What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 127 Stack Exchange communities.

One of the main reasons I use bounties is to increase the attention and exposure to a question I have.

The problem is, I myself rarely answer bounty questions, even if I have a good contribution to the question.

This is simply because when I'm bounty hunting I want to win the answer, rather than just contribute to the answer. (I'm speaking practically, and ignoring the idealistic behaviour)

I think that upvoted answers on bounty questions should have a higher point value to encourage participation, even if the odds of actually winning the answer are lower.

Thoughts?

Update

Good comments!

  1. rep-mining? the whole rep system is based on incentives. Why have reputation if you don't care about it? Human behaviour modification is predicated on the system of 'paying attention to what you measure'.

  2. One true answer? we all know that the accepted answer is not always the right answer. Sometimes the right answer for one person is not the right answer for someone else. I'm under the impression that the purpose of SO is to create a repository of knowledge, not just a Q&A forum. If it's just Q&A, then sure, only the accepted answer counts, but if it's a knowledge base, then surely every good answer is worth being availabe?

share|improve this question
add comment

2 Answers

up vote 2 down vote accepted

I think that the bounty offers a good enough incentive. If you know that your answer isn't good enough to actually answer the question, or it isn't better than the rest of the answers there, don't post it--regardless of bounty. If all you have is a +1 on another answer, then just comment or edit the answer to include your little "aha!" After all, you were late to the game.

My opinion is that rep-mining is a behavior that should not be encouraged. Anything we can do to get people to help get at least one right answer there instead of a bunch of pieces put together to make up a whole answer should be encouraged.

share|improve this answer
1  
Hear Hear! Also, the OP is going to look at every answer and select the one that best answers their question, so you should upvote anyone who's answer influenced YOUR answer, because if you win all the rep-bonus, don't they at least deserve 10 measly points from you? Especially since it doesn't cost you anything? –  devinb Jul 10 '09 at 20:51
    
Valid perspective, but I think misses the point of SO: 1. a knowledge base, and 2. incentivized answers. –  John Weldon Jul 10 '09 at 21:02
    
Not sure why the OP didn't just comment here, but anyway: Paying attention to what you measure is human nature, but it's not (and shout not be) encouraged. The "one true answer" is a bit of a strawman on your part. I said we should help toward the goal of having at least one complete answer rather than a bunch of partial answers. Often there are alternatives, so there's no one "true" answer, and if you have one, please do post it! Example of my point: stackoverflow.com/questions/1111715/… –  Eric Jul 10 '09 at 21:08
    
I should note: I started typing then walked away, the comment about the OP commenting is invalid :) –  Eric Jul 10 '09 at 21:09
    
I agree with 99% of these comments. I DO upvote other answers that merit it. My only issue is what are we trying to incentivize with reputation? If we don't want people attempting to increase their reputation, then why have it? My purpose in suggesting this is to make the incentives work better, not worse. –  John Weldon Jul 10 '09 at 21:27
    
Reputation isn't an incentive. It's a measure of how much the site trusts you as a user--how productively active you've been. Rep is supposed to reward productive behavior, and I don't think that spamming a bounty question with partial answers falls into that category (a little extreme, granted). The bounty is there as an incentive to answer that specific question because it promotes answering obscure, difficult, or otherwise unnoticed questions, which makes the % answered stat go up, which is, therefore, productive. –  Eric Jul 10 '09 at 21:37
    
Good point. Nevertheless, if you measure it, it's still an incentive. Human behaviour. We can agree that its not ideal, but I expect better results when I try to work with normal human responses. –  John Weldon Jul 11 '09 at 20:38
add comment

Don't agree with this.

Stuff posted on bounty questions is already much more likely to be voted on in general, because

  1. there's a giant [+200] (or whatever) on the question title bringing attention to it in every list it appears on.

  2. it's on the featured tab with a very small list of questions

  3. the featured list is sorted by time, like eBay, so the most recently ending stuff is always at the top.

share|improve this answer
    
Thanks for the 'official' response :) I still think that I would take the extra effort to write a meaningful response on a bounty question if there were more incentive. –  John Weldon Jul 11 '09 at 20:35
    
I do not agree. I seldom check the featured tab, or vote questions and answer, and the +200 does not attract my attention. The question title does, regardless of the bounty. Fact is that bounty questions are more likely to be specialized, and less people have an interest into it. For example: "Any reason to overload global new and delete?". I'm neutral to this question in terms of interest. I never had to overload new and delete, nor I will in the future. Having a bounty does not attract me more towards reading it or voting it. –  Stefano Borini Aug 1 '09 at 17:10
    
Also, if a bounty is just a payment for visibility, why does a 550 bounty and a 100 bounty have exactly the same visibility and rules ? –  Stefano Borini Aug 1 '09 at 18:54
add comment

You must log in to answer this question.

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