-7

I've come across different questions at different networks that get closed, the reopened, then closed again...

It's often about questions that represent an edge-case for what's on/off-topic on the corresponding site, or "maybe-opinion-based" questions.

What often happens is that a few people don't think the question is fine and vote to close it while people who think it is can do nothing. Then, when enough "close" votes are reached, the opposite happens (reopen votes until the threshold is met)

I don't know if you think this is fine but for me it's chaotic. A question is either on topic or not, but it can't be on-topic on Tuesday and off-topic by next Friday only to be on-topic again in the weekend. I can think of a few alternatives.

  • Closing a question becomes definitive - This appears to me as way too harsh and losing value for the sites (maybe a good edit could save the day).
  • Creating a "do-not-close" vote - Questions wouldn't get closed until the "close vs do-not-close count" reaches a certain point (5 more votes for closing, or maybe at least 3 and 66% of total, or any other criterion). Then, questions shall only be reopened by a moderator after an approppriate edit is made.
  • Put the "closers" on trial (i.e. make us reputation-responsible for what we decide to flag and close). This means that sound close-requests should be awarded reputation while unsound ones should imply a loss in reputation

NOTE: This is my recent experience:

I present you a closed question with 3 reopen votes while this post is being made. This question will probably be closed and reopened as well. Meanwhile, this question has gone through the entire process. I was writing an answer when it suddenly got closed for a few days. Now it's reopened but it's really painful to take the time to rewrite the entire thing

EDIT: This question was closed as duplicate, but the linked question is more than 10 years old. A large majority there agreed with the second proposal but nothing has been done since! I would like to understand the reasons why. By the way, if this question gets reopened, it would be the most-meta question ever!

21
  • In the close review queue we do have a leave open vote.
    – rene Mod
    Commented Aug 26, 2019 at 15:37
  • @rene I have never seen any such queue. I am talking about votes made on the posts themselves
    – David
    Commented Aug 26, 2019 at 15:38
  • I've only seen that queue ...
    – rene Mod
    Commented Aug 26, 2019 at 15:39
  • So not playing is an incentive ...
    – rene Mod
    Commented Aug 26, 2019 at 15:46
  • The example question on Chess is a bit special. That got closed by one mod and got kicked from the re-open queue twice by another mod. Not sure what the do-not-close and reputation/'trial" stuff would have helped there
    – rene Mod
    Commented Aug 26, 2019 at 15:49
  • 2
    @SonictheAnonymousHedgehog So... It seems everyone wants it. What are we waiting for?
    – David
    Commented Aug 26, 2019 at 15:51
  • 1
    @David It already exists. We do have a review queue where people can disagree with close votes. Commented Aug 26, 2019 at 15:52
  • 2
    The question hasn't been closed as a duplicate yet; the banner you're seeing is only shown to you since someone voted to close as a duplicate. Also, if you read the top answer to that post and its tagging as completed, you'll see that this already exists. Commented Aug 26, 2019 at 15:54
  • 1
    @SonictheAnonymousHedgehog but leave open votes only exist on review. If you open the close dialog directly on the post you can't vote to leave open. You would have to wait for it to be closed, before your disagreement can be backed with a (re-open) vote. And that is bascially what the OP asks here and that doesn't exist so the dupe doesn't apply if your argument stays as it is. There might be other reasons why it is a good dupe nevertheless but not under your current argument.
    – rene Mod
    Commented Aug 26, 2019 at 16:05
  • 2
    @rene This question doesn't really add any new arguments beyond what is in that question, and it doesn't show any research into how the current system is implemented. Commented Aug 26, 2019 at 16:07
  • 1
    @SonictheAnonymousHedgehog that is at least a better argument for the dupe, thanks. Lack of research has never been a close reason. That doesn't make the dupe any better
    – rene Mod
    Commented Aug 26, 2019 at 16:08
  • 2
    @David: "but nothing has been done since! I would like to understand the reasons why." Shog9's answer explains why. You may not agree with his reasoning, but that's still the reasoning. And personally, I find it adequate. Commented Aug 26, 2019 at 18:14
  • 3
    Well, considering your tone, and the fact that your suggestion is not an exact match for the answer, it looks pretty reasonable to have such a disparity. For instance, I see nothing in the answer related to Put the "closers" on trial (i.e. make us reputation-responsible for what we decide to flag and close). That's just punishing curators for trying to keep the site clean; easiest way to ensure nobody ever does any, and now you're left with a site where you can't find anything, because it's full of junk.
    – fbueckert
    Commented Aug 27, 2019 at 13:42
  • 2
    Well, reputation is the level of trust the system has in your ability to contribute to it. With the gamification aspect, people care a whole lot about imaginary internet points. The system signalling that your attempt to help clean it up will lose some of those points means less people will do it; it already requires 1 rep to downvote answers, so that happens less often than it should. Encourage behaviour you want to happen, discourage that which you don't. I want curators to be encouraged to help clean it up, even if they get it wrong sometimes.
    – fbueckert
    Commented Aug 27, 2019 at 16:13
  • 1
    Do you honestly think that what Stack Exchange needs is more downvotes and closed questions? Yes. Further, punishing people just ensures they don't do anything. Do you fire your janitor the instant he misses a mess? Seems like a great way to ensure you never have a janitor.
    – fbueckert
    Commented Aug 27, 2019 at 16:18

1 Answer 1

11

You're showing your bias. Why isn't this a request for a "do not re-open" vote?

If there is genuine disagreement, these "just keep it open" or "just keep it closed" solutions always expose that bias… not the actual voting.

"Close wars" are a bit of a misnomer. If there is genuine disagreement, folks have to wait in line to vote the other way in groups of five, back and forth, until you run out of interested parties. Even if there are 10,000 people wanting to keep it open and 20 against, you still have to run through four iterations for the flip-flop to end; hence the yo yo effect.

I wrote up a solution eons ago, and it went something like this:

A Solution to Abate (most) Close Wars

The basic idea to make make sure everyone can vote to "close" or "keep open" so the majority/momentum will quickly drive the decision in one direction or the other without forcing that whipsaw effect.

When someone votes to close, change the prompt to read something like this:

close(+1)? yes or no

or

close(+1) or keep open

If someone votes to close, you increment that counter.
If someone votes to keep open, you decrement that counter.

There are ways to make that prompt clearer, but:

If the total vote hit 5, the post is closed (as it is today).
If the total vote hits -5, the prompt goes back to normal.

If there is a true lack of consensus, the vote may oscillate between +5 and -5 for a while, but the majority of the time, it is more likely that a general consensus will quickly be reached without forcing the question through that close-reopen-close-reopen cycle.

Of course, if a question is closed (incorrectly), you can reverse that process in the other direction.

Re-open(+1)? yes or no

or

Re-open(+1) or keep closed

A bonus is that the system would no longer have to worry about aging votes (where close-votes eventually expire after a certain period of time). With the +/- votes shown, someone is free to counteract those random close-votes which tend to accumulate over time.

7
  • 1
    Would this mechanism also do away with the close and re-open queues? Or would you envision a review queue is still useful to reach the status quo sooner?
    – rene Mod
    Commented Aug 26, 2019 at 17:15
  • Not sure. The original design of the review queue was to bring attention to situations that might otherwise not be seen, so not entirely sure how this would affect that design specifically. Commented Aug 26, 2019 at 17:35
  • 1
    The problem with this solution is that it creates other problems. Here's the thing: a lot of people (particularly on SO) would vote to keep poor questions open. But they don't have the opportunity to do so, and by the time the last close vote hits such a question, they're not looking at it anymore, because they're sifting through the newer questions. As such, bad questions are properly closed. If these people had the power to preemptively stop those who are interested in quality from closing questions, you can basically kiss quality on SO goodbye. Commented Aug 26, 2019 at 18:17
  • The only way this would be functional is if "keep open" powers were only granted to 20K rep users; 2K rep is just too easy to scrape together. Commented Aug 26, 2019 at 18:18
  • 1
    @NicolBolas I think the Stack Exchange community has an obsession with policing "bad" questions. I understand that off-topic questions or plain spam should be closed, of course, but hose cases should be a rare exception, not something like 1 out of every 5 questions
    – David
    Commented Aug 27, 2019 at 7:26
  • This reminds me of voting on "tag synonym". Commented Aug 27, 2019 at 12:06
  • 2
    @David: "but hose cases should be a rare exception, not something like 1 out of every 5 questions" I will gladly stop closing bad questions when people stop asking bad questions. Changing our standards so that "bad questions" are rare is not the way to make the site better. Commented Aug 27, 2019 at 13:21

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