Don't let flag-to-close flags linger for months in flag purgatory

Question

Here is the close queue:

Here is an active flag I have:

I currently have 5 flags pending in the close queue. Some, like this one, are months old. This doesn't feel right and may discourage newer users from flagging things that need attention with the assumption that a flag may be ignored for 4 months, so why bother?

Here are some possibilities:

1. Bump flags that have been languishing for more than 30 days to the front of the close queue (or weight questions with said flags a bit higher). Won't solve the issue, but will make it less egregious
2. Punishing a misbehaving mod by forcing them to go through every sub-3000 user's flagged close votes in the queue and free them from flag limbo
3. psubsee's suggestion, "a new 10K flag list for "stuck" flags, which would be any close flag that is unhandled within 30 days."

^{This question has been heavily edited under the assumption that flags didn't get refreshed every day. Apparently they do, thanks to Bill and Asheesh, so sorry about the misdirection. Languishing flags still an issue though!}

Answer 1

The way to deal with this would be to let flags to close expire, just like votes to close do.

Votes to close are known to expire, for the reasons explained eg in answers to this question: Why do close votes expire?

closing is supposed to be a (mostly) spontaneous action by the community to deal with closures on their own... community... didn't reach even a minor consensus as to whether the question should be closed

Nothing like this happen to flags to close cast by users lacking CV privileges.

Note how per recent changes flags like that are no longer targeted to moderators only, but instead, go to same review queue as votes to close. Given this change, one can argue that the reasoning for votes expiration can now be applied to flags.

_{Submitted this answer per advice from animuson here.}

---

It is worth noting that this approach has been mentioned at September meeting of SE team:

more aggressive dismissal or aging of "recommend closure" flags...

Answer 2

First off, close flags don't expire because they don't have to. There's no implicit action tied to close flags; no matter how many of them a post collects, they don't bring it any closer to being closed. There's no concern that a gradual accumulation of close flags will eventually cause reasonable questions to be closed.

That said, it is a problem if they go too long unhandled: a flag that everyone is ignoring is effectively useless.

Just aging them away doesn't fix this; that just converts an effectively useless flag into one bearing a great big system-applied "YOU WASTED YOUR TIME" stamp. There needs to be some way to determine an actual resolution for the flag, even if that resolution is "you're wrong, this doesn't need to be closed".

The simplest solution here would be to make a single Do Not Close review decline a flag, just as a single Close review marks it helpful. This could even be enhanced to increase the number of responses required to decline a flag according to the number of flags on the post - just as we do for the Low Quality queue. Given the disparity in privilege level here, we would probably want to cap the number of "do not close" reviews required however, to prevent edge cases wherein a post would stick around in /review long past the point where it would normally be dequeued today.

A somewhat more messy solution would be to create a new flag status to reflect flags that aren't being actively reviewed by anyone. This might be useful if we opted to deprioritize flags that haven't yet attracted enough signal to warrant further attention.

Answer 3

I had a pretty big change of heart here when re-reading the blog post that first announced these flags:

One thing we realized is that the mod flag dialog ends up being training wheels for closers. That is, users who do not yet have the right to cast a close vote (earned at 3k reputation), but do have strong feelings that a given question does not belong based on our standard set of close reasons.

Four years later, this is how many new users learn to use the site's moderation tools - and as such, it's important that they don't learn bad habits. Accurate flagging is a critical part of maintaining the quality of these sites; careless or unnecessary flagging bogs down the process for everyone.

So with that in mind, flags are no longer allowed to linger. The solution isn't as elegant as what I once hoped for, but it should suffice:

1. Close flags age the same way that close votes age. That means there's a hard limit of 14 days after the last vote or flag that they can remain active. Both votes and flags can now be re-raised if they fall into this scenario, which I think provides a reasonable compromise between the ideal of automatically escalating ignored flags and the practicality of lots of these flags being not particularly urgent.

2. Triage review (currently Stack Overflow only) instantly disputes flags if the consensus is not "unsalvageable". This provides nearly-instant feedback without requiring a lengthy close review.