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.

If I were to post an answer on a (eg) question tagged as 'cryptography' and receive two upvotes, one from joe-bloggs and one from a recognised expert in the field, it would be nice if that second vote carried some extra weight. In some ways this will help with the 'fastest gun in the west' problem, in that the earlier answers can hopefully be overweighted by the opinions of experts voting on later answers.

I'm not quite sure of an easy way to judge 'experts', but I think this weighting system would be quite valuable if it could be implemented somehow.

One option would be to allow users to nominate themselves as experts in up to five particular tags which they can only change infrequently (to stop gaming). Any votes they place (up or down) against questions in one of those tags could then carry a weight of 1.5 (rather than just 1 as it is now). These 'expert' tags should appear in the user profile page for maximum transparency.

The half vote shouldnt be displayed against the answer score, but can instead be tagged in the backend that the answer has received an additional half vote. When it receives a second half vote then the score will jump by 2. This means that only when two experts have agreed does the answer move by an extra point.

Similar Question: Indicate How Trustworthy the UpVotes Are

Note however that this is not an exact duplicate. The previous question discussed (extensively and exclusively) using Rep as a means of weighting votes. This system does not rely on rep.

share|improve this question
2  
@random. I disagree that this is an exact duplicate. I'm not suggesting that rep is used to weight answers. –  PaulG May 12 '10 at 11:18
    
Aren't you asking about weighting votes? –  random May 12 '10 at 11:32
2  
Yes, but not by rep as the earlier question suggested. –  PaulG May 12 '10 at 12:15
2  
@PaulG - having 1000 upvotes on a tag may be because you are good on this topic, but that doesn't mean that you are more competent to judge what is valid or not than someone who is an actual expert on the topic, who simply doesn't spend as much time on the site. Votes should really remain equal. Indexing on reputation or "targeted reputation" is ultimately the same. –  Gnoupi May 12 '10 at 12:56
    
ok, i took out the bit where I mentioned a possibility of using rep (which I specifically said was a bad idea) - it seems to have confused people. –  PaulG May 12 '10 at 13:12
    
illustration on Meta, I'm declaring myself an expert in "feature-request", so my votes will have more values on them? In general, if someone does that on smaller tags, it could lead to people having suddenly more power on some questions, just before they said they were experts. I understand the idea behind this request (and the other). Have votes from people who know the subject value more than "profanes" ones. But there are other criterias which can make me recognize a good answer even if I can't judge the topic: the way it is written. –  Gnoupi May 12 '10 at 13:20
    
A good answer is not only a right answer, it's also a well presented, formatted answer. And this, even if I don't know fully the language, I can judge. –  Gnoupi May 12 '10 at 13:21
    
we'll have to disagree on that. I think a badly presented right answer is better than a well presented wrong answer. Using this system the experts wouldnt rely solely on how well formatted it was. –  PaulG May 12 '10 at 13:29
    
@random. Is the final verdict that this non-dupe is staying closed then? Shame, I'd have liked the chance for some discussion better than this. –  PaulG May 12 '10 at 17:25
1  
You state that you removed the part about rep, but this post is still surrounding a system that is primarily judged by reputation. You suggest that "experts" be the ones who get weighted votes, but the only example you have for determining who is an expert is "users above a certain rep threshold". As such, your request still duplicates the concept of "People with higher reputation should be given greater weight in their upvotes", and devinb's answer still applies: "The voting system is the one part that is even and fair to everyone. Let's keep it that way.". –  Grace Note May 13 '10 at 12:22
1  
@random. oh hai, thanks for finally replying. Might be nice if you can make sure its a dupe before closing and then disappearing for a week next time :) meta.stackexchange.com/questions/49744/… –  PaulG May 19 '10 at 16:18
1  
@random I believe you are way too much overzealous when closing for duplicates.. I believe I already told you but "exact duplicate" means exact not similar :/ –  Andreas Bonini May 21 '10 at 11:07
1  
When the question covers the same topic, but with different words, then yes, that's a dupe. Don't be fooled by the thesaurus. And there is not enough closing as duplicates. @kop –  random May 21 '10 at 11:47
1  
@random. So discussion about weighting votes is not permitted because it has been brought up once (now twice) before?! Are you trying to invoke Godwins law?! See also: meta.stackexchange.com/questions/50507/… –  PaulG May 21 '10 at 13:47
1  
@PaulG: Most people who are ignorant of a particular subject don't vote on it at all (I don't even open questions on Java, Flash, Ruby On Rails, etc.) All that this feature hypothetically does is allow anyone, anywhere, to declare their votes worth more than anyone else's, regardless of their actual level of expertise. –  Aarobot May 22 '10 at 14:59
show 10 more comments

3 Answers

up vote 4 down vote accepted

I've already made a point about the philosophical problem with this feature in the comments, but I wanted to point out a few logistical problems as well.

Questions have more than one tag, and tags can change. Users, according to this feature, can also have multiple "expert" tags. So:

  • What is a vote worth when a question matches two or more of a user's "expert" tags?
  • What happens to "expert" votes if the "expert" tags are removed from the question?
  • Conversely, what happens if the "expert" tags are added later, should the votes be recalculated on every retag?
  • What happens to all of the past questions/answers if a user decides he's really not an expert (changes his own tags)?
  • Do half-votes affect answer sorting? Should they?
  • What does this look like on a rep audit? Or the rep chart? How are users supposed to reconcile what appears to be 12 votes with only 100 rep? Get ready for the bug reports!
  • What about downvotes? Normally a system would either round up or down. To get the same behaviour for downvotes, it would have to round down for upvotes and up for downvotes. Confusing, to say the least.

This is really a logistical and impelmentation nightmare, and it doesn't add any trustworthiness whatsoever (do you trust someone who says "trust me, I'm an expert" just because he says so?). I predict and firmly advocate status-declined for this one.

share|improve this answer
    
All very good points, most of which I hadn't considered. Thanks. –  PaulG May 22 '10 at 18:50
add comment

I think that the biggest problem with a system like this is determining who the experts are.

Automated

  1. Reputation - this isn't reliable, as has been discussed ad nauseum in other places, but it boils down to the fact that having the free time to answer questions on SO does not necessarily equate to expertise or good judgment.
  2. Votes-to-answer - easily gamed by waiting for questions that look like they'll be popular. Some of the highest-voted answers on SO are some of the least illuminating, helpful as they may be.

Manual

  1. Nomination/voting - honestly, who has time for this? It takes long enough to elect moderators on SO, and they don't have a direct impact on the "flow" of the site (in their capacity as moderator, they're police, not the drivers).
  2. Self-nomination - this seems like the least desirable method. Yes, it addresses the time constraint and isn't rep based, but I don't think that what something is not based on can automatically lend it any validity.

Putting aside the question of "who", say, for the moment, that I'm an expert. Right now, I'll frequently upvote answers to questions that I've also answered, especially when one is similar or complimentary to my own. If I'm an expert and the other person is not, that now means I can either upvote them and have their answer now automatically worth more than mine (the expert to begin with!) assuming that all other votes are equal, or I can not upvote a perfectly valid and acceptable answer. I don't consider either of these circumstances to be A Good Thing.

share|improve this answer
    
Thanks, good reply. Regarding upvoting and answering in the same question though - you already face this dilemma surely? I hear what you're saying though, and if this system were in place your vote would only award an extra (bonus) point if another expert had already upvoted it (or later upvoted it). Its a fair point though. –  PaulG May 21 '10 at 23:06
    
@Paul: No, I don't face this dilemma now, as all of the votes are worth the same. If I were an expert, that means that I have to hope that another expert will come along to upvote me, rather than just anyone with more than 10 reputation. –  Adam Robinson May 22 '10 at 4:20
add comment

I reopened your question per your request, but this proposal is even more toxic than the one it's a semi-dupe of.

One option would be to allow users to nominate themselves as experts in up to five particular tags which they can only change infrequently (to stop gaming). Any votes they place (up or down) against questions in one of those tags could then carry a weight of 1.5 (rather than just 1 as it is now).

Excellent, because I have determined that I am an expert in everything. What could possibly go wrong here...

share|improve this answer
1  
I'd like to draw your attention to the bit where it says 'in up to five particular tags'. Oh, and thanks for re-opening :) –  PaulG May 21 '10 at 9:35
    
I suspect this answer could have been 'Bad idea because the sky is green' and it would still be getting upvotes. I'm getting that people dont like the idea, I'm just not getting why. –  PaulG May 21 '10 at 13:24
add comment

You must log in to answer this question.

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