0

Close vote review queue grows fast! Currently it has 101.7k questions pending. My idea, based on a simple research from this answer is to introduce another queue that will partially balance the close vote queue. It should work as follows:

  • Questions from users with less than, let's say, 10 rep won't be shown immediately on the site. Instead they would go to a buffer, namely: Novice questions approval queue.

  • This queue would be presented along with other queues on the Review page, and would be available for users with, let's say, at least 2000 rep.

  • For each presented question, in most simple version, there would be two options: Approve and Reject (possibly also Edit & Approve).

  • The new queue could of course offer some gamification infrastructure: small rep rewards for reviewing for users, lets say, under 3000 rep, and analogously to other queues: badges.

  • If in the meantime OP would manage to earn enough rep, their question would get published automatically.

Roughly relating to these statistics if we assume that all users that were to ask "good questions" (score 3 or more) would be completely discouraged by this policy and give up asking at all, we would lost 2% of "good questions", while the Close votes queue length would be reduced by 23%. What's more important, it seems that the queue would at least grow slower then now.

Differences from First posts queue:

  • First posts queue handles both questions and answers. The latter do not seem a big problem on SO.

  • This queue would somehow shield the page and close vote queue from very low quality questions.

  • First post queue is based on the post being ... a first post :) This queue would work with regard to rep.

EDIT: In response to some points made in the comments&answers:

  • This would not prevent people from posting completely. Earning 10 rep isn't much work, and would give them right to post without review.

  • Any arguments in regard to gamification encouraging bad behaviour are not specific to this proposal - they describe valid problems, but in regard to the gamification idea in general.

  • I think that this would introduce better behaviour of the queues (instead of one Close votes queue): for example it's very easy not to approve a question by a lazy programmer, who didn't even bother to spell correctly. Such easy to review questions are currently hidden in thousands of other, more complicated cases in the Close votes queue.

  • two smaller problems are easier to handle than one big problem :)

32
  • The first post queue accomplishes the goals that this queue would, more or less, already. It doesn't prevent the question from being seen, but it provides the opportunity for any needed moderation to take place, and the items in the queue tend to be handled quite quickly.
    – Servy
    Jan 3, 2014 at 16:00
  • 1
    @Servy But the first post queue doesn't help at all with insanely fast growing close vote queue.
    – BartoszKP
    Jan 3, 2014 at 16:02
  • 1
    One of the nice things is you require no rep to ask a question or answer it. This changes that to be "you are free to ask questions if you have a minimum rep or if enough random people think it is acceptable". Jan 3, 2014 at 16:04
  • 1
    @psubsee2003 The whole site's model is based on "what random people think" about your questions and answers :)
    – BartoszKP
    Jan 3, 2014 at 16:07
  • @BartoszKP as a collective, 5 or 10 is hardly a collection opinion. Jan 3, 2014 at 16:08
  • 3
    Also, adding a rep award would absolutely destroy this queue. Look what little bronze, silver, and gold bits of pixels have done to the other review queues. Users blindly click to get rewards and don't really care about the quality. Jan 3, 2014 at 16:08
  • @psubsee2003 Yet that's how all queues work.
    – BartoszKP
    Jan 3, 2014 at 16:08
  • @psubsee2003 That's the general problem with gamification, not specific to this proposal.
    – BartoszKP
    Jan 3, 2014 at 16:09
  • @BartoszKP right, but you are preventing someone from posting at all based on a small sample, not a collective opinion. Jan 3, 2014 at 16:09
  • @psubsee2003 Well, not preventing entirely. Earning 10 rep is not much work, and would give you right to post also.
    – BartoszKP
    Jan 3, 2014 at 16:10
  • @BartoszKP It would be a heck of a lot more work if you make it really hard to post questions.
    – Servy
    Jan 3, 2014 at 16:21
  • @Servy I agree - but isn't that what we expect from OPs? We expect that they are ready to give some effort when solving their problems. We don't like people who want to get the solution for free. Hopefully this would improve quality.
    – BartoszKP
    Jan 3, 2014 at 16:23
  • @BartoszKP You said that getting 10 rep is really easy, I was simply saying that that's going to be much less true if this change is made. And while I do agree that asking good questions is hard and takes work, I don't see this as a good solution. Users will just end up re-posting their questions as they don't get posted, they won't be exposed to lots of people for editing and comment feedback, and those they are exposed to are less likely to be active in the relevant tags, the feedback would be harder to see for the OP, the queue is open to abuse, and I don't see it reducing the workload.
    – Servy
    Jan 3, 2014 at 16:27
  • @Servy I said that it "isn't much work", while it of course requires some work :) Comments could be enabled for reviers, feedback would be possible. As for your other objections - I get your point. I can't prove that the dynamics of increasing workload will get better, I just feel like it will (see my edit) :) It's in fact a divide-and-conquer approach - two smaller problems may be easier to handle, than one, currently big, problem.
    – BartoszKP
    Jan 3, 2014 at 16:31
  • This new queue will just get even bigger than the close votes queue and will make new users upset because their question still hasn't been approved.
    – ɥʇǝS
    Jan 3, 2014 at 23:14

2 Answers 2

8

I have several problems with this issue, one will completely ruin this proposed review queue and make it completely worthless, and the other goes somewhat against what Stack Overflow is about.

Questions from users with less than, let's say, 10 rep won't be shown immediately on the site. Instead they would go to a buffer, namely: Novice questions approval queue.

All fine and dandy, but Stack Overflow (and Stack Exchange) is all about being an open site where anyone can ask a question or leave an answer without any rep. Now you are putting a roadblock up to possible prevent users from doing that.

I certainly agree that 10 rep is a small amount, but I object to the notion that "as a new user, you can only post a question if a random sampling of 5 people who may or may not be qualified to judge your post think it is acceptable". It just smacks of elitism.

And what about sites that still allow unregistered users? This would effectively require them to have all of their questions "approved" before it would appear.

And when questions do get rejected, since they didn't get posted, how do you propose to provide feedback to the user? Given what happens with closed questions getting reasked, I image that even more users will just keep reasking until there question makes it through (and given the state of reviewers and the review queues in general), you are going to end up with users who reask eventually getting their post through.

The new queue could of course offer some gamification infrastructure: small rep rewards for reviewing for users, lets say, under 3000 rep, and analogously to other queues: badges.

Gamification will destory this queue. I know you have said this is a problem with gamification itself and not this specific idea, but I disagree. Any time any kind of gamification is added to the site, it causes problems. Reviewers are falling over themselves to get hats, and before that it was little colored pixels (badges) that ruined the original review queues over 15 months ago, and we still haven't really fixed it completely (although it is better). The moment you add rep to the game, you will literally render the queue worthless and cause more harm than good. Instead of reviewers working in the best interest of the site to review posts and make objective decisions on quality, you will end up with a not insignificant percentage who review for speed and not accuracy/quality so they can review any many posts as possible and game as much rep as possible.


And let's not forget what Shog has mentioned several times recently. The problem in the close vote queue is not new questions, it is the long tail of backlogged questions in specific tags. So this does nothing to solve that immediate problem.

1
  • All right, you've convinced me. Thanks for your feedback!
    – BartoszKP
    Jan 4, 2014 at 16:50
1

First like Servy said we already have a queue for first posts and new users. It does pretty much the same thing.

Now about your comment :

@Servy But the first post queue doesn't help at all with insanely fast growing close vote queue.

I don't think splitting the queue into two different queues really a solution. I mean, trading an insanely large queue for two very large queues isn't a solution.

3
  • Well yes, you got a point there that roughly speaking it's just splitting the queue into two queues. It's impossible however to predict how dynamics of these two queues will behave. So I say, let's give it a try. I think it will improve the quality of this site.
    – BartoszKP
    Jan 3, 2014 at 16:12
  • What do you mean by "Predict how they will behave". They are the same issue with different name no ? Jan 3, 2014 at 16:13
  • Not exactly. Only a specific subset of questions would hit this novice queue, and I say it's a subset that's easier to review. Questions from "lazy programmers" for example are very easy to diagnose. They are in the close queue already of course, but hidden in thousands of other, more complicated cases.
    – BartoszKP
    Jan 3, 2014 at 16:15

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