What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

How about introducing a badge that rewards downvoting, but only of the sensible kind?

A measure for sensible downvoting would be a high percentage certain number of downvotes on content that has subsequently been

  • Deleted
  • Closed (except migrations and duplicates)

Bronze, Silver, and Gold badges would vary in numbers, say

Bronze: cast 50 downvotes on content that was subsequently closed or deleted

This would encourage downvoting (which is great and needed), but of the measurably sensible kind only.

What good actual numbers would be for Bronze, Silver, and Gold would be a separate discussion. Feel free to suggest numbers in your answers, as well as name suggestions (the only I managed to come up with was Whistleblowerand that isn't really accurate. I'm more imagining the old lady calling the council over some abandoned litter in the streets, or broken windows... but in a good way of course.

share|improve this question
19  
Isn't it that downvoted questions are not necessarily candidates for closing/deletion? –  Time Traveling Bobby Jun 5 '12 at 9:30
13  
Also (3) Downvoted to oblivion. There are answers which deserve the downvote but not the deletion. Deletion doesn't happen that much; this may lead to people selectively downvoting on deletion candidates only. –  Manishearth Jun 5 '12 at 9:30
    
@Garden not necessarily, no. Hence the required percentage of "accurate" votes ("accurate" under that very, very narrow definition) shouldn't be too high even for the gold badge. –  Pëkka Jun 5 '12 at 9:32
2  
@Tim yeah, I thought about that but I was worried that could make the badge mis-useable: a horde of people could downvote stuff they don't like and help each other earn the badge that way. Different from votes, deletion / closing decisions can be independently reviewed and rolled back –  Pëkka Jun 5 '12 at 9:33
1  
as someone on SO closing in on a 1:1 up:down ratio, I like this idea. Concerned in general about side effects, though. –  Andrew Barber Jun 5 '12 at 11:44
2  
Ah, I see now that you made an exception for self-deleted posts. But then, what's the point? Answers that have to be deleted because they're not an answer don't need to be voted down. And vice versa. –  slhck Jun 5 '12 at 12:14
    
@slhck true, some legitimate downvotes wouldn't count toward the badge because of self-deletions. But there's still plenty of not-user-deleted content around that's closed. If there weren't, there would be no need for mod and 10k+ deletions.... the exception wouldn't apply if a user deletes a question after its getting closed. –  Pëkka Jun 5 '12 at 12:19
    
These badges would be so easy to get on FSO :P –  Lix Jun 5 '12 at 12:27
    
I like the name Whistleblower. I might also propose Gatekeeper or Crow's Nest –  gobernador Jun 5 '12 at 13:55
    
good idea IQ badges :D –  user187464 Jun 6 '12 at 15:33
1  
Not sure we need a badge, but one thing I want to comment on: "Except deletions by user." Why the exception? We encourage users to self-delete. There's a badge for it when the score is -3. If the user self-deletes, seems like an accurate downvote to me. –  Anthony Pegram Jun 7 '12 at 21:25
    
@Anthony I was thinking about the odd case where a downvote makes a user delete their contribution, but you're right, there is no reason not to take self-deletion into consideration really. –  Pëkka Jun 7 '12 at 21:30
    
@Arjan true, I didn't think of that side effect :) –  Pëkka Jun 8 '12 at 11:07
    
closed votes from people with 3000+ points not from those with 250+ points? There is not much distinction: stackoverflow.com/faq#reputation. Closed and Downvotes should probably be separate, to be eligible for this badge I think downvotes starting at 125 points is too small. –  Jeremy Thompson Jun 11 '12 at 4:07
3  
with down-votes on 4,587 deleted posts (as of 2 months ago) I feel like Pale Horse Rider –  gnat Mar 13 '13 at 8:33

10 Answers 10

up vote 37 down vote accepted
+500

I disagree. Not all that should be downvoted should be closed, and vice versa.

  • A question can be very good, well put, well thought-out, but is off-topic, or not constructive.
  • A question can be valid, but badly formatted, or research wasn't applied to it ("RTFM questions which aren't duplicates"). There's no reason to close, only downvote.

Also

  • You're encouraging gang-voting, people see a close (4) question and rush to downvote it before it gets closed.
share|improve this answer
13  
+1 This badge will almost be guaranteed to drive away all newbies that don't get it right their very first time. I'm scared to even think about such consequences of every single slightly bad question getting down to -40 or lower. –  Mysticial Mar 12 '13 at 17:35
1  
This is unfortunately a good argument. –  Pëkka Mar 12 '13 at 17:36
3  
@Pekka웃 To give an idea how how strong the badge incentive is: I got ~90 upvotes in a 48-hour period during Winter bash because one of the hats were given for upvoting. And we've seen what's happened with the review badges. –  Mysticial Mar 12 '13 at 17:43
    
"A question can be very good, well put, well thought-out, but is off-topic, or not constructive [or a duplicate]." Exactly. No reason to downvote those IMO. –  ɥʇǝS Mar 18 '13 at 3:08
    
@Seth I agree. The downvote tooltip says, "This question does not show any research effort; it is unclear or not useful." Duplicates might show lack of research effort, but a very well written question shouldn't be downvoted as unclear. I suppose that, in some sense, any "off-topic" question is "not useful", but I think that's a stretch. Questions can be (generally) useful but off-topic for StackOverflow; they just should have been asked somewhere else. –  Joshua Taylor Mar 10 at 22:40

I like it, and if it's applied retrospectively I'd probably get the Gold variant on every site I participate, I'm particularly proud of my deletion instincts and I do feel that down voters are excluded from badges. We need some badges to award our garbage collectors, people!

That said I wouldn't feel right earning the badge on Programmers, where I'm a moderator, as all I'd have to do would be down vote every post I delete, just before I delete it. And I don't know how complicated this will get if we add an extra constraint to exclude down votes on posts you've later deleted.

share|improve this answer
1  
I totally agree. Its weird that there's 3 badges for Flagging but only 1 (critic) for downvotes, close votes or delete votes. –  Some Helpful Commenter Jun 7 '12 at 21:29

I like this proposal, seems useful to promote the usage of down-voting, but some considerations have to be made since when down-voting we are stating that:

This question does not show any research effort; it is unclear or not useful (click again to undo)

So, the down-vote alone will start being used even if the question demonstrates efforts from the user to achieve is goal.

It is necessary to consider that questions may get closed by being exact duplicates or off topic, that doesn't necessarily mean that the user haven't sweated to try to solve the issue:

exact duplicate

This question covers exactly the same content as earlier questions on this topic; its answers may be merged with another identical question.

off topic

Questions on Stack Overflow are expected to relate to programming or software development within the scope defined in the FAQ. Consider editing the question or leaving comments for improvement if you believe the question can be reworded to fit within the scope. Read more about closed questions here.

I like this solution if practicably applied to situations where the down-voter gets rewards if the question gets closed under this grounds:

not constructive

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or specific expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, see the FAQ for guidance.

not a real question

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, see the FAQ.

Questions closed by `too localized` option

This is grounds for a deeper discussion, but my first choice would be to exclude the awards on this topic.


Deleted questions should reward the down-voters since if the question got deleted, the community agrees that it was a really a bad question!

See the FAQ, where it reads:

Questions that are extremely off topic, or of very low quality, may be removed at the discretion of the community and moderators.

share|improve this answer
    
Very good points. I guess downvoting on duplicates should indeed not be encouraged. I have mixed feelings over off-topic questions though. But that can be discussed. –  Pëkka Jun 5 '12 at 9:59
    
Zuul simplest solution would be to only count down votes to answers. Those hurt a bit, and thus are less likely to be abused. /cc: @Pekka –  Yannis Jun 5 '12 at 10:03
    
@Yannis that would be a solution indeed, but there's so many bad questions that need downvoting.... –  Pëkka Jun 5 '12 at 10:05
    
@Pekka I know. But if it helps get us down voting badges, I'm all for it... –  Yannis Jun 5 '12 at 10:06
    
@Yannis indeed, what Ziv says below might be a strong argument for what you suggest... –  Pëkka Jun 8 '12 at 15:33

I don't like the idea. I downvote questions because they're bad (and it's the only way to fix them), not because I get a badge. Gamification is good and all, but my reasons for downvoting is because something deserves it, not because I'd get a badge. Voting really needs to be a civic thing more than anything else.

share|improve this answer
7  
And creating a situation where we downvote for badges we end up with lots of people downvoting for the wrong reasons, outside of what the community consensus would be on the matter. –  Journeyman Geek Jun 5 '12 at 12:20
2  
The badge wouldn't encourage arbitrary downvoting - it would encourage looking closely before you cast a downvote. Is it bad content that is likely to be closed or deleted? I don't see how that is a bad thing or egregiously abusable. (for the record, I would probably not get the badge because I downvote plenty of stuff that I find terrible, like stuff you can Google in ten seconds, but that doesn't always get closed.) –  Pëkka Jun 5 '12 at 12:22
8  
Honestly, though, this logic works to question why we have badges at all. Why would I upvote 100 competing answers if it didn't give me the Sportsmanship badge? Think of it less as motiviation and more of commendation. You have been a good and productive user, and so we recognize you. –  gobernador Jun 5 '12 at 13:49

I really don't like this "accuracy percent" idea; you're basically discouraging people from downvoting on bad but not closable posts; there are lots of ways your post can be bad but not outright closable.

It's also often the case that a good answer salvages a poor question, or a poor question is kept around because it's popular. It doesn't make sense that I should have to save up my downvotes only for stuff I know is going to be closed, or be locked out of a badge forever because 50% of my 400 downvotes aren't on closed posts.

Alternately I can see this being useful if it's instead downvote X number, not percent, of posts that go on to be closed. There you keep the idea of "accurate downvotes" without the problems % based badges have (hello Unsung Hero).

share|improve this answer
3  
True — there are so many posts that just lack basic research effort but are otherwise perfectly on topic, scoped, and answerable. –  slhck Jun 5 '12 at 12:45
3  
Fair enough - an absolute number would make sense. –  Pëkka Jun 5 '12 at 13:35
    
I propose a hybrid of the two, like the Tenacious badge: "Zero score accepted answers: more than 5 and 20% of total". Perhaps if it said "Downvotes that resulted in closure: more than 50 and 40% of total" –  gobernador Jun 5 '12 at 13:51
    
Updated to remove the percentage - after some thinking, I agree that could be harmful. –  Pëkka Jun 5 '12 at 18:08
    
Again a problem that would be solved if only downvotes on answers count towards the badge. –  Yannis Jun 5 '12 at 18:42
    
@YannisRizos so what, only count downvotes on answers that go to be deleted? –  Ben Brocka Jun 5 '12 at 19:08
    
I think it is har dto judge what a good downvote is even when you seem to be basing it on some objective criterion. –  Michael Chernick Jun 7 '12 at 21:18

What about:

“being the first person to downvote X number, not percent, of posts that are then downvoted by at least two other people”

share|improve this answer
1  
Hmmm. An interesting idea too, but very, very susceptible to horde downvoting... a chat room team could help each other get the badge, which would be much harder using closevotes. –  Pëkka Jun 6 '12 at 15:26
    
@Pekka, agreed but this would be able to be detected from voting history –  Ian Ringrose Jun 6 '12 at 22:03
    
I agree with this idea. Otherwise it would encourage anyone who sees a -1 question to downvote. –  Junuxx Jun 8 '12 at 20:02
    
@Junuxx why? Only a subsequent closing would make the downvotes "pay" towards the badge. Other peoples' downvotes wouldn't (and I think that's a key element of the idea) –  Pëkka Jun 8 '12 at 20:13
    
@Pekka: Yes, but don't you think that questions with many downvotes easily attract close votes? –  Junuxx Jun 8 '12 at 20:18
    
@Junuxx some do, some don't - plus, an unjust closing can be reverted by 5 users or a mod. It takes 5 votes to close, I wouldn't expect badge-related misuse to become rampant this way. Still, I only now understand the brilliance of Ian's suggestion - I overlooked the "then downvoted". Hmm, interesting. –  Pëkka Jun 8 '12 at 20:21

Prophet of Doom

50 downvotes on posts that are later:

  • Deleted
  • Closed (except migrations and duplicate)

Many duplicates are good questions, but just use different keywords.

Doomsayer

500 downvotes

Pale Horse Rider

5000 downvotes

share|improve this answer
3  
Yeah! I like it. And yeah, fair enough re duplicate closings, I'll add those to the list as exceptions. –  Pëkka Jun 7 '12 at 22:36
4  
Pale Horse Rider - When the Lamb opened the fourth seal, I heard the voice of the fourth living creature say, "Come and see!" I looked and there before me was a pale horse! Its rider was named Death, and Hades was following close behind him. They were given power over a fourth of the earth to kill by sword, famine and plague, and by the wild beasts of the earth. –  gnat Jun 8 '12 at 11:17
7  
60, 600 & 6000 as this is pure evil. HaHaHA –  Jeremy Thompson Jun 10 '12 at 7:04

I think this is likely to end up being almost the same as the Marshal badge in that you just make sure to downvote any answer/question before you raise a flag on them.

So if these badges would be introduced I'd suggest that there should be some kind of divider between this and the Marshal one so that any posts that you flag and downvote, would only count for one of the badges.

share|improve this answer
    
Hmm, I'm not sure I agree: If I flag a post, and action is taken, why should it not count towards both badges? I don't see actual abuse in the pattern you describe. Don't forget, a moderator will independently review your flag, and only if they agree there's an issue, there will be action. –  Pëkka Jun 7 '12 at 22:46
    
@Pekka: Not abuse as such, but there won't be any difference, everyone with one of the badges will almost automatically have both (if they want to at least). –  ho1 Jun 7 '12 at 22:48
    
true, that might indeed happen. –  Pëkka Jun 7 '12 at 22:48

Your goal of encouraging downvotes when warranted is an excellent one. The trick here is figuring out the "when warranted" clause - how can we separate justified, "honest" downvotes from downvotes submitted merely to game the system?

You propose that a good indicator of a warranted downvote is whether or not the question was, ultimately, rejected (in one way or another). Let's consider this.

Is the "ultimate fate <--> warranted downvote" correlation correct?

Somewhat. In most cases, a post being closed/deleted is ample de facto justification for a downvote. However:

  • Downvotes != Close votes. They serve two different functions - sometimes related, sometimes not. There's a huge difference between "this is a lousy, low-quality question but it's firmly on topic and will get reasonable responses" vs. "this is an awesome question, but it needs to be migrated to a different site." So the correlation is far from reliable - and you really don't want to encourage people to start coupling the two too tightly.
  • Answer deletion is quite rare. A downvote on an answer is first and foremost an "I disagree" vote; in most such cases, there's no reason for the post to be deleted even if it scores very low. Spam and comments-as-answers are the exception, but they need flagging - not downvotes. So you're really mostly addressing questions here.
  • Edge cases. You're going to have a lot of edge cases to contend with here. e.g., You're already limiting your scope to particular close reasons and not others. What about questions that are subsequently reopened? What about questions that are only closed much later? What about questions in the middle of close/open war? etc., etc.

In summary, I like the heuristic you're proposing as a rough estimate of what you're trying to measure, but I'm not sure it translates easily into a formal well-defined badge.

Would this badge promote gaming and abuse?

Oh, yes. You can't take a true measure of what you want - "well-justified downvotes"; if you could, maybe you could automate voting, and you wouldn't need SE :P But what's left is an approximation - which depends on behavior of other people - and which is fairly easy to manipulate.

Just as you're trying to use the result to test the original action, people can choose their actions to aim for the desired result. For example, one could systematically search for questions with low scores or with any close votes (it's right there in the Review panel!...) and downvote them willy-nilly. Or create lousy questions using sockpuppets, and jump in to downvote them. Or downvote lots and lots of new questions, because it's free and some of them are going to be closed.

So your approximation may be useful for academic purposes, but if it's introduced as a badge, it will be extremely easy to game - in ways which will be directly detrimental to the site at large.

share|improve this answer
    
Eloquently put and good points. I'm not that worried about the edge cases - they are too rare to be a factor IMO, and the system could wait for a while before calculating the badge stats, during which time most wars and reopenings will have been resolved one way or the other. Re gaming - hmm. I'm not sure I agree whether the risk is that huge, but fraudulent voting is indeed undetectable and this would have to be thought through very carefully. Unfortunately, the fact that downvotes on questions are free is a problematic factor here - gaming would be much less attractive if they weren't. –  Pëkka Jun 8 '12 at 11:16
    
Answer deletion is quite rare how do you know that? –  Some Helpful Commenter Jun 8 '12 at 15:50
    
@SomeHelpfulCommenter: Well, that's my own experience moderating Writers.SE, which I acknowledge works very differently than SO and others. Nonetheless, comparing answer deletions with question closures+deletions seems to me like a whole 'nothe ballpark. Questions have lots of close reasons - what deletion reasons are there for answers, except self-deletion, spam, comment-as-answer, and offensive content? –  Ziv Jun 8 '12 at 16:12
    
@Ziv what's wrong with the deletion reason list you gave? Isn't that enough? –  Some Helpful Commenter Jun 8 '12 at 16:18
    
@SomeHelpfulCommenter: All but self-deletion deserve flags, not downvotes. Self-deletion is more indicative, but based on OP's actions, and hardly a sound base for such a basge. –  Ziv Jun 8 '12 at 16:27
    
Speaking from the perspective of Gaming.SE - answer deletion is our bread and butter. We get more incoherent crap answers than I know what to do with. –  LessPop_MoreFizz Jun 11 '12 at 12:50

Alpha and Ringleader

The problem with this proposal is that it conflates downvoting with voting to close, and while the two are related they're not the same. A badge that rewards downvoting based on whether questions are closed will confuse the two ideas more than they already are.

Badgest that reward accuracy seem like a good idea, so I'd suggest that we modify the proposal so that there's a badge for accurate voting (up or down) and a badge for accurate closing. They'd work like this:

Ringleader: 80% accuracy among first up or down votes on 40 questions. Accuracy means that your vote was followed by 3 or more similar votes.

Alpha: 90% accuracy among first close votes on 40 questions. Accuracy means that the question was closed as a result of your vote.

So, to determine eligibility for Ringleader, you'd look at the last 40 questions where the user in question was first to vote up or down. If their up or down vote was followed by 3 or more similar votes in 80% of the cases, they get the badge. Similarly, for Alpha, you'd look at the last 40 questions where the user was first to vote to close. If the question was closed due in part to their vote 90% of the time, they get the badge.

The idea in each case is to encourage users to vote early, but to carefully consider whether others will follow their lead. You don't want to be the first to vote to close if you're not pretty sure that others will agree.

share|improve this answer
1  
Hey, these are really cool alternatives that help address the main problems. –  Pëkka Mar 12 '13 at 20:13
    
Both of these badges would be actively bad. Voting exists so users can express their evaluations of a post. Awarding people for up/down voting the same as other people would undercut the meaning of voting. It would encourage people to vote the way people think others would like, instead of to vote their conscience. This would be destructive in the same way letting people see each others' votes would be destructive. As for the Alpha badge your propose, there's nothing inherently good about questions you vote to close being closed. Questions get wrongly closed (and reopened) all the time. –  Eliah Kagan Mar 13 '13 at 4:36

You must log in to answer this question.

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