Consider this question on How to run/debug multiple web application projects with-in the same solution?. I put out a 200 bounty on the question, but did not get an answer from anyone other than myself that worked.

In fact, my answer is also the highest up-voted on it at this point. I know that allowing OP's to retract a bounty can be a slippery slope; in fact I think that once the bounty is put up it should stay up. But why not allow the OP to answer their own questions with a bounty and reward it to them if they receive the same up-votes as others?

Yes, I am fully aware that bounties are a risk. People will use them for numerous reasons such as trying to draw extra attention to their issues. However, in my specific question, the only valid answer is the one that I gave. Had it been anyone else's question, I would have received the bounty, but since it was my own I did not.

  • 1
    (-1) I don't want the 'retract a bounty' feature, and I don't want the 'award the bounty to myself' feature. – devinb Aug 10 '09 at 16:58
  • 1
    Is there any other reason to use a bounty than "trying to draw extra attention to my issues"? I can't currently imagine what numerous other reasons there might be. – Tomalak Aug 10 '09 at 17:40
  • What is to stop people from creating three accounts, and just bounty-swap when the occasion calls for it? – Cloud Jan 17 '14 at 22:24

Bounty is a risk.

Remember, you're not entitled to answers, even on bounty questions. You put your arbitrary number on the line and even then others didn't see the cause worthy enough to answer well.

It sucks, but that's what comes with risk.

  • Welbog is only worth 3,013 theoretical dollars on the internet. – Troggy Aug 10 '09 at 18:12
  • $15,038 if you count all the sites. – Welbog Aug 10 '09 at 18:21
  • 2
    That makes sense, but what if you kept doing research for that question you asked and you find the answer. Shouldn't you be awarded for your hard work!? In my scenario, I spent a few days searching for the answer to this question: stackoverflow.com/questions/25941675/… and eventually found this answer: stackoverflow.com/questions/22832435/… after I posted a bounty. I provided a solution to my problem with help from that link. Shouldn't I be awarded a bounty? – Termato Sep 22 '14 at 17:23
  • 4
    You could have done that research before you posted the bounty, @Termato, and then you'd have gotten your bounty back by virtue of not having risked it in the first place. The idea being, do research before you post a bounty. – Welbog Sep 22 '14 at 17:28
  • 2
    In this scenario, I spent a 2 days researching the topic before I posted my question, then researched 2 more days before I posted my bounty. It's not like I wasn't researching. I just wasn't able to find what I needed to because of the similarities between my issue and Razor View causing errors after compiling the code issue (very similar wording). Yes, I agree, I should have done the research more and now I have to pay for it...but I just thought it would be fair that if the answer was found, why not award it to who found it, regardless of who it was? – Termato Sep 22 '14 at 17:38

There is not a process for this, and I don't expect there will be one. The impression I have gotten is that a bounty is treated advertisement for your question and nothing more. If you don't want to use the bounty system because you don't want to lose the rep, then don't. But if you use it, your reputation is gone; it's the price you pay for the increased likelihood your question will be seen.

  • 4
    Exactly, bounty is not even to consider as "a risk". It's a price you pay for advertising your question. You lose it the moment you ask the question, there is no reason why it should be refunded. – Gnoupi Aug 10 '09 at 15:44
  • 2
    It's not a refund...it's the award for answering the question. If the original asker does the research and eventually answers the question, why shouldn't they get the same reward that everyone else would? – beska Aug 10 '09 at 17:15
  • 1
    @beska, I believe the goal of the bounty is to provide added incentive for users to take the time to provide quality answers to your question. The OP should not need an added reward for solving their own issue, it was their problem to begin with. – Timothy Carter Aug 10 '09 at 18:24
  • Why on earth not? If I post a quality problem that several people might have, and then I discover a good answer that hasn't been posted, shouldn't the site encourage me to post that response? They've paid the bounty. Economically, it's a sunk cost, and they're on the same field as everyone else (with the exception of the 50 bonus points, which could just be removed for users whose answers become accepted) I would also think that the user shouldn't be able to accept their own answer...but if many people vote it up to be accepted, shouldn't that be rewarded? – beska Aug 10 '09 at 23:41

But why not allow the OP to answer their own questions with a bounty and reward it to them if they receive the same up-votes as others?

Because that's missing the point of having a bounty: encouraging other people to tackle a tough problem they can't be bothered to answer otherwise. Sweetening the deal. This is consistent with other features: you can't up-vote your own answers, and you don't get the normal reward for accepting your own answer.

Frankly, i'm still a bit uncomfortable with the concept of bounties period. It's dancing very close to the idea of paying people to answer questions. But there are several factors that stand between them as it stands now: rep has no inherent value, there's no open market for rep, and you must wait two days to put a bounty on a question.

Oh - and you can't retract a bounty or award it back to yourself.

If you aren't comfortable with this, then just don't use bounties. The system should work just fine without them...


People would game the system, since bounty comes with 50 bonus points from SO, it's not a zero-sum game. If you could reward yourself the bounty, we'd have people answering their own questions left and right. Since bounties are not subject to the rep cap, you could, theoretically, build yourself up to Jon Skeet levels rather quickly just by accepting your own bounties.

Ergo: No.

  • 1
    Why the upvote requirement for the award then? Seems hard to game at that point. – RSolberg Aug 10 '09 at 16:00
  • @RSolberg: All you need is two friends or two sock puppet accounts to do it. – Eric Aug 10 '09 at 16:01
  • 3
    Just remove the 50 points if the award goes to the asker. Easy. – beska Aug 10 '09 at 17:16

I'll go out on a limb and say that I think this is a good idea, with some caveats...primarily, that the 50 point bonus should be removed, so as to eliminate any desire for people to game the system.

People are making the point that you're spending reputation for higher visibility, that it's a "risk", and that it shouldn't be refunded. But we're not talking about a refund. You spend the rep for the higher visibility. It's now a sunk cost, and over and done with.

Now, like everyone else who is interested, you do the research on the question, and you manage to come up with the best answer. Why shouldn't you reap the same benefit as anyone else who answered the question? The point of the site is to encourage the best answers to good questions. If the best answer, as voted on by the people who are interested, happens to eventually be posted by the asker, isn't it deserving of the bounty?

EDIT: Shog9 has a point that makes sense: would people spend much time answering bounty questions when the author could just take back the bounty by accepting their own answer? To that effect, I'd say that that's easily solved by not letting the person not accept their own answer on questions with a bounty (can they do that now? I would think not...) They could accept any one else's answer. But if the original author posts their own well thought out answer, and people vote it up enough, it would eventually become auto-accepted. I don't see that as a bad thing.

  • 2
    Would you put time into answering a bounty question if the author could respond, "oh yeah, i figured that out myself" and take back his bounty? – Shog9 Aug 10 '09 at 17:32
  • 1
    No, which is why I point out in my edit, that the user shouldn't be able to "accept" their own answer. However, if the answer is popularily voted to be the best answer, that's another story. – beska Aug 11 '09 at 14:21

You spend your own points to entice people to answer your question in detail.

If you are allowed 'Take Backs' then you have essentially lied to everyone who spent time to answer the question.

By putting up a bounty you are saying "I will give x rep to the best answer". The money's on the table. It's not yours to take back anymore.

  • 2
    You're saying two different things. "I will give x rep to the best answer." What if the people, by sheer number of votes, decide that your answer is the best answer? Haven't you then earned getting your original rep input back? – beska Aug 18 '09 at 18:01

You must log in to answer this question.

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