0

I posted a question a couple of months ago that got several answers. At the time, I clicked to "accept" the one that seemed best to me, and when I did I got a +2 bonus to my reputation.

In the time since then, I've been reading things that have helped me to come to a much more comprehensive answer to my question than any of those that were posted, so I returned and added my own answer.

According to the FAQ, this is a good thing to do. The Stack Exchange blog even says that

it is not merely OK to ask and answer your own question, it is explicitly encouraged.

Of course if I'm bothering to return after months to post my own answer, that answer will be my new "accepted" one. Yet when I selected my new answer, I found that it gave -2 to my reputation.

I suppose the philosophy behind this must be that it wouldn't make sense for users to be able to auto-generate their own reputation by just creating and accepting their own answers, but two reputation is such a small amount that I doubt this would present a realistic problem. Instead what happens is it feels like a slap on the hand to lose a small amount of reputation in return for having bothered to provide a more thorough answer to the community.

The interests of the user who posted the answer that I had originally accepted should also be considered. Because she already has several votes in her favor, losing my "accept" vote now makes it possible for her to get the gold Populist badge. She loses fifteen reputation, but that seems a small price to pay for a chance at a gold badge, and I doubt most users would object to being given such an opportunity.

Why do you lose reputation for changing the "accepted" answer to your own? Should this policy be changed?

6
  • 1
    It saddens me no end that you see deduction of only 2 reputation as a punishment.
    – Mohit
    Apr 15, 2013 at 3:47
  • 2
    @Mohit My point is that it's a nudge: a small adjustment that could have a conscious or unconscious impact upon the behavior of the site's users. Within the site's system, any points or badges lost work as negative extrinsic reinforcement. The -2 functions as a needless, albeit minor, slap on the wrist for anyone who answers their own post.
    – Graham Herrli
    Apr 15, 2013 at 4:37
  • @3nafish Wow, the greedy vibe in this post really puts me off. It looks to me that this blinds your logic. Of course the +2 is deducted, because you remove the action that got you the +2 in the first place. Arguing against that is just ignorant. And a +1 to Charles who pointed out the fact than the previous answer poster will lose 15 points in the process.
    – AndroidHustle
    Jun 5, 2013 at 8:21
  • 1
    @AndroidHustle It's not greed. (If I consciously cared about small amounts of points, I wouldn't downvote nearly as much as I do.) It's a basic acknowledgement of the reward circuits in our brains: the small amount of negative feedback (-2) for a good action could help subconsciously to train us never to post back with our own answers. (An even smaller -1 certainly seems to have subconsciously trained most of us to upvote more often than downvote.) As to your attributing the mention of the lost 15 to Charles...you might want to re-read my post.
    – Graham Herrli
    Jun 5, 2013 at 13:02
  • @3nafish Ok, I get the minor objection to having the +2 deducted when you marquee your own answer as the correct one. But in your case you got another 30-something from other up votes, surely ones circuits in our brain would recognize and appreciate that. If you get points for that, plus you're keeping the +2 for an action that is withdrawn, as I said it really feels like a case of eating the cake and keeping it too. And the for Populist badge, sure, that would be a nice scenario, but it so RARELY happens.
    – AndroidHustle
    Jun 5, 2013 at 13:41
  • @AndroidHustle Those 30 were for answering the question though, not for accepting it. They're two separate issues. I'm not so certain our neural circuitry would conflate rewards systems for the two, especially given that the -2 follows immediately upon the acceptance, whereas there's a delay before any points for answering, and that the negative reinforcement is a certainty whereas the positive reinforcement is not.
    – Graham Herrli
    Jun 5, 2013 at 14:11

3 Answers 3

Reset to default
8

You're not punished for answering your own question - in fact there is even a badge for doing so. However you don't earn the +2 votes for accepting your own answer.

What happened to you is that previously you accepted another answer and received the +2 reputation for accepting an answer. However now you have un-accepted that answer and therefore that +2 rep was removed. You didn't get it back because you aren't awarded rep for accepting your own answer.

If you were to receive rep for accepting your own answer then it's an opportunity to 'game the system' by asking easy questions, answering them yourself and accepting that answer. Yes, it's only 2 rep, but it's still not what the system was designed for. It rewards other people helping you out.

4
  • Yes, but I'm not talking about "gaming the system." I already had the 2 rep from having chosen a previous answer. What I'm suggesting is that a user shouldn't lose those 2 rep for later deciding to change the accepted answer to his/her own. I agree that there shouldn't be a bonus for accepting my own answer, but there's no reason to have that slight negative nudge either. Why not just keep the rep the same when the "accepted" answer changes?
    – Graham Herrli
    Apr 12, 2013 at 16:32
  • 5
    Because the +2 is a "reward" for recognizing another user's answer, which gives them +15. When you mark your own answer, you lost -2 but they lost -15. They shouldn't get the points if they didn't have the best answer, but you should also not get to keep the two points while they lose 15.
    – Charles Wesley
    Apr 12, 2013 at 19:18
  • 1
    @CharlesWesley I agree with your statement that They shouldn't get the points if they didn't have the best answer but it doesn't follow logically that because they no longer have the best answer, the person who bothered to write a new best answer should lose points as a direct result of having added in that new answer.
    – Graham Herrli
    Apr 13, 2013 at 2:17
  • 4
    You're not losing points because you wrote a better answer, you're losing points because you're not recognizing someone else's answer. The +2 is a bonus for awarding points to another. Your reward for writing a better answer is up-votes by other users. The gist is you don't get to reward yourself, that reward has to come from others. If you choose to not recognize another user's answer, you don't get the points. Pretty simple.
    – Charles Wesley
    Apr 15, 2013 at 17:53
3

Your proposal would allow gaming the system.

  1. Accept someone else's answer. Gain 2 rep points.
  2. Post your own answer.
  3. Transfer acceptance to your own answer.
  4. Profit!

And now you've dishonestly acquired two rep points. If your own answer is better, it should acquire votes anyway, and each of those is worth 10. That's more important.

1
  • 2
    There's a logic gap at Step 3. You have nothing to gain by transferring acceptance to your own answer. It's a way of marking that answer as correct in your opinion, but it doesn't have any other inherent benefit. (Accepting your own answer doesn't move it to the top of the list.) If it's the two points you're after, you would already have those at Step 1, so there's no point in taking the additional steps to "game the system."
    – Graham Herrli
    Apr 13, 2013 at 16:51
2

Why do we punish users for returning to answer their own questions?

Well, we don't. We just don't reward it automatically. You still get +10 rep per upvote on your answer.

The rep bonus for accepting answers is really just meant to encourage people "paying it forward" by accepting answers which makes the question "answered enough" and gives the answerer a boost of reputation. To me it makes some sense to encourage accepting other people's answers rather than always accepting your own, so the +2 rep is a slight nudge in that direction. If you're answering your own question it should really be for a reason better than +2 rep.

Automatically earning reputation, even in tiny amounts, is a risky thing to build into the system. All you have to do is ask, answer and accept a bunch of questions that slip by unnoticed/ignored and you can have yourself some free rep. Not much, but if you can abuse it it will be abused, and I dont' see a big win here for giving people 2 free points for accepting their own answers...I'd much rather we give them larger amounts of rep by humans upvoting those posts.

3
  • 1
    I'm not saying it should be possible automatically to earn reputation. My claim is that if you decide to switch to your own answer a while after having chosen another, you shouldn't automatically lose reputation for changing. With the current setup, the slight nudge to support accepting an answer becomes a slight nudge against adding your own answer.
    – Graham Herrli
    Apr 12, 2013 at 21:38
  • @3nafish But in that case you shouldn't be rewarded for +1's on your new answer. You can't keep the cake and eat it too.
    – AndroidHustle
    Jun 5, 2013 at 8:23
  • @AndroidHustle Why not? Accepting your answer doesn't increase its likelihood of getting +1's. (It doesn't move it to the top of the list like a normal accepted answer.) It's just a way of saying "having thought about this a lot, I've decided that this turned out to be the best answer to my question."
    – Graham Herrli
    Jun 5, 2013 at 13:07

You must log in to answer this question.

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