63

I've noticed that the number of reviews required to remove a post from a queue varies. For example, to accept/reject a suggested edit, you need two votes of the same kind. If the votes are conflicting, the number goes higher. This seems to be the case for the other queues as well. (Additionally, I think that a mod-review is binding)

So, my question is: what are the criteria for a review to be marked as a "completed" review? Also, when is a review considered invalidated (i.e. "no longer reviewable")?

Formerly When is a post removed from a review queue?.

Return to FAQ index

4
  • 1
    Don't suggested edit simply need 2 agreeing votes on all sites, expect SO where it is 3?
    – Bart
    Commented Jan 19, 2013 at 17:29
  • @Bart: Yeah, I think that's the case. But I'd like to know about the other queues as well. Commented Jan 19, 2013 at 17:31
  • What I know is: moderator review actions are binding :P
    – slhck
    Commented Jan 19, 2013 at 17:37
  • @slhck: yay! moar powaz for me! Commented Jan 19, 2013 at 17:39

1 Answer 1

71
+50

A review is considered "completed" if a single user's review immediately removed that post from the review queue (e.g. making the final review that closes or deletes a post). If the review task is removed due to other circumstances outside the queue (e.g. a moderator deletes a post that is being reviewed, outside the queue), it is marked "invalidated" or "no longer reviewable".

The following lists the general criteria for a review task to be removed from a review queue.

All review queues

  • One moderator action - the moderator's review will be the binding, final outcome
  • Post is deleted[1]
  • Post is locked

Close votes

  • Three "Leave Open" reviews and "Edit" reviews combined
  • Post gets closed
  • All close votes and recommend closure flags on the question expire, are retracted, or are invalidated for another reason
    • Remember that if there are pending votes to close question A as a duplicate of question B, and question B gets closed as a duplicate of question A, these pending votes will be invalidated. The same thing happens if question B (the proposed duplicate target) gets deleted.
  • Someone starts a bounty on the question
  • The question ends up in the Triage review queue

Reopen votes

  • Three "Leave Closed" reviews
  • Post gets reopened
  • All reopen votes on the question expire or are retracted
  • The question was added to the queue through an edit after being closed, and the user who edited the question later flags it as spam or rude/abusive, and there are no active reopen votes from others cast more than 15 minutes ago.

Suggested edits

  • Two "Approve" reviews
  • Two "Reject" reviews
  • One review from the post author
  • Community user making a review:
  • The post is rolled back
  • The tag ceases to exist (tag wiki edits only)

Low quality posts/answers

  • Six "Recommend Deletion" or "Delete" reviews for answers (four on Stack Overflow)
    • Note that a "Delete" review (by a trusted user) that brings the answer up to three delete votes will already delete the post regardless of review queue logic, and therefore complete the review task (see "All review queues" above)
  • One "Edit" review
  • "Looks OK" reviews: a default number set by the site (2 on Stack Overflow, 1 on other sites), plus the number of pending "very low quality" and "not an answer" flags cast by users.
    • For example, if there are no active flags (i.e. the post was added to the queue due to failing the system's quality test), a single "Looks OK" review will dismiss it on most sites (two reviews on Stack Overflow). However, if there are three "very low quality" flags cast by users, four "Looks OK" reviews are required to dismiss it (five on Stack Overflow).
    • This check considers the current state of flags, so if a new flag is cast while the post is still being reviewed, an additional "Looks OK" review will be required to dismiss it.
    • If the review is completed with a "Looks OK" outcome, but the "Looks OK" reviews did not constitute a majority of reviews (i.e. there were as many or more reviews in favor of closure or deletion as "Looks OK" reviews), an automatic moderator flag will be raised.
  • The post gets closed
  • The post is edited outside of the review UI and the new revision passes the system's quality test
  • Any active "not an answer", "very low quality", or "low quality (auto)" flags on the post are dismissed by a moderator (for any reason, helpful or declined)

First questions, First answers, and Late answers

  • One "Looks OK" review
  • One reviewer performs an action and clicks "Other action"
  • For First questions and First answers only:
    • One reviewer uses the "Share feedback" option
    • The post becomes 14 days old
    • The post scores 3 or higher
    • The post ends up in another review queue
  • For First questions only: the question gets closed
  • For First answers only: the author's reputation becomes 125 or higher

Triage

  • Three "Looks OK" reviews
  • Three "Needs community edit" reviews
  • Three "Needs author edit" and "Flag" reviews combined
  • A moderator dismisses any pending "very low quality" or "low quality (auto)" flags on the question

[1]: With the exception of the Suggested Edits queue, though this is considered a .

7
  • 2
    +1 for summary. I didn't know that only single action is required for first posts or late answers. Is it really so little - no consensus needed? Commented Dec 1, 2013 at 16:31
  • 2
    @ŁukaszL. Yep, but we're fighting it.
    – animuson StaffMod
    Commented Dec 1, 2013 at 16:32
  • 3
    Technically First Posts and Late Answers require both an action and a click on "Done"... the amount of times I've accidentally clicked "skip" :-). Commented Mar 26, 2014 at 14:45
  • Shall we make it explicit that "Skip" is not an action? Commented Oct 8, 2017 at 16:34
  • 3
    If a post is in both First Posts and Late Answers, how do actions from one queue affect the other? (I believe the same reviewer won't see it in the second one after handling it in the first, but does that affect anybody else?) Commented Oct 8, 2017 at 18:09
  • 2
    For invalidating reopen reviews: It appears contrary to the purpose of reopen reviews to invalidate the review queue task because the user who edited the question voted to reopen. If anything, such a vote should be an indicator that the review should be given a (somewhat) higher priority in the reopen review queue, not kicked out. Sure, if the editor flags as spam/R/A, that's definitely an indicator that they don't think it should be reopened, but a reopen vote is a strong indicator that the user thinks it should be reopened.
    – Makyen
    Commented Mar 18, 2021 at 16:31
  • @Makyen I think the bug is fixed with the last review queue workflows update (cf. bullet point 5. under "Changes to the Reopen votes queue"). Commented Jul 25, 2022 at 15:04

You must log in to answer this question.

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