What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 128 Stack Exchange communities.

First, a good-faith request: please don't take this too seriously. I've had this idea, ran it by a couple of people and decided it's not completely worthless. Just forgetting about it seemed a waste. I'm sure it's far too radical to be an actual realistic possibility on StackExchange. Still, here it goes - perhaps it'll spark some other, less radical yet realistic ideas.

Motivation:

To increase reward for posting many complex answers to obscure questions, and reduce reward for posting simple answers to basic but very popular questions. Not because the latter is not useful - absolutely not. But rather, because the system is currently disproportionately biased in favour of the latter.

In other words, I'd like to award the answerer on the basis of how close his/her answer is to being the best answer available, rather than on the basis of how popular it is.

For answers:

Instead of awarding rep for every upvote or downvote of an answer, award rep according to the rank of the answer, as sorted by total votes.

The exact details can vary a lot, but one idea is that the rank awards a multiplier, while the weight of each vote depends on the total number of votes. Thus, the top answer might get a multiplier of 3, the second top answer 2, and all others 1. Moreover, a question whose total vote is +1 would get 10 base rep, while one with 3 upvotes would only get 10+9+8 = 27 rep. The maximum base rep would thus be 10+...+1 = 55, at which point any further upvotes are just "popular question" upvotes and effectively don't count.

Under this specific example (which is by no means the only way to implement the general idea of using rank), a top answer with +30 votes would get the answerer 3*55 = 165 rep - and that's it; never more. A top answer to an unpopular question with just +3 votes would still get the answerer 3*27 = 81 rep - only a factor of 2x out with a really popular top answer.

One idea for handling accepted answers is that its multiplier is the max of the rank-based multiplier and, say, 2 - so if it's top by votes already then it gains zero from being accepted, but if it's not top then it gains quite a lot, based on votes.

Not considered:

Obviously this is very sketchy. Among other things, it doesn't consider rep for questions at all, or rep loss for negative vote totals in any detal.

Finally:

Please, if you're going to reply, don't nitpick on the very specific numbers I posted as an example - the example hasn't been carefully analysed or fine-tuned. Rather, concentrate on the idea of awarding answer rep on the basis of the rank of the answer.

share|improve this question
5  
@rom, I thought it was clear. –  jjnguy Aug 9 '10 at 20:23
1  
OK, sections swapped. –  romkyns Aug 9 '10 at 21:23

4 Answers 4

I'd rather not set up another incentive to downvote other people's answers. We already have tactical downvoting, but this would be strategic downvoting.

share|improve this answer
4  
I'd upvote you for the good point about strategy, but then I'd lose my lead. –  Grace Note Aug 9 '10 at 20:52
    
Ah, good point. I know my opinions tend to be the minority opinions here, but I always felt answerers should not get a chance to downvote other answerers - the loss of a single potential expert's downvote is a drop in the ocean. –  romkyns Aug 9 '10 at 21:21

There are a few failings with this

Multipliers: Unfair exposure

More popular tags get more votes. This means more upvotes for the highest ranked answers. This means that getting the "right" answer on a C# question would be much more valuable than getting the highest rank on an obscure tag.

Multipliers: Rich get Richer

It is well known that the top-ranked answers get more upvotes after they become top-ranked. This is simply based on exposure. People don't like reading, and so they'll read the first few (sorted by votes) and maybe upvote them. So those users already enjoy a reputation bonus, your system will reward them for their good fortune. This is like offering a scholarship only to those who have won the lottery.

Multipliers: Conflict - Rank vs Accepted

Oft-times there are answers that are higher voted than the accepted answer? Which one would your system reward? Either way, it is wrong. In one case, you are rewarding a user who was not as helpful as the "accepted answer". In the other case, you are rewarding someone how has an answer that the community doesn't agree with. Both of these reward someone at the expense of another.

Multipliers: Logistics

Grace Note mentioned, reputation would be absolute gibberish if this was implemented. Your reputation would change wildly, by up to 55 points on a whim, only based on the fact that someone else got an additional upvote. Reputation would be meaningless because it is constantly in flux. Also, ties would be a problem.

Multipliers: Answers are not relative!

Answers are supposed to be considered independently. That is to say. If someone says "I need to take two strings and put them into one string", there are multiple completely valid ways to answer the question. However, because people are fickle, they won't get upvoted evenly. Normally, this would be no more than a 10-20 point difference, however your system internalizes and glorifies these differences by rewarding them unevenly for their equivalent and equally useful efforts.

Multipliers: Obscure tags often have zero votes

How does a multiplier help me if I have zero votes to begin with? Normally, this doesn't matter, because I'm losing out on only 10 points. But in this case it would be 30 points per vote.

Multiplier: Higher bonus for gaming

Normally, if I downvote someone nearby, I get a little celebrity from being higher than them (at the expense of my soul), in your case, I would get extra vote plus a multiplier. This increases the incentive to be higher in the ranks, rather than increasing the incentive to be better.

Diminishing Vote Value = Diminishing Votes

So an answer has +6, but it is 100% correct and well written. But still, I don't bother upvoting it, because it would only yield +5 for the user. My vote is less valuable there than it would be elsewhere. Whereas if I vote on a correct, but poorly written answer that has zero votes, I am giving that user +10 points, even though their answer wasn't NEARLY as good.

Diminishing Vote Value = Diminishing Quality

If I toss off a quick response, maybe just a link, I'll get an upvote or two, and they'll be worth 10 or 19 points. If I write a good response, it'll be worth upwards of 34. If I throw my soul into a response, it'll be worth only 21 points more, even if I get 20 more upvotes, meaning that there are a lot of people who like the response. So where's the incentive to be stellar? I'll give you a hint, there isn't one.

Dimishing/Multiplier: Downvotes

Downvotes would end up becoming very complicated in your system, because with a single downvote I could bump someone from having the highest voted, to second highest voted (or tied) and suddenly they would lose a chunk of reputation. This would make me less likely to downvote even though my objections still remain. The content of the post hasn't changed, but the voting schema will have changed my voting pattern.

Anything that encourages strategic voting is bad.

Most importantly.

This system encourages being higher in rank as being more important than being higher in quality

share|improve this answer
    
Hmm, Exchange doesn't work for discussions because there's no way for me to reply to your points in detail :) First: any rating system will have this many drawbacks. One could surely write a whole thesis with Meta material on all the drawbacks of the current system. Second, re "most importantly": the current system doesn't encourage quality in any direct way either. Lastly, I disagree with many of your other conclusions, but alas can't really elaborate here :/ –  romkyns Aug 9 '10 at 21:18
    
@romkyns It's not the most elegant, but you can always do something like "re:Rich get Richer" at the start of an individual comment for each point. –  Grace Note Aug 9 '10 at 21:24
    
@Romykns, Perhaps this will shorten the work you have to do. We should have an assumption at the baseline that I believe many of these failings exist in the current system. However, I feel that your system enhances (makes worse) the problem. At no point would I claim that the current system is perfect. Hopefully, that should cover your general points (which usually take up a lot of space) and you can focus on specifics. Although, I should note, I'll be leaving soon, so I might have to pick it up tomorrow. –  devinb Aug 9 '10 at 21:29
    
Sorry, I decided not to. Even if we end up convincing each other one way or another (which I suspect we won't), that end result isn't worth the time it would take. Thanks for your input though. –  romkyns Aug 10 '10 at 9:20
    
@romkyns, my pleasure. –  devinb Aug 10 '10 at 11:38

This doesn't take into account that answers are rarely going to maintain a consistent ranking. It will become a nightmare to try and track what is changing on your reputation. It's also less under your control because now your reputation is based on what people think of everything else, not just the content of your post.

Suppose a question gets a pretty good answer with 3 votes to start. It's all good until two days later, someone posts a much better answer that gets 6 votes. Disregarding that ties aren't considered yet, the shift from when it's 3-2 to 3-4 yields a significant change. Suddenly, a user jumps from 38 reputation gain to 102 in 2 votes. Meanwhile, the first answer drops from 81 to 54 through no fault of that answerer.

share|improve this answer
    
Hey, I borrowed your answer to incorporate into mine, I hope you don't mind. –  devinb Aug 9 '10 at 21:08
    
It's not the answerer's fault, of course, but posting a better answer is under their control. –  romkyns Aug 11 '10 at 11:06

I'm surprised no one mentioned a very important "side effect" of your proposal: it makes stack overflow harder to learn.

Usability and a low entry barrier for new users (who care about the "reputation game") is very important.

share|improve this answer
    
If you mean learning the details of the scoring, then it's already arcane and anything but easy to learn. If you mean something else then please elaborate, since I don't see how anything else changes. –  romkyns Aug 11 '10 at 11:07
    
@rom: yes, that's what I meant. –  Andreas Bonini Aug 11 '10 at 20:03

You must log in to answer this question.

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