13

While looking over the idea on increasing the weight of down votes, I started thinking about the distribution for reputation scores and wonder if the accepted answer bonus of +15 is still a good number.

REP SCORES:

  • +10 per upvote
  • -5 per downvote (proposed/planned?)
  • -1 per cast downvote
  • +2 per acceptance of answer
  • +15 per accepted post

Since a down votes may be increased 150% from 2 to 5 points, the accepted answer will not be worth as much compared to a down vote before. Saying that you will get +15 for an accepted answer and potentially lose that by a couple of down votes doesn't seem to reward the user who answered the question enough.

Recently I've seen a few jQuery and CSS questions where folks were coming in and literally swiping the previously posted answers (almost verbatim at times) and then place a few down votes on the other posts to put themselves towards the top of the list. In these cases, even if you get awarded the accepted answer bonus, you still got dinged pretty good from the other users tactics.

Should the accepted answer bonus be increased from +15 to +20, +25, maybe even +30? This would seem to reward the answerer a bit more and help recover some of the points lost for the odd down votes. I also think that this could potentially improve the quality of answers as folks really have more to shoot for and also increase the number of folks trying to get "the answer."

My only concern in this is that by increasing the rep score for answers, the behavior of answer swiping and tactical down voting would increase.

6 Answers 6

7

Sounds like you are trying to propose a solution to a problem that would further cause the problem to flourish.

But with that said, I always thought that an accepted check mark only being worth 1.5 upvotes was a little strange and that the bonus should have been much bigger. The only really good upside to a check mark is that it will allow you to go over rep limit (but only if you are already there, not prior to hitting the rep limit).

So I wouldn't mind seeing the Check mark be worth 30-50 points myself. I think that would be a great encouragement for people to focus on actually trying to win the question rather than just trying to post quick answers hoping for a couple quick upvotes (since +2 is already worth more than a check mark alone).

1
  • lol, whats up with you calling me out! At least I noted that very concern in my thread at the bottom in italics... :) Hopefully that type of issue can be addressed at a later time anyways.
    – RSolberg
    Jul 21, 2009 at 16:30
7

I think the accepted answer should receive higher rep, but not for the reasons being discussed in this thread.

I think accepted answers should receive higher rep (I would suggest +50) to encourage people to answer specific, technical questions.

I've used SO for over a year now and I've been discouraged by the following observation: There is very little incentive to try to work through a specific question to provide an answer that will probably only benefit the asker. Maybe that is by design. However, I have put significant effort into certain questions and received plenty of gratitude from the asker, only to receive (at most) 25 points for my trouble. That's assuming the asker both accepts and upvotes my answer.

I've slowly learned that answering those types of questions is generally not worth my time. On the bright side, I now understand why some of my very specific, technical questions have also gone unanswered.

I know I'm not the only one who feels this way.

2
  • While I agree with your post here as I've experienced something similar, I don't see the connection with the link you posted. It seems unrelated as it's a discussion about SQL; is there something in that thread I should be looking at?
    – greatwolf
    Jan 16, 2011 at 1:40
  • The link should take you to a comment by @KM (I included an anchor tag). The full text of KM's comment is: "all the answers should be wiki too. it is just insane that questions and answers like these get so many up votes. I thought this was a technical forum? you can solve someone's problem by providing some difficult to write code and get one or two up votes, yet answer a question like this and get loads of up votes. that's really lame if you ask me."
    – mwolfe02
    Jan 16, 2011 at 15:46
1

I don't think that increasing the value of accepted answers will make people downvote to put their answer on top any more.

There is already huge incentive to be on the top of the list of answers.

However, I don't necessarily see the need to increase this amount. We don't need to introduce more rep into the system.

2
  • "don't think that increasing the value of accepted answers will make people downvote to put their answer on top any more" true, that wasn't the point... It was to try and reward the folks who end up with the accepted answer a little more to give them back some of what may have been taken away by folks gaming the system in addition to increasing the overall quality of answers like TheTXI pointed out.
    – RSolberg
    Jul 21, 2009 at 16:32
  • Well, I was just trying to point out that your reason for not increasing the value, might not be valid.
    – jjnguy
    Jul 21, 2009 at 16:46
1

Since a down votes may be increased 150% from 2 to 5 points, the accepted answer will not be worth as much compared to a down vote before.

Assuming legitimate down votes, I think that quote from you is a perfect reason to not increase the weight of an accepted answer.

I have seen quite a lot of questions where I wondered whether the guy asking the question is really the best person to judge which answer is the best. That's my top reason against raising the rep gain for an accepted answer.

You should probably raise the real question and ask what could be done about illegitimate down votes.

4
  • Yes, I pointed out the concern, but if down votes will hurt you by 150% more than they used to, then you sure as heck better have more upside to your post in the form of the accepted answer bonus.
    – RSolberg
    Jul 21, 2009 at 16:37
  • 1
    In the most traditional sense of programmer Q&A, the correct answer would be the one that actually worked for the OP, meaning the OP would be a suitable judge. It's only when you start getting into philosophical and open-ended questions that the OP might not be the most useful judge.
    – TheTXI
    Jul 21, 2009 at 16:41
  • Only if you assume illegitimate/strategic/whatever down votes. I for one, rarely see stuff like that.
    – innaM
    Jul 21, 2009 at 16:42
  • Even for clear cut questions there maybe answers that work and answers that will work even better. There are even answers that work, but raise new issues or hurt you code in some ways.
    – innaM
    Jul 21, 2009 at 16:45
1

Sometimes one or more posters do not understand the original question but still get upvotes. The asker is in a better position to judge this than a casual reader. So yes I think the accepted answer should carry more points.

0

If you outweigh the increased downvote with a higher rep for accepted answers, then where is the benefit of increased downvotes?

1
  • In theory, the accepted answer probably shouldn't be getting downvoted. As the OP, I'll award the answer to the first person who got the answer to me rather than the person with the most votes. I know others do this as well.
    – RSolberg
    Jul 21, 2009 at 16:56

You must log in to answer this question.

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