Overflow users will ever visit Meta. In my mind, consensuses born on Meta are something you can point to when there's disagreement - but I don't think we cany expect people to adjust their gut reactions to Meta's findings, that'll never work – Pëkka Nov 23 '10 at 17:20
But on a different note, I don't really see the issue here, seeing as your answer has two upvotes? If there were a huge gap between the most upvoted answer and yours, I could see your point but with those numbers?... I bet that over time, your answer will become the most upvoted one if it is the best one – Pëkka Nov 23 '10 at 17:22
Yeah, that's probably true. When I first posted to Meta, it only had one up-vote. I think my primary reaction was seeing that whole 1 vs 3 situation sitting there still a day after I answered the question and remembering all the other self-answered questions in my profile that ended up that way. This particular question topic may be popular enough that that situation won't happen this time. – Chris Redford Nov 23 '10 at 17:29
@Rejoice Well, there are things we can do and already do with the SO interface that force people's gut reactions to be adjusted. For example, when they first start on SO, they may want to go through down-voting and up-voting all over the place. That's their gut reaction. But we force them to gain reputation first and thus learn more about what the rest of the community thinks are good questions and answers before they themselves can vote. That is one of the possible solution categories I was thinking people could offer. – Chris Redford Nov 23 '10 at 17:33

I've self-answered one of my questions, late, and gotten massive numbers of up-votes for it. So there's certainly community support for that sort of thing.

Keep in mind that folks tend to vote out of self-interest, elevating stuff they see as useful or interesting... Which doesn't necessarily correspond to "exact solutions to the OP's specific problem". This creates a bit of a pitfall for self-answers, because the author is less motivated to explain in detail something they've already figured out for themselves. If you pretend you're explaining the solution to someone who isn't you and may not have quite the same problem you had, you'll likely do a better job of writing for a broader audience - which is what you're doing after all.


Your answer gradually pulled ahead of the other one over the years since you posted it. Writing for the ages ain't all that bad...

We're working on further popularizing the notion of self-answering by integrating it directly into the Ask form - hopefully this will help dispel remaining prejudices against the technique and leave folks to focus on the quality of the content itself.

share|improve this answer
I appreciate the advice but I personally feel like I already use those principles when writing my answers (write for a broad audience). I was really hoping someone had some ideas about how we as a community can reduce the stigma for self-answers and the preference for (inferior) other-answers. It's not just bad for the self-answerer's reputation, it also promotes sub-par answers when a better answer is available. – Chris Redford Nov 29 '10 at 16:09

First of, in your example, I would have mentioned the kind of research I did, and if I had an issue with what I found. If you mentioned beforehand that the Wolfram article wasn't sufficient in its answer and why, maybe it would have motivated someone to answer with a more complete answer, instead of being contented with what was already given.

Furthermore, for each breakthrough made during the search for an answer, I would have made an update to the question to reflect the new information. Commenting incomplete answer in addition could also help motivate the answerer to dig a bit deeper. Also the comment would show the voter that it is incomplete. Down-vote "could" help, but it is to use carefully.

At last, if you find the answer and it was not given, simply answer the question, and maybe add a last update edit to point that you find the answer and how.

tl;dr version: Show that you do the legwork and edit often. It would legitimize the answer in the eye of some users.

share|improve this answer
Thanks for the suggestions. I am a little wary about editing too much. I took the time to carefully craft and edit a detailed update to hard problem and got screwed on it because my many edits turned it into Community Wiki. Now the question is sitting on my profile and getting many up-votes that will never effect my reputation probably precisely because I put so much effort into editing it:… – Chris Redford Nov 23 '10 at 17:23

I doubt this had anything to do with self-answering, honestly. Misnomer's answer was in 3 minutes after you posted the question, when the question was probably getting maximum traffic. You posted your answer nearly 2.5 hours later, when the question was long gone off the "newest questions" list and probably had far less traffic. You never mentioned in your question that you'd already read the Wolfram article, so of course someone was going to post it, and Misnomer's answer seems good -- I would've upvoted it too. Someone who posted a good answer immediately having more upvotes than your hours later answer is unsurprising, and a 2 upvote difference is if anything low

share|improve this answer
Yeah, that's a good point about the timing. It sucked in this case because getting the correct answer took about an hour of sifting through the Wolfram article and related articles. I guess the answer will probably get up-voted in time if people find it useful. – Chris Redford Nov 23 '10 at 17:10

You must log in to answer this question.

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