What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 122 Stack Exchange communities.

Say you and 5 of your buddies get together and decide you're going to get to 10,000 reputation in 6 months. How difficult would that be? Here is my approximate math - 10 points each for an upvote plus 15 for the accepted answer(accepting the answer too often might start to look suspicious but bear with me it's an approximation) - you would need to ask/answer approximately 154 "fake" questions - with some variation in your voting not taken into account. That seems do-able in six months - right?

Now I originally proposed the idea of a vote history - so you could see who is working together to game the system - Now I understand the drawback of this would be that it would destroy vote anonymity(which I don't really value that much for myself - so I didn't consider how important it was to everybody).

So all that aside - I want to propose another idea about how to combat "mutual upvoters". You could have a vote density measurement within tags on each person’s profile. So assuming these 6 guys all upvoted each other and nobody else upvoted them and they did it within just one tag (or a variety of tags - it doesn't matter in this case since they would be consistent in their upvoting) - each of them would have an upvote density of 20%. If it was just 5 of them it would be 25% - three of them 50% and 2 of them 100%.(or if you did this based on points - the percentages would be slightly different - I'll let you do the math) So your goal would be to have a low density of upvotes. Most people currently on SO rarely answer questions from the same guy over and over and over again - so the average person would have a low upvote density(probably well below 5%). You could probably do the same with a downvoting density within each tag - in this, you would want to have a high density of downvotes (that way if just one cranky guy is repeatedly downvoting someone - you will be able to see that).

share|improve this question
9  
Extensive mechanisms to detect cross-voting are already in place. Maybe even something like this is - but SO won't make details public, because that would defeat the purpose. –  Pëkka Oct 31 '12 at 19:03
3  
As an aside, it's not difficult to get 10k legit rep. even in 2 months. Many have done it. –  NullUserException อ_อ Oct 31 '12 at 19:28
1  
@NullUserExceptionอ_อ Keep in mind that the kinds of people who turn to vote fraud tend to be people that can't earn rep so easily. –  Servy Oct 31 '12 at 19:31
add comment

1 Answer

You're at a rather large disadvantage when proposing countermeasures against vote fraud as you can't know which measures are already in place. I can't tell any details, but you can be sure that the scenario you mention is something the current tools can deal with (Disclaimer: I'm a moderator on Skeptics and Biology, so I've seen those tools).

I also think that it just doesn't make any sense to make such statistics public, regular users can't do anything about vote fraud, you always need moderators for that (or the automatic script). There would be no advantage to publicize such a statistic, even if it doesn't violate the privacy of votes.

share|improve this answer
1  
Which votes are fraudulent is oftentimes subjective or open to interpretation. THe moderator gods can't always catch everyone. –  Stepan1010 Oct 31 '12 at 19:09
7  
@Stepan1010 Of course we can, we wouldn't be gods otherwise. –  Yannis Oct 31 '12 at 19:12
    
@Stepan1010 If mods can't do it, regular users most certainly can't.... tools or no. –  Andrew's a Unitato Oct 31 '12 at 19:13
6  
@Stepan1010 "Which votes are fraudulent is oftentimes subjective or open to interpretation." Which is why these numbers shouldn't be posted publicly to be met with a hail of downvotes. –  Bill the Lizard Oct 31 '12 at 19:14
1  
@Stepan1010 If you feel that votes were reversed incorrectly, or not reversed when they should be for you or someone else, you can always send an email to "team@stackoverflow.com" to discuss the issue with someone. –  Servy Oct 31 '12 at 19:14
    
@Andrew Barber - Users and employers can use it to evaluate and make conclusions about the reputation score of a user. –  Stepan1010 Oct 31 '12 at 19:16
1  
@Servy - This is not about me. I am happy with my experience on the website. But there is always room for improvement. –  Stepan1010 Oct 31 '12 at 19:17
1  
@Stepan1010 - Employers don't care about your reputation score on Stack Overflow, Google Groups, Yahoo Answers, or any other forums that are out there. They care if you can do a job or not. –  jmort253 Oct 31 '12 at 19:43
    
@jmort253 -Obviously employers care if you can do a job. But it is oftentimes difficult to tell who can do a job before they are hired. That's why they spend $ on HR departments. I value the reputation score - when I am sorting through people's answers. I notice a correlation between the reputation of a answerer and how helpful and well written their answer is(just a correlation). I would assume employers use any number they can get their hands on(20 years experience with blah blah blah). But I would be interested to know how many people agree with you on this. –  Stepan1010 Oct 31 '12 at 20:24
    
@Stepan1010 Of course there's a correlation between a person's reputation score and how helpful their answers are. That's the whole point of the reputation system - you upvote to recognise users that are posting well written, useful answers. –  Anthony Grist Nov 1 '12 at 9:32
    
@Anthony Grist - I was trying to be diplomatic. I value the reputation score. It is one of the defining features of the stack network. I'm just interested how many people/employers don't - like jmort. –  Stepan1010 Nov 1 '12 at 12:06
add comment

You must log in to answer this question.

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