Updated: As the preferences expressed seem to endorse the change, I have asked the Community Managers (CMs) to carry out a one month test of reducing the number of close/re-open votes to 3 to see how this works. We don't yet know when this will happen as the CMs have an embarrassment of work on hand, but -- if we get prior warning -- we will let you know when the experiment starts.

Recently Stack Overflow reduced the number of votes needed to close/reopen questions from 5 to 3 after a successful experiment to see if it was worth it.

The results were that it did improve the rate at which bad questions were closed but didn't result in good questions being incorrectly closed.


Do we want to go down the same route here on Genealogy & Family History?

As background, 2019: a year in moderation reveals that of 96 questions closed last year, 95 required moderator intervention to close them! A change to the number of close votes needed would make it easier for the community to close questions that need closing quickly, and also to re-open questions that have been improved.

Please vote on the two answers provided (or add your own answer.)


Yes, we should lower the threshold -- it will improve the speed of closing and re-opening questions and better allow the community (rather than the moderators) to indicate which questions are acceptable and which aren't.

For the avoidance of doubt this is my preferred solution but I can't vote for my own answer.

  • 1
    This looks fair. It's fairly quiet here, and I suspect that few users vote regularly, so lowering the threshold is likely to help. I rather doubt we'll see any incorrect closures as a result.
    – AndyW
    Jan 9 '20 at 11:23

The results at Threshold experiment results: closing, editing and reopening all become more effective indicate "that [lowering from 5 to 3] did improve the rate at which bad questions were closed but didn't result in good questions being incorrectly closed", even for our largest site, so I am keen to see it implemented here too. However, even though it looks like 3-vote closing/re-opening may be spreading to other sites (see Testing three-vote close and reopen on 12 network sites) it looks like we are not amongst the sites to be trialled.

To me this is independent of who (moderator/community) cast the last close vote on questions last year being 95/1. That ratio is interesting, but I do not find it concerning. I only get concerned when questions are open that should be closed, or questions are closed that should be open. That only 2 questions were re-opened last year, both by moderators casting the last re-open vote, suggests that question closures here were rarely contentious. Even those 2 questions may have been edited before re-opening so they may not have represented a close/re-open "dispute" but questions closed correctly, fixed and then re-opened correctly.

@Shog9's advice to moderators is what I try to follow when deciding whether to close/re-open vote:

When you see a post you think should be closed, close it. When you see a post you think should be re-opened, re-open it. If you're not sure, don't do either.

  • 1
    The change wouldn't stop mods voting on the clear-cut questions, but it would allow us to give the community space to act on the less-clear ones and so signal their views.
    – user104
    Jan 9 '20 at 10:33
  • @ColeValleyGirl I think that more close/re-open votes before closure/re-opening is always better unless it delays the closing of an open question that should be closed, or delays re-opening of a closed question that should be open (perhaps because it has been edited).
    – PolyGeo Mod
    Jan 9 '20 at 10:41
  • I think our focus should be on accurate and quick closing/re-opening rather than who performs that task.
    – PolyGeo Mod
    Jan 9 '20 at 10:43
  • We can focus on more than one thing at a time -- accurate and quick closing/reopening and allowing the community to handle more of such are not mutually exclusive, if we can lower the threshold. There's at least one site that has a threshold of one, but I think that'd be taking things too far.
    – user104
    Jan 9 '20 at 10:46
  • Lowering the threshold is what this answer supports but not because of any concern about the 95:1 ratio. This answer supports it because, in the absence of a moderator being around to complete quick closure/re-opening, there is more likelihood that the question will already be closed/re-opened before a moderator sees a question needing that action.
    – PolyGeo Mod
    Jan 9 '20 at 10:51
  • To that point, do we know how many vote-eligible visitors we have at any given time? It may simply be that there aren't enough people (that are able/interested to vote) online before we moderator gets round to closing the question anyway. Jan 14 '20 at 13:33
  • @PeterKuehne There's no way of knowing who is online at any one time (deliberately so, so voting is anonymous). We do know that questions can accumulate a couple of close votes before moderators get to them, so that some questions that who up in the stats as closed by moderators are actually closed by a moderator as the third vote...
    – user104
    Jan 16 '20 at 8:44
  • That would imply there's at most a 2:1 ratio of users able/interested to vote to moderators, which to me implies either too many moderators or, more likely, not enough community engagement. Lowering the threshold could help in improving that, getting more people to vote and keep the questions/answers tidy, rather than relying on "the mods will get to it". Jan 16 '20 at 9:27
  • @PeterKuehne I think the moderator:community close ratio is something that finds its own balance depending on each community and the moderators elected/appointed to it. I don't mind whether the last close/re-open vote comes from a moderator or any other close/re-open voter. I only get concerned when questions are open that should be closed, or questions are closed that should be open.
    – PolyGeo Mod
    Jan 16 '20 at 10:54

No, we should leave the threshold as it is -- it shouldn't be easier to close and re-open questions.

You must log in to answer this question.

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