18

I know this is how things should work now, and maybe not that much of a priority, but I wonder if there is an easy solution to this.

On a question (now deleted by its author), I voted to close as 'Unclear what you are asking'. Second, a gold tag badge holder voted to close as duplicate and instantly reopened again (obviously because he saw the duplicate didn't match the question):

In the process, all existing close votes were removed (in this case only mine, but potentially four).

What are your thoughts on this? Is this intended, and should this be kept this way, or should we preserve votes when a single user closes and reopens (like retracting does now)?

As clarification: this is NOT a feature request, it is an analysis of a situation that causes me to question what happened and I wonder if and, if so, what to do about it.

|improve this question
  • 5
    Indeed, the power to clear existing close votes is something that is not part of the dupehammer as far as I can tell, so better be restricted. – Shadow9 Nov 19 '14 at 15:56
  • 2
    Or it is a hidden feature. – Patrick Hofman Nov 19 '14 at 15:57
  • 3
    Better yet, make a hammer-close + hammer-reopen by the same user an auto-flag and have it bear dire consequences – John Dvorak Nov 19 '14 at 16:59
  • cv'd btw. Meta effect go! – John Dvorak Nov 19 '14 at 17:00
  • 2
    For posterity, do you think you could grab a screen shot of the revision history relating to the close vote section so that people without 10k on SO can see the issue? – user213963 Nov 19 '14 at 18:38
  • Done, @MichaelT. – Shadow9 Nov 19 '14 at 19:34
  • I think @Jan's idea is a much better alternative. I doubt this happens very often, and when it does, the best option is to warn the user not to do that. Modifying the system to preserve cleared close votes (basically unclearing them) is a lot of effort for not much of a result. – animuson♦ Dec 16 '14 at 3:28
  • @animuson: A better aternative than what? This isn't a feature request, just an analysis. I understand it takes some effort to code a fix for this, but I wonder if it is necessary in the first place. – Patrick Hofman Dec 16 '14 at 9:55
  • @Patrick A better alternative than preserving the votes, which you do mention in the question. – animuson♦ Dec 16 '14 at 13:32
  • @animuson: Okay. That maybe is a big maybe... Not sure if that is my own opinion... – Patrick Hofman Dec 16 '14 at 13:34
  • 1
    This happened to me too (see this meta.math question). After reading this thread, I still think this behaviour is undesirable. The discussion here focuses too much on deliberate misuse (unlikely from gold badge holders) and too little on honest mistakes that destroy useful information (as happened in Patrick's and my case). – joriki Jul 4 '15 at 11:57
4
+50

My thoughts about it:

Is this intentional?

Yes, kind of. The dupehammer gives the user power of a moderator, who can use this to clear pending close votes. Uncommon, but not unheard of.

It is a big problem?

Not really, for the same reasons described here, in short because the user is already trustworthy plus those actions can be easily seen in the Revisions, and Reopening a question bumps it up the Active Questions list, making it easy to catch.

Can it be misused?

Yes, a user who was never elected as moderator can cancel decision made by other four high rep users with a single vote. For example four high rep users decide a question is off topic because [reasons here], but the dupehammer bearer disagrees. Instead of waiting until it gets the last vote and casting ordinary reopen vote, he/she use the power of the hammer to simply clear those votes.

How to prevent this from being misused?

If a question got more than two non-duplicate close votes, duplicate close vote will not be binding even if coming from a gold badge user.

|improve this answer
6

This is no different than a mod-closure. If the question gets closed all pending close votes are consumed. A tag-badge closer is really just a highly-localized moderator in this case (duplicates only, in specific tags only). For that matter, it's the same as a community closure -- but with the reopen requiring more people, which seems to be part of your point here.

In all closures, reasons that aren't the one that ultimately "won" get lost along the way along with their votes. If two people vote "too broad" and three vote "unclear", "unclear" wins and the "too broad" reason drops out. The votes contributed to the closure, but not for that reason.

Yes it's frustrating -- once you've voted to close a question you can never, ever vote to close it again, no matter what happens to it in the meantime, so if you voted to close and it got reopened (perhaps with edits you deem insufficient to fix the problem), there's nothing you can do directly. This is true no matter how the question got closed or what specific votes were cast.

Perhaps what you are looking for is an "oops" button -- I'm a mod or badge-holder and just unilaterally closed that and then realized I shouldn't have. This would mean retracting a close vote that has already taken effect. This would add complexity to the interface and confuse some people, so I doubt it would be considered without more evidence that this is a problem, but that's the approach I would take if I wanted to pursue a change here.

Alternatively, if the problem is that your vote is locked once consumed, attack that aspect (applying it to all closures). I expect that this has come up on MSE before.

Finally, remember that if you can't recast your own close vote, you still have meta and chat as tools to encourage others to vote.

|improve this answer
  • 1
    Disagree with the description as "a highly-localized moderator". Gold badge holders are trusted to identify duplicates, and that's it. They are not trusted any more than other 3K users when it comes to the closure of questions as unclear, too broad, or any reason other than a duplicate. Therefore, they should not be able to unilaterally wipe out closevotes of those kinds. – user259867 Dec 21 '14 at 23:35
  • @Behaviour I meant with respect to closure -- in this one specific (localized) way only, badge holders act like moderators. So any concern about badge-holder closes should probably be treated a little more broadly, at least to mod closes, right? Ok, I think I see what you're saying about other vote types, but unfortunately that happens with any closure -- if a question comes in with one off-topic, one unclear, and three too-broad, it'll be closed as too broad and the other voters' reasons are lost. – Monica Cellio Dec 21 '14 at 23:40
  • The reasons may be not displayed; the closevotes themselves are not lost; they take effect in your scenario. Which is rather different from them being simply cancelled. – user259867 Dec 21 '14 at 23:49
  • The votes take effect with a badge-close, too. Look again at the screen shot in the question; all votes are factored in, as they are for any closure. It's just that in this case SE doesn't wait for 5 votes. – Monica Cellio Dec 21 '14 at 23:51
4

I am going to throw an answer in myself, and see how it is received.

Is this intentional?

No. It isn't.

It is a big problem?

Currently not that much I think. There are not that many gold tag badge holders, and those are considered to be trustworthy (they have a considerable reputation, which gives them quite some power to vote to delete, undelete, etc.)

Can it be misused?

Yes. As pointed out, it can be an easy and effective mechanism to buy yourself some time before you answer, receive quite some upvotes before it is reclosed again. Have I seen this happen? No. But it is possible.

How to prevent this from being misused?

The change to keep the other votes when a gold tag badge holder retracts might be too much for this issue. The solution from Jan Dvorak as comment seems to be the easiest and more controllable solution:

make a hammer-close + hammer-reopen by the same user an auto-flag and have it bear dire consequences.

|improve this answer
  • 3
    make a hammer-close + hammer-reopen by the same user an auto-flag and have it bear dire consequences. maybe with a time frame. No need to flag if the user comes back 4 days later and realizes he closed it wrong.. right? – ɥʇǝS Dec 21 '14 at 21:43
-6

I don't understand why you'd want that. When a post is closed and reopened, the original votes to close have taken effect, they aren't reapplied.

Scenario 1:

  1. Q1 is closed as a duplicate of Q2 by users A, B, C, D and E.
  2. Q2 is reopened single-handedly by gold badge holder G.

Scenario 2:

  1. Q1 is closed as a duplicate of Q2 by users A, B and G.
  2. Q2 is reopened single-handedly by gold badge holder G.

In scenario 1, the gold badge power is used minimally: to reopen a question that was closed as a duplicate. In scenario 2, the gold badge power is used twice. In scenario 1, users A, B, C, D and E cannot vote to close that question again (but G still can). In scenario 2, users C, D, E and G are still free to vote to close the question, only A and B can no longer to it. So in scenario 2, G's reopen action leaves the question more closeable than in scenario 1.

I don't see how it could make sense to add more close votes on the post in scenario 2 than in scenario 1, when altogether fewer people voted to close.

|improve this answer
  • 3
    In scenario 2 though you need seven close votes to get the question closed. This can be significant on sites where getting five close votes can be a challenge at times. – user213963 Nov 20 '14 at 1:30
  • @MichaelT And in scenario 1 you need 10. Your point? – Gilles 'SO- stop being evil' Nov 20 '14 at 1:32
  • 6
    In scenario 1, it may not have been closed as a dup (and thus no dup hammer reopening). The problem is that by closing as a dup (no matter what other close votes where) and then reopening the question the previous votes are discarded. This is only a problem with dups and the dup hammer because the dup hammer trumps other votes. Consider 4x unclear, and then a dup hammer and reopen. If one more unclear was cast, the dup hammer wouldn't have been able to reopen and the question could continue its process as being closed as unclear. – user213963 Nov 20 '14 at 1:35
  • @MichaelT And the problem with that is? You keep saying that the outcome is different, but why is a different outcome more desirable? (Anyway, it's usually better to close a question as a duplicate than for other reasons — at least that provides some information which may be relevant, if not to the asker, then to future visitors who interpret the question in the same way as the close voter) – Gilles 'SO- stop being evil' Nov 20 '14 at 1:48
  • 2
    @Gilles: This wasn't a feature request. It is just an analysis of the situation and the question 'should this happen'? When a non-gold tag badge holder votes as duplicate and retracts, all other votes remain. I wonder if that should count for gold tag badge holders too. – Patrick Hofman Nov 20 '14 at 8:34
  • @PatrickHofman Voting to close and then voting to reopen is not the same thing as retracting a close vote. If you're the last close vote (which is always the case for a gold-tag-badge-powered duplicate close), you can't retract your close vote. The behavior here is the normal design, it does the right thing in the scenarios outlined here, I don't understand why you want to make an exception. – Gilles 'SO- stop being evil' Nov 20 '14 at 9:17
  • 1
    @Gilles: I understand it is 'by design' now, and I am not sure if we should change that. I can be an issue when the GTB holder mistakenly closes and reopens. But as said, not sure if it is an issue at all. – Patrick Hofman Nov 20 '14 at 9:19

You must log in to answer this question.

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