8

After reading Why do you not review? and stumbling across Do we need a event to clear out the Close Queue?. It occurred to me that the problem might be a lack of incentive.

After 1000 such reviews, It would seem that anyone who is concerned with their rep or badges might lack motivation to continue, stop reviewing and just go back to answering questions. There's nothing wrong with that IMHO, but there may be a flaw in the policy that provides 0 rep for reviewing the close queue and badges that can only be earned once topping out at 1000.

  • 1
    Excellent question! (I've got no answer, but just wanted to share that I'm wondering what I'm going to do once I hit that gold badge...) – Fabby Feb 21 '15 at 18:10
  • The problem it can be monotonous - similar fairly low quality questions with the occasional old one being duped, and once in a while tests to see if you are still awake. If it could vary the topic a bit it might be more interesting. However am fairly sure moaning about it on Meta so it ends up in the Hot posts thing as a reminder to look at it might help :) – Wilf Feb 21 '15 at 19:55
  • 1
    @Wilf I'm pretty sure the only thing that will help is actively working the queue, but it never hurts to try. – Elder Geek Feb 21 '15 at 19:57
  • @ElderGeek - on it now (fairly bored already - better than finishing physics though :) – Wilf Feb 21 '15 at 19:59
  • 1
    And it saying go away you have done 20 does not help - suppose it helps with bots and stuff... Nearly finished Physics :D – Wilf Feb 21 '15 at 21:07
  • @Wilf Agreed, although I think it's go away after 40 when the queue is over 1000.... – Elder Geek Feb 21 '15 at 21:20
  • 2
    Raise the limit per day. Right now it is 20 per day, and 40 if queue length > 1000. Make 40 if queue length > 500. – muru Feb 22 '15 at 1:34
  • 2
    I asked this in one of the linked threads you quote. Unix/Linux almost always has a single-digit close queue. I wonder what the difference is, and whether we could learn from them? – Sparhawk Feb 23 '15 at 13:39
  • 2
    Also, another option to help clear the close queue here is to lower the threshold from five votes to four votes. – Sparhawk Feb 23 '15 at 13:40
  • 2
    @Sparhawk I think U&L has a much higher standard of moderation. For example, you can see Gilles and Anthon (who are not diamond mods, IIRC) regularly patrol the questions, doing retagging and general cleanup, something I have seen 200_success do on Vi and Vim as well. I can almost predict when Gilles will drop by and correct the tags if I post a question. I suspect a bunch of users regularly go through the close review queue as well, and never let it go out of hand. Here, it's much more haphazard. – muru Feb 25 '15 at 13:44
  • @Sparhawk I'm assuming the calculations for close reason require a tie-breaker (majority) which would be impossible to guarantee with an even number of votes.. – Elder Geek Feb 25 '15 at 14:13
  • 1
    @muru That seems reasonable and simultaneously depressing, that it depends on so few. We need to hijack Gilles and hold him at askubuntu! – Sparhawk Feb 25 '15 at 21:55
  • @ElderGeek I'm not 100% sure how it works, but I'm fairly sure that you need five "yes" votes to close. I'm not sure how the "keep open" votes affect this. At very least, it's not "majority from five". – Sparhawk Feb 25 '15 at 21:56
  • @Sparhawk. Sorry about the confusion. I was referring to the close reason not whether it would be closed or not. – Elder Geek Feb 25 '15 at 21:59
  • 1
    I see that we are now under 1000 again, so boosting the number allowed to 40 when it's over 1000 seems to work eventually or this question drew some interest to the close queue. Perhaps reducing that threshhold from 1000 to say 800 would keep it more manageable. – Elder Geek Feb 25 '15 at 22:08
8

Yes

I would work the close queue more regularly if there was at least some incentive for this exceedingly boring, difficult and thankless job.

  • 1
    I'm not completely against additional reward for working on the close queue, but it shouldn't be a completely mechanical task either. The metric for awarding something there should be "you turned more than hundred similar poor questions into one good canonical question". The problem with that however would be that it should be a team effort, with more than one user or a very close circle of friends working on one topic… that may just be the point of view from someone who tries to avoid chatrooms. Also it shouldn't encourage, that one user seemingly "owns" a topic with one monolithic answer. – LiveWireBT Feb 21 '15 at 23:14
  • 1
    @LiveWireBT Agreed. It's certainly not a mechanical task. In order to make good choices one has to Read the questions, ascertain if there are any quality answers that would make it a candidate for merging with a known duplicate and flag as such if so, ascertain if it can be improved and do so if possible, ascertain if it can be answered with the information available, request more information if not, (just a short list of examples). I'm not sure how the current system would go about tracking turning 100 ?s into 1, but I've only been here a hair over a year so I may have missed something. OOSpc – Elder Geek Feb 22 '15 at 19:12
6

Idea (that never is going to be implemented ;) )

The review method might be set up smarter. Currently it shows 1 review per item. I believe it would be a bit better for some of the review options if more than 1 topics are lined up below each other and you could mark them all before submitting them.

Where this might be implemented for users with more than a specific reputation and only of the amount of reviews is above a threshold (like 1k+).

Like the other day: Takkat went on a frenzy marking a LOT of topics about "canon" printers against a duplicate. At some point closing them was more a drag than anything else ;)

Oh and if I had my say reviewing would be linked to hats >:-D Here is hoping no meta-SO user sees this

  • Point taken. Do you mean being able to mark say off-topic as well as unclear (as an example)? I did notice the "canon" topic coming up often in the queue. Hats, a pittance of rep, whatever, honestly I think close determinations are sometimes more difficult than answers (especially if you are attempting to do it properly). After a period of time policing the close queue is a difficult, thankless job that I think far too many of us avoid. – Elder Geek Feb 21 '15 at 18:57
  • 1
    No I meant that the system would line up say 5 questions that got marked as say duplicates from say Takkat that are similar say about a "canon" printer. If one of these is marked as a dupe it is likely all of them can be marked as dupe as (1) Takkat is a trusted user and (2) those question are easily visually checked for being accurately duped against a topic you saw at the 1st reviewed topic. Kind of a list like: here is one that got dupes, these X are duped too. And have 1 action deal with all. – Rinzwind Feb 21 '15 at 19:06
  • Ahh, that would be helpful (under certain circumstances). I'm more interested in finding a way to provide an incentive for those who are put off by the queue to work it. – Elder Geek Feb 21 '15 at 19:12

You must log in to answer this question.

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