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.

"Voted on 600 questions and has better than 3:1 answer/question vote ratio" is the current text.

"Voted on 600 questions and 25% or more of total votes are on questions" is my proposed text. (Or should it read "or less"? I'm still not completely sure what is intended.)

"Better" in the original doesn't indicate which way is preferred (and it's not clear to me one way or the other is better, merely that a certain threshold is the requirement for this badge). This is even slightly shorter, in both characters and as displayed for me in a proportional font, which should help for a badge with one of the longest current descriptions.

share|improve this question
18  
I can't believe I have voted over 6600 times and missed this badge! And it isn't easy to see how far I have to go (or whether I should be voting on questions vs answers). –  Marc Gravell Jan 2 '10 at 16:41
    
There's another possible ambiguity with comment votes. (I've assumed so far they are counted entirely separately.) –  Gnome Jan 2 '10 at 17:57
5  
+1 - I'll admit I thought I was slightly confused as to the criteria for this badge until I read the possibilities here... now I have absolutely no idea what you have to do to get this badge - lol... –  John Rasch Jan 2 '10 at 18:05
2  
What about downvotes? I probably downvote questions more often than I downvote answers, since I think it's more important to bury the crap questions than bad answers (which are usually edited or deleted after their shortcomings are pointed out, anyway). –  Ether Jan 2 '10 at 18:43
    
Re downvotes: it appears they count in the statistics just the same: e.g. look at the Community user (stackoverflow.com/users/-1/community), who doesn't even have 600 upvotes in total but received the badge. –  Ether Jan 2 '10 at 18:57
    
downvotes or upvotes are valid. Note that RichB has this badge on Meta, and check out his voting stats.. –  Jeff Atwood Jan 3 '10 at 0:58
12  
The wording is now (2010-01-03 08:37 -08:00) 'Voted on 600 questions and 25% or more of total votes are on questions'. This is a gold badge. What good behaviour is it rewarding? It seems to be rewarding people who vote on questions and penalizes those who vote on answers. Well, I suspect I'm unlikely to get the badge; I vote for answers, not questions - I was under the illusion that they were much more important. –  Jonathan Leffler Jan 3 '10 at 16:39
    
Jonathan: there is another meta question discussing that, this feature-request was only to get the wording clarified without changing the meaning of the badge. –  Gnome Jan 3 '10 at 17:57
add comment

4 Answers

up vote 9 down vote accepted

I see several interpretations here. (I've no idea which is correct, but I'll share my guess after listing them.)

Voted on 600 questions and has better than 3:1 answer/question vote ratio

  1. Voted 600 times, 25+% of votes were for questions.
    (Assumes "better than" a 3:1 ratio means <=3.)

  2. Voted 600 times, <=25% of votes were for questions.
    (Assumes "better than" a 3:1 ratio is 3+.)

  3. Voted on 600+ questions, and voted on at least 3x {number of questions voted on} answers. For example, if you voted on exactly 600 questions, you won't get this badge until you've voted on at least 1800 answers (or more if you keep voting on questions).
    (Assumes "better than" a 3:1 ratio is 3+.)
    discounted; see update

  4. Voted on 600+ questions, and voted on no more than 3x {number of questions voted on} answers.
    (Assumes "better than" a 3:1 ratio means <=3.)

Any other interpretations?

I'm guessing #3 is what is intended. It is a gold badge, after all. But what exactly is meant by "better than" is still ambiguous.


Update: In light of this question from this user who apparently got the Electorate badge with a grand total of 1276 votes (current count only)... interpretation #3 is hereby incorrect.

share|improve this answer
    
Maybe you should stop reverse engineering and wait till the guy who implemented it jumps in ;) What was his name again... –  Ladybug Killer Jan 2 '10 at 15:45
1  
nah, speculating is fun. i'm sure he'll jump in when he's ready; this was just a thought experiment of the "what does it mean" variety. –  quack quixote Jan 2 '10 at 17:56
    
Your example in the update is the best proof that the badge description is outright wrong. Remember, a 3:1 ratio is not the same as a fractional 3/1: it's actually equivalent to fractional 4/1. (However, the user doesn't have enough total votes for this incorrect calculation either.) –  Ether Jan 2 '10 at 19:04
1  
@Æther: well, 3:1 only converts to 4 if you're taking a total number of parts and discarding any interesting info about their relationship. 3:1 is equivalent to fractional 3/4 (three parts of a total of four) for some conversions of ratio to number; for others (think unit conversion, or multiplying recipes), 3/1 is a perfectly sane equivalence. but i don't think the badge description is wrong, just extremely ambiguous to the point of utterly misleading ... –  quack quixote Jan 2 '10 at 20:28
6  
#4 is correct, and I updated the description in the database. We want people to vote for a lot more questions, so if your ratio of answer votes is too high (eg you have voted for FAR more answers than questions) you can't achieve the Electorate badge. –  Jeff Atwood Jan 3 '10 at 0:56
9  
@Jeff - #4 seems counter-productive to me. Aren't people just going to start voting on questions exclusively, just to get their ratios down? I think the 600 question votes would be enough incentive to get people to vote on questions, without having to set up a ratio which may stop them from voting on answers entirely. I much prefer interpretation #3, for what it's worth. It requires a total of 2400 votes, which seems gold worthy, but doesn't have the negative behavioural modifications. –  womp Jan 3 '10 at 4:58
add comment

Hmmm... not sure how I'm ever going to get this badge (or how far away from it I am). I tend to vote for any question I find interesting (and well-asked), and any interesting answers. There are more answers than questions, so I think my ratio is shot to heck. :-(

Then again, who needs to be in the electorate if you have the (hopefully benevolent) [♦ dictator] badge ;-p </joke>

share|improve this answer
    
I was thinking that people with whacked ratios, such as Jon Skeet, may not, due to max votes/day, even get this badge for another year or more. (Obviously a dedicated effort could get it in less time, but I'd hope this badge doesn't become the sole reason to vote.) –  Gnome Jan 2 '10 at 17:05
    
I strongly agree on the well-asked criteria. I'm much less likely to vote up if it's not well-asked, yet lacking that is rarely a reason to vote down for me. Of course, 90% of questions are crud, because 90% of everything is... –  Gnome Jan 2 '10 at 17:09
    
Without more detailed statistics on your profile page (whether visible to just you, or to everyone), it's not possible. Sadly, repeated requests for more such statistics have been turned down. It's unfortunate the data dumps (and stackql.net) are not more current, or this information would be reasonably easily calculable to those who are interested. –  Ether Jan 2 '10 at 19:05
add comment

I'm all for clarifying the wording.

I'll also take this moment to share a few thoughts on the badge itself. Taken literally:

voted on at least 600 questions

would mean 600 votes for questions, not counting votes on answers.

better than 3:1 answer/question ratio

means more than 3x as many votes on answers as on questions. I'm not sure that's what is intended, but that's what the text says. You might dispute the word "better", but with a literal reading it could only mean a number greater than 3 for the ratio.

That means you need at least 2401 votes to get the badge: 600 on questions and 1801 on answers. Now I know this is wrong based on the update to quack's answer, but it is accurate based on a strict literal interpretation of the text. The actual definition probably just reverses the meaning a "better".

So taking this interpretation, is this a useful badge? As the stated purpose of badges is to encourage specific positive behaviors we need to consider what behaviors this badge encourages. In this case, it seems clear that the badge is intended to increase votes on questions. In that respect, I think might do okay, but I don't know that it will have a lasting impact unless it's effective at creating good voting habits as well.

My main complaint is that this is a one-time gold-level badge. I'll grant it's "bigger" than a silver, but it doesn't quite live up to what I think a gold should be. If you really want to encourage voting on questions long term, I think you're going to need some kind of ongoing incentive. A recurring silver badge for every 250 or so question votes might just about do it.

Also, one additional side effect of this badge is that it will shift some reputation from question answerers to question askers. I think a lot of people consider this to be a bad thing.

share|improve this answer
    
Somewhere else (I forget where) the comparison was drawn that "better" can mean less, as in "better than 75% mortality rate". –  Gnome Jan 2 '10 at 22:46
    
I just saw quack's update, so a lot of this changed. –  Joel Coehoorn Jan 2 '10 at 23:07
1  
@Roger: It was here: meta.stackexchange.com/questions/34340/… :) –  Ladybug Killer Jan 2 '10 at 23:15
    
consider carefully. 600 votes is a LOT. If you used all your votes, every day, on nothing but questions, it would take you 20 days. And that's assuming you never voted on a single answer -- every answer vote potentially harms your odds of getting Electorate. –  Jeff Atwood Jan 3 '10 at 4:19
    
@Jeff: "every answer vote potentially harms your odds of getting Electorate." - That's a funny statement coming from someone who just said that Electorate doesn't discourage voting on answers. –  mmyers Jan 4 '10 at 15:54
    
I know at least one user who is only voting on questions now until they get the badge. (It's not me... I already have it. I mean one other user.) –  womp Jan 6 '10 at 4:07
add comment

How can this badge even be calculated, given that voting histories are not tracked? I was under the impression that there was no record of what I voted for: votes are tracked by their targets (questions and answers have votes), but user data is only saved for the number of votes in total, other than downvotes which are visible in my recent activity (as well as negatively affecting my reputation).

So, to rephrase: how is this tracked in the database? Just how much information about my voting history is saved?

share|improve this answer
    
interestingly, the item after this one on the front page right now touches on this subject: meta.stackexchange.com/questions/34323/… ... check the answer comments where Jeff talks about being unwilling to expose such history... given that, i'd say your impression of no record saved is just plain incorrect. –  quack quixote Jan 2 '10 at 18:47
3  
You can view old questions and your votes are displayed to you; they must be saved somewhere. Plus preventing duplicate votes requires tracking some past history. (Preventing dupes doesn't require tracking up or down, and that's seemingly not required for this badge either, but it does track that, because it displays it to you.) –  Gnome Jan 2 '10 at 18:55
1  
@Roger: you're right: I can see what I previously voted on, so that information must be saved. I conflated "this information is not in the public data dump" with "this information is not saved at all" -- two very different things. –  Ether Jan 2 '10 at 18:59
3  
Voting history is not exported in the monthly data dump. That would be bad. But it is saved in the actual database used by the site. –  Joel Coehoorn Jan 2 '10 at 21:46
add comment

You must log in to answer this question.