23

Due to large number of reviews waiting do be done - at the time about 46,000 - did you consider - maybe temporary increase daily close votes limit from 40 to higher number ?

I know there is limit daily 50 question that we can vote to close, however review limit could be higher - not every review need to be closed.

Improve this question
10
  • 19
    Can't get to that top spot fast enough? ;)
    – Bill the Lizard
    Jan 23, 2013 at 2:13
  • a workaround would be to vote-close from outside of queue, making "fake skip" in it
    – gnat
    Jan 23, 2013 at 9:25
  • 1
    Two weeks later it's up over 47k. Given that cleaning up dross is surely one of the most important activities in keeping SO's quality level up, and it's clear that we as a community are not keeping up with the influx of dross at the moment, I think this limit should be raised a lot.
    – Vicky
    Feb 5, 2013 at 9:20
  • Or it could be the close-votes that are, uh, dross. I've noticed a ton of off-topic votes on questions that are clearly on-topic. It seems to be the catch-all vote for any question some elitist cliques don't like. Regardless, the tide has turned. Let's see if it's permanent (two weeks is hardly any time at all, try 6 to 8 weeks instead) or if the count will continue crawling downward. =)
    – J. Steen
    Feb 6, 2013 at 12:44
  • @J.Steen, surely if the close votes are dross they will attract "leave open" actions which cancel them out, and they disappear from the queue? Or is that not how it works?
    – Vicky
    Feb 6, 2013 at 12:56
  • @Vicky 5 do not close votes are needed, I believe.
    – J. Steen
    Feb 6, 2013 at 13:00
  • 2
    Reviewing the comments at Huge close votes review queue... it seems like 46k was the low water mark of the queue. It has been over 50k for months and is now (June 2, 2013) 53.7k. I think we should do let votes accumulate and rollover or increase votes by formula if not this. I'm in favor of a flat raise to 100 votes per day until the backlog is cleared.
    – Old Pro
    Jun 2, 2013 at 18:11
  • I'm adding another bounty to this question because I think this is really really important. There are a number of good suggestions on the table about how to deal with it now.
    – Vicky
    Jul 26, 2013 at 12:15
  • 1
    Oh gosh. Can't have this! When I've used up my close votes, I take that as a sign that I've spent too much time on SO already for the day.
    – Cody Gray - on strike
    Jul 26, 2013 at 12:17
  • 1
    Looks like they are listening! meta.stackexchange.com/questions/195352/…
    – Sergio
    Sep 1, 2013 at 16:57

2 Answers 2

Reset to default
14
+100

For the last two weeks the CV queue doesn't seem to decrease. It is more like stagnant between 47-49k questions (I haven't seen it fall below 47k, as of writing it is around 48.5k).

Wouldn't it be better if the limit would be relaxed (to for example 60, with the 50 close vote limit still in place), BUT reviewers wouldn't get review points after the first 40 review (so robo-reviewers would have no real reason to continue reviewing apart from not noticing that they are past this limit).

This could be also improved that only the first 20 reviews would count for the badges (like for other queues), but if someone wants to they can continue reviewing until the second, more relaxed limit (like 40 or 60). This would also discourage robo-reviewers in the close vote queue a bit (but encourage others to do more reviewing, and empty the queue faster)

Improve this answer
1
  • 3
    Yes, +1. I'd happily help out a bit more with no extra point/badge incentives.
    – halfer
    May 20, 2013 at 22:17
5

As Manishearth points out, the daily limit is one way to catch robo reviewers. I would like to suggest an alternate approach that balances the need to stop bad reviewers with the need to resolve the open queue items.

I suggest that we increase the frequency of the review audits as the number of daily reviews increases. I currently see an audit quite rarely, perhaps once every forty reviews. What if we increased the rate after forty reviews to (for example) an audit every fifteen questions, and increased it again after 80 reviews?

As the audit frequency increases, real reviewers may eventually take a break, but this will be a self-imposed custom limit, rather than an arbitrary one. This provides the double benefit of allowing motivated reviewers to continue chipping away at the queue while increasing the chances that a robo-reviewer will be "caught."

Improve this answer
7
  • 1
    Ok, good point.
    – Joe
    Jul 17, 2013 at 19:12
  • This is better than not doing anything, but has the problem of discouraging real reviewers from using all of their (additional) votes. If they take breaks because of frequent audits, as you state, then there's a good chance they'll just get fed up and quit reviewing for the day (whether intentionally or because they were on break at UTC 00:00), thus wasting a chunk of their votes. Robo-reviewers won't be discouraged at all, unless they fail enough audits to get banned. While that chance does increase, it's still downright trivial to bypass audits.
    – Esoteric Screen Name
    Jul 17, 2013 at 20:53
  • @EsotericScreenName I addressed this in the post. Reviewers will be free to continue or stop whenever they choose. When the audits become too frequent, they may choose to stop. Regardless of whether the motivated reviewer chooses to stop at 45 or 105, the number of reviews completed above the current limit of 40 will be an improvement over the current situation where the hard limit is enforced for everyone. Robo reviewers will be stopped by the automatic ban after failing too many audits.
    – George Cummins
    Jul 17, 2013 at 20:58
  • @GeorgeCummins And I addressed that in my comment. The net effect is to increase both the robo-to-real vote ratio and the absolute number of real votes, with the former increasing faster than the latter. Robo-reviewers will be stopped after getting banned, but ban evasion is too easy.
    – Esoteric Screen Name
    Jul 17, 2013 at 21:00
  • @EsotericScreenName "with the former increasing faster than the latter." Do you have data to back up that claim? I suspect that the robo-reviewers will be stopped rather quickly as the audit review frequency increases. Since the ban lasts for one week and "real" reviewers can continue to contribute daily, I see only a net positive.
    – George Cummins
    Jul 17, 2013 at 21:04
  • @GeorgeCummins robo-reviewers will be stopped rather quickly Do you have data to back up that claim? Because it's obviously still a large problem despite frequent audits; e.g., the suggested edit queue.
    – Esoteric Screen Name
    Jul 17, 2013 at 21:07
  • @EsotericScreenName You'll note I prefaced my sentence with "I suspect." It isn't a claim, thus it should be clear that I do not currently have data to support it. This comment thread appears to be degenerating beyond the point of usefulness.
    – George Cummins
    Jul 17, 2013 at 21:11

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