I have read all the discussions around the 48 hour delay on accepting ones own answers. To quote Jeff Atwood:

You must wait 2 days from the time you originally asked your question before you can accept your own answer. This gives other users a chance to answer the question in good faith, and earn the accepted answer.

Let's suppose you ask a question. You finish typing it out, but don't press Post Your Question. Instead, you go off and do a bit of research. You find a solution. Instead of discarding the question, you decide to share your new knowledge with the community because you're just that type of person. So you check Answer your own question after reading that it is explicitly encouraged. You enter your answer and submit both the question and answer.

Knowing that your answer solved the problem, you try to accept your own answer to indicate as much to everyone. You are then confronted with

You can accept your own answer in 2 days.

Now, your intention is to come back in two days and accept your answer.

You come back two days later, but in the meantime, others have provided answers in the hope of getting accepted. However, your intention has always been to accept your own answer because you know it is correct. You have just lured them into wasting their time providing answers that you already know will not be accepted. You accept your own answer and that's that. But not quite... you get annoyed when you realise that your effort to share knowledge has been sabotaged by your correct answer not appearing at the top.

Can we have the system changed to work like this:

  • Provide an additional "Answer and accept your own question";
  • Always keep the accepted answer at the top, no matter whose answer it is;
  • Don't count votes on the question or answer toward the user's reputation.
  • If the user decides to accept another person's answer after the fact, allow votes on the question only to affect the user's reputation.

If you're going to say that you encourage users to ask and answer their own question, you should not sabotage their efforts to do so by demoting their answer. It's just pointless. You already have measures in place to prevent gaming, such as not counting the answer towards the user's reputation.

  • i think you should do something like this in that case I am answering this question Q&A style...leave a note to others – TheFisherMan Jun 2 '12 at 6:32
  • 7
    That's a good example. @Presley: despite the OP posting an answer, others might come up with a much easier solution, or more robust, or more generic, ... It's great that the question is given some more attention by not having the self answer accepted right away. – Arjan Jun 2 '12 at 6:39
  • 1
    The accepted answer status can be re-assigned to any other answer any number of times, so the delay is pointless and just lures people into false hope, unless they genuinely think their solution is better than all others posted. – Distortum Jun 2 '12 at 6:44
  • 1
    @SteveTaylor unless they genuinely think their solution is better than all others posted what if the new answer get more upvotes then OP answer , in that case others also think that its a better solution than the self-answerer's answer – TheFisherMan Jun 2 '12 at 6:47
  • 1
    @PresleyDias it is always up to the person who asked the question to decide which answer to accept. I have seen many cases of the accepted answer not being the highest voted answer for questions where the OP didn't answer. – Distortum Jun 2 '12 at 6:53
  • Indeed, Steve, in many cases the OP accepts a different answer than the community feels is best. So then how did the poster of that higher voted answer waste their time? – Arjan Jun 2 '12 at 7:01
  • @Arjan because the user who asked and answered had the intention of accepting their own question and will typically follow through on their intention. People are generally favour their own advice over others'. – Distortum Jun 2 '12 at 7:10
  • Very good question. I don't understand why it would be rated down. Totally agree with you. – msbg Oct 13 '12 at 23:17

The person who asks the question does not always have the best answer. For example, this is an example where I suspect the OP already had the answer in mind when posting the question, but it turned out to be a bad answer (and bad question).

Had others not stepped in, we'd be left with a QA pair that would poison the Internet.

Yes, self answering is encouraged but should not be done blindly and narcissistically (that's what blogs are for). The 2 day hold period encourages others to contribute to the content, and the best answer should be ranked higher no matter who posted it.

Placing a less ideal answer on top wastes the time of unsuspecting visitors who come by looking for and answer.

  • Nice example. As an aside: note that self-accepted answers are not sticky on top. – Arjan Jun 2 '12 at 8:37
  • Indeed. Which is why better answers should be given a chance to be posted and upvoted to offset self-answers that turn out to be not as good. Q's marked answered too soon may not attract as many eyeballs. – Shawn Chin Jun 2 '12 at 8:52
  • Oh, and the posted example really stood out because the OP was downvoting every answer that came in and yet his own sounded like a badly remembered textbook solution. It's an extreme case, but highlights the reason why the 2 day hold period before self answers should remain. – Shawn Chin Jun 2 '12 at 8:58
  • "Poison the Internet"? I wouldn't go that far. – casperOne Jun 3 '12 at 3:44
  • 1
    @casper how about "cause many sleepless nights"? – Shawn Chin Jun 3 '12 at 9:37
  • @ShawnChin It's a broken window. How much agita that causes you is subjective, but I don't let it keep me up at night =) – casperOne Jun 3 '12 at 17:22

Who cares if the answer doesn't get accepted? That's only a check mark and 15 reputation. If they have a better answer than you and they want to put the time and effort into writing it, chances are that it will generate them upvotes like any other answer. I wouldn't consider that wasting their time. Plus, keeping the question marked as not having an accepted answer will also keep more visitors flowing through it. If there's a brand new question with an accepted answer marked already, how many people do you really expect to visit it?

Anyone can see that your question and answer were posted at the same time and that it's likely that you will accept your own answer. That shouldn't discourage them from posting their own answer if they have further or better information. Like has been said so many times: the accepted answer only marks the one that helped the OP. In your case, your answer helped you. Good job. But, maybe there's that ever-so-slight chance that theirs will be more helpful. You never know.

  • 1
    By your second paragraph, if they can see that the question was answered when it was posted, and they know it will be accepted by the person asking, then the checkmark being applied at the time one answers shouldn't dissuade them either. Accepted answers and knowing what answer will be the accepted answer should have the same effect, so why not allow someone to accept their own answer immediately? – casperOne Jun 2 '12 at 6:38
  • @casperOne: That makes sense, but there's still the traffic issue. – animuson Jun 2 '12 at 6:39
  • @animuson: "Anyone can see that your question and answer were posted at the same time" - Not easily if it has been pushed down. If I came across a better answer after the fact, I would definitely make it the accepted answer. It happens often. Users shouldn't be tricked into thinking that a certain answer isn't accepted when it is - they just don't know it yet. – Distortum Jun 2 '12 at 6:43
  • Re: The traffic issue, perhaps, perhaps not, but one has to look at the long-term effects of accepted answers on posts vs non-accepted answers on posts. I'd be willing to wager that accepted answers get more views over the course of time, but we have no way of really quantifying that... – casperOne Jun 2 '12 at 6:44
  • @casperOne: Yes, but we're only talking about the initial 48-hour period of the question's existence before it gets accepted. – animuson Jun 2 '12 at 6:45
  • 2
    In that case, is it really important that traffic be driven through the question? The intent of the self-answered question was not to gain exposure because you don't know the answer, but because you wanted to share the complete solution. One could argue traffic is more important only when you don't have an answer. – casperOne Jun 2 '12 at 6:50
  • @casperOne: Assuming that the answer is correct and doesn't need commentary from other users. I would argue that it does need traffic to verify the information, verify its on-topicness, and allow other users to pitch in their alternate ideas. If you just create it and let it sit, it could be a bad answer that we'd end up advertising as an only and accepted answer to the problem described, simply because it didn't receive enough traffic to say otherwise. – animuson Jun 2 '12 at 6:57
  • I'd say yes, @casperOne. More traffic means more feedback. When someone has the same problem sometime later, then of course still the self answer might be the best. But the more specialists have looked at it, the more certain we are about that? And reading someone's self-answer to realize it's okay (and then even upvote it) does not feel like a waste of time to me. – Arjan Jun 2 '12 at 6:58

Remember that these sites are not some help forum, but intend to build libraries of high-quality questions and answers. Keeping that in mind, why do you think your own answer is the only or even the best solution to your question? Surely the folks who posted additional answers think their answer has some merit too.

So, no, I think the 48 hours is required to ensure more people read the question and give it some thought. Only then we get the best answers.

your intention has always been to accept your own answer because you know it is correct

That reads to me like: I know I always have the best answer and don't want to learn about other options. That's fine, but that is still not what these sites are about.

  • No, it is a good faith intention. When someone has answered their own question, it is typically because they tried something and it worked so they are sure of it. Other answers may well be along the lines of have-you-tried-this or have-you-tried-that. – Distortum Jun 2 '12 at 6:38
  • 1
    Without seeing a specific example, I disagree Steve. Sure, if the other answers are bad, they should be downvoted. If another answer ends up to be better, then I guess you would eventually accept that other answer, wouldn't you? – Arjan Jun 2 '12 at 6:42
  • 1
    People generally ask questions in SO because they're performing a task and hit a brick wall. The point is, they are performing a task at the time. One would expect, then, that they would post an answer after trying something that does work. My example: stackoverflow.com/q/10860032/584670 – Distortum Jun 2 '12 at 6:48
  • I can see how in that example you would not expect a better answer, @Steve. But there are no other answers in that example. Do you feel that even reading the question and its self-answer to realize it's fine, is a waste of time for people? I don't. – Arjan Jun 2 '12 at 7:04
  • I asked and answered very recently. Give it time. :) – Distortum Jun 2 '12 at 7:15
  • 1
    @SteveTaylor: That something does work does not mean that it is the best solution. Other people can come up with alternatives for many questions that may be better than what the OP thought of. These people should be allowed the chance to post, and the OP should be allowed the opportunity to reward these replies that may be better than his own answer. – Nicol Bolas Jun 2 '12 at 9:42

You have just lured them into wasting their time providing answers [..]

I disagree with this one completely (though it's cited a bit out of context). Instead I'd say you just encouraged someone to contribute potential valuable information that might help future visitors (and the OP if he is interested).

This is very useful sometimes, even if it's not the most helpful answer for the OP. Here is a real-world example of mine. His answer was posted later, but imagine he found that way 5 minutes after posting the question. He would have accepted his answer and I would have skipped the question because it's marked as solved in the question list. Instead I answered and provided an answer that helped a lot of people over time (given the votes), though not the OP directly.

Overall, I have to say you are never sure that your answer get's accepted when you answer someone elses question. This is not a special thing relevant for this scenario. In the end the goal on SO is to trick people into contributing valuable information for others. Rep is one way, but not the only one. This change would reduce the amount of information contributed, just for a slightly faster individual rep gain. I'd say this is a net-loss for the community.

You must log in to answer this question.

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