# What to do about long review queues?

There are now often very long review queues, especially the close vote queue. Accumulating the necessary 5 close votes typically takes more than 24 hours, often twice as long. What can be done?

Over at math SE, I get 50 close votes a day, but here at CV I only get 20. Why the difference? Can we increase the limit?

• (+1) This is peculiar. I thought that the VTC limit for a user grew in accordance with reputation in a manner consistent across all of SE, but that can't be true if you have more rep on CV.SE than on math.SE, yet a larger VTC limit on mathSE. FWIW, I am the cause of the large number of old questions with neural-networks,machine-learning, etc. tags being closed. This is because a large slice of these unanswered questions are duplicates, unclear, or too broad to be answerable. – Sycorax Aug 18 '18 at 18:48
• ... I'm not undertaking this closure effort to cause anyone trouble. I am trying to improve the quality of the content with these tags, since they are a bit of a "Wild West" right now. I'm also undertaking to provide answers where I can, and upvotes (or bounties!) where appropriate. In the long-term, I'm attempting to gain the gold badge for machine-learning and neural-networks which will allow me to close-as-duplicate questions unilaterally, and alleviate the queue backlogs for these tags. – Sycorax Aug 18 '18 at 18:52
• @Sycorax Godspeed! – amoeba Aug 18 '18 at 22:11
• @amoeba It's ... slow going. I'm about half-way there for each of machine-learning and neural-networks. – Sycorax Aug 18 '18 at 22:47
• @Sycorax I know. I still never got one for pca. – amoeba Aug 19 '18 at 7:28
• I'm responsible for a bunch of recent flags (and comment -> CW answers). Apologies for the extra work, I'm trying to clear up the unanswered regression questions list a bit. – mkt Aug 19 '18 at 12:24
• I'm also contributing to the long review queue! that is not really the problem, it is good that Qs that should be closed gets closed. But that's not happening effectively now, the question is what to do with that. Either getting more reviewers (how?), or that existing reviewers get more votes. That seems more practical, but how? – kjetil b halvorsen Aug 19 '18 at 14:06

# Solution: Theme Week!

I'm just spitballing, but what if we have like a week-long theme where we propose a (prolific) tag and the goal is for participants to collaborate in cleaning up that tag? Cleanup goals could include (1) identifying duplicates (2) identifying "blind spots" where a canonical thread is required (3) answering neglected questions as well as anything else we think might be necessary. The idea is that because everyone is working towards the same, common goal in a focused way, participation will be more rewarding.

If the "teamwork" aspect isn't enough reward in itself, I think that some high-rep users (I'm including myself) could be persuaded to award bounties to users who make significant contributions to the effort. We'll clearly have to work out the details, but I think it is within the mechanisms of CV.SE and the spirit of the SE organization to reward these kinds of computation.

(This is a wholesale rewrite of my answer, which was more of a comment. Interested readers can find it in the edit history.)

• I think this is a good idea, discussion of which tag(s) should be themes belongs in its own thread. Meanwhile I am still waiting if anyone knows about how/if vote limits can be increased. – kjetil b halvorsen Aug 22 '18 at 16:00
• I suspect the VTC limits are governed by some sort of SE-wide policy -- perhaps asking on SE meta would be the best place to find a solution? – Sycorax Aug 22 '18 at 16:02
• Will do so! ... – kjetil b halvorsen Aug 22 '18 at 16:04
• @kjetilbhalvorsen See meta.stackexchange.com/a/147981. Apparently the limit is 20 but it gets doubled if a queue has a "large backlog". – amoeba Aug 22 '18 at 19:51
• This is a great idea! – mkt Aug 23 '18 at 6:29

The current context seems to be a permanent backlog of ~40 items in the Close Review Queue. In answer to the original question "What can be done":

1. Suggestion: we could recruit 10 more reviewers to each review 20 a day and then taper off as the queue gets manageable.
Problem: how to find them?

2. Suggestion: we could find out why Kjetil has 50 votes on another stack and see if we can get the same here.
Problem: he and others who currently review might not want to do 50 a day and concentrating so much in the hands of a few reviewers is not best practice as their idiosyncrasies dominate the site.

3. Suggestion: moderators could have occasional purges when things get beyond non-mod users' capabilities.
Problem: community moderating is preferred, see 2 above.

4. Suggestion: reduce our efforts to clean up the site by only issuing votes-to-close three or four times a day which if we then review 20 keeps it in balance (I think).
Problem: this defeats the aim of having a site of good quality questions and answers.

I agree this all comes from the Department of the Blindingly Obvious but perhaps others may have contributions.

• Re #1, we might as well recruit the tooth fairy. There are only so many users >3k rep & they may not want to spend a lot of time every day reviewing close votes--there's nothing we can do to induce them to do so. Re #2 &3, SE sites will necessarily reflect the idiosyncrasies of their most active users; that's mitigated by a large enough pool of highly active users (ie Stack Overflow), but nowhere else--it's inescapable. Re #3, moderators do do occasional purges, they would need to be less 'occasional' until there is sufficient user review activity. The tension in #4 is also intrinsic to SE sites. – gung Aug 22 '18 at 13:26
• (The Department of the Blindingly Obvious responds to itself.) – gung Aug 22 '18 at 13:26

I don't yet have enough rep to review the close votes queue, but have been doing a fair amount of flagging lately. If there's been a small surge of similar activity by other low-rep users (which I think would be a good thing), that might explain the extra work y'all are facing. In which case, the problem should reduce as more of us breach the 3k threshold and can help with reviewing.