This confusion came up in comments on another Meta question.

On the site I use most actively (RPG.SE), I've noticed that the counts for each review queue appear to only indicate the reviews available to me requiring my action. Just to be clear, I'm talking about this number here:

0 questions in close vote review queue

Since I'm a 10k user, that's pretty easy to tell: even after doing all my reviews, and the review page saying there's 0 items in each queue, I might still have that 10k notice in my top bar saying there's [5] reviews pending, or some other number. So they must still be there, just waiting on other users to finish them, and it just doesn't show them to me. This is fine, I like this behaviour.

However, on Meta SE about an hour ago, there was 1 item in the close vote review queue. I reviewed that item. The count stayed at 1 on the review page. When I clicked to re-enter the queue in case there were new items, I was presented with this:

"There are no items for you to review."

Another user in that meta question I linked claims this is normal behaviour for them on their own site (Christianity.SE I think?).

What's going on here? Are the review queues supposed to be customised just to what I personally can actually review at the time? Is there a bug, are sites customised differently, or is there just some kind of caching thing leading to a time delay on the review count getting updated? (As they say, it is always caching, but is caching really the culprit here too?)


1 Answer 1


Caching. And some complicated eligibility rules.

Most of the time, the top bar will indicate what's still TO DO for the site, while the numbers under /review will indicate what's still TO DO for you. But when reviews are being done quickly, they can become out-of-date briefly. The system recalculates them every few minutes, so generally they won't be out of date for too long...

...But there are still some edge-cases where there'll be items in a queue that you aren't eligible to review but which are disqualified because of tests too expensive for the system to run until/unless you actually try to review them.

It's unlikely you'll encounter the latter very often, but if you see a discrepancy hanging around for more than 5-10 minutes, chances are that's what it is.

  • I assume this applies to suggested edit queue too? I had just a similar problem with it over at SFF Commented Mar 6, 2017 at 16:15
  • Probably, @Gallifreyan. We're gonna try & change the rules for how that indicator shows up, but to some extent this is unavoidable given it's an asynchronous system.
    – Shog9
    Commented Mar 6, 2017 at 21:33
  • Is This new layout for review stats and button supposed to be the fix for this issue?
    – NSNoob
    Commented Apr 7, 2017 at 10:21
  • 1
    Well... It's a fix, @NSNoob. It ain't ideal; the ideal implementation would light up the indicator when - and only when - there are tasks that you need to do when you haven't been reviewing for a while. But it is an improvement: there's a delay, no bogus number, and a check for some reasonably-sane number of pending tasks.
    – Shog9
    Commented Apr 7, 2017 at 21:54

You must log in to answer this question.

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