Since asking this question and answering it myself (having read Jeff Atwood's take on this on mSO), I've done a little more reading and realized it is somewhat controversial.

I guess each community decides for itself how it feels about this practice. How do we feel about it here on dba.se?

Do we prefer people

  1. not to ask/answer questions they already know an answer to
  2. to wait a certain period (1h?, 1d?, more?) before answering
  3. to answer immediately to save others the trouble of re-doing your research
  4. to use the CW flag (on question or answer or both)


There is a new feature that allows a question and answer to be submitted at the same time - making it all the more obvious that self-answering is acceptable and even to be encouraged


4 Answers 4


This site should have no preference. It should be entirely up to the OP when and whether to answer their own question.

When I ask a question I know the answer to, sometimes I don't answer immediately because when I do my answer may limit the scope or potential for other answers. Other times it is so obviously the correct answer then I post it immediately. Only the OP would know whether this would be useful or not.

I see no reason to suggest a limitation on the the OP. The quality, number, and depth of answers is not limited in any way if the OP answers. In almost every way it is no different than if someone else answers the question. Excellent answers are great and bad/mediocre answers will be out-voted by competing answers.


I personally think that posting the answer to your own question is a great idea.

There have been times when I run into a question, do the research, solve the problem and promptly forget it. If I'm able to ask the question that I want answered and then immediately answer it, it's a way to 1) save my knowledge for myself later and 2) pass my research along.

Often, researching problems takes considerable time and I like to save myself and others from repeating this effort.

Having said all of that, there is currently a wait time to answer your own question. It only applies for people with reputations under 100. But, the wait time is 8 hours, which I find to be a bit too long.

I can understand the 100-rep minimum, but I'd like to see that wait shortened to 4 hours at most.


I favor to wait at least 3 hours before answering ones own question.

This is to give those who started immediately to answer the question time to finish their answer, before being confronted with the solution by the OP.

Here on dba we often have to setup some test cases and try out things first.

It is not the simple writing down of things one knows already or have in there private toolbox.

When after that time there is no suitable answer, I feel it is OK when the OP posts his answer.

The argument that it saves others the trouble of re-doing research is not important here. In many cases the solutions can be quite different and cover aspects the OP didn't think of and that diversity is one of the benefits of the answers on this site.

Here we often have different approaches to solve a problem and the important measure is not which is the accepted answer to a question, but the upvotes to the answers.

I think making a question or answer CW to early does not help to get this picture, of what might be the promising attempts.

  • 1
    That's a good point, you don't want to discourage people right away from trying to come up with a good answer, however, if you look to the people with the highest scores, you'll notice that they go ahead and tackle the problem anyways, without regard for the other answers that exist. Also, I would love to see lots of people's takes on the same problem, so go ahead and tackle an interesting question even if it is self-answered. They may give you the checkmark anyways, you never know.
    – jcolebrand Mod
    Aug 8, 2011 at 16:04
  • @jcolebrand, if I remember well, this already happened to on SO. And I've saw happening sometimes: even if the OP answer his own question, (s)he accept another answer if find to be THE answer. Oct 14, 2011 at 23:00
  • @bernd_k: I think the limit would be 24 hours to accept his/her own answer. Answering per se is not a problem. Oct 14, 2011 at 23:02

I once answered a question on MySQL's authentication flaw on installation : MySQL : Why are there "test" entries in mysql.db?. After a week, I accepted my own answer. I said it before, and I'll say it again: It's Like Kissing Your Sister. YUCK !!!

Notwithstanding, it does give a chance to share knowledge and provoke conversation (and potential controversy).

Should anyone ask a quesiton and answer it the same/next day, such question should be moderator-reviewed to see if the content of both question and answer contributes to a learning experience.

You must log in to answer this question.

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