15

UPDATE: While the testing period has ended, this change has been well-received by the community so we have opted to not reset the votes needed to close/reopen to 5 while we look at data.

Please let me know if you have any concerns about this change in the planned process or if you feel the number of votes should be changed back at least temporarily. I'll be back in a few weeks with the results of the project.

Part of understanding the impact of three-vote closure is to understand what impact it's having on y'all. I'd like to ask anyone who feels up to it to write an answer to this question and to tell us about your experience over the past few weeks.

The sorts of things I'm interested in knowing (feel free to address all or none or add your own) -

  • Did this help?
  • Did it hurt?
  • Did it make you more interested in close/reopen reviewing? Less?
    • Are there things other than 3-vote close impacting your interest in reviewing?
  • If you can't vote but do flag, did this make you more/less likely to flag?
  • Would you prefer to go back to 5?
  • Should it be permanent?
  • What would you like me to look into when it comes to analyzing the data from the test?
  • Were there any side-effects (good or bad) that you want me to be aware of?

These are merely prompts, so feel free to add anything I'm missing that you think is worth saying. This is focused on how you're feeling about it and what data you want me to look at, so don't feel like you need to analyze the data over the last 30 days - though you're welcome to if that's how you figure out what you feel.

If you have any outstanding questions, also feel welcome to ask them here.

The test will run until about Monday the 21st of June, at which point we'll be resetting the votes to close up to 5 while I review the responses here and dig into the data that we've been collecting.

6
  • 2
    I already wrote what I would like to contribute here.
    – Doc Brown
    Jun 4, 2021 at 21:27
  • 2
    10K tools page shows 9 reopens by 3K votes in 30 days that passed since experiment start. Can you help with similar stats for 2-3 months before experiment? I would want to compare and estimate the impact of the change but 10K tools can only show data for last 30 days
    – gnat
    Jun 6, 2021 at 14:01
  • 1
    just out of curiosity, does this include the "gold badge holder autoclose" feature, as on SO?
    – Nate T
    Jun 12, 2021 at 21:06
  • 2
    @JonSkeet meaning that gold badge users can unilaterally duplicate close a question? If so, that feature is network-wide but tends to be more common on SO because it's easier to attain a gold tag badge there.
    – Catija
    Jun 12, 2021 at 21:13
  • 1
    Good to know, thanks. didn't realize that it was network-wide.
    – Nate T
    Jun 12, 2021 at 21:20
  • 1
    Today, I realized that here on SE.Meta, still 5 close/reopen votes are necessary. Is this intentionally? I mean, meta participation is a lot lower than on the main site, which should make a lower threshold for close/reopen votes even more reasonable.
    – Doc Brown
    Aug 6, 2021 at 6:02

1 Answer 1

Reset to default
14

I absolutely love it. Please consider leaving it in place.

As a moderator, there have been significantly fewer flags asking for a moderator vote to close a question. Before, questions would get to 3 or 4 close votes and not get that 5th vote. They may even start to attract low-quality answers. This change lets the community handle them much faster and get them closed until problems can be resolved.

However, this does require the community to leave comments on questions explaining, specifically, what needs to happen in order to make the question a good fit (assuming it can be made into a good fit). Drive-by down votes and close votes don't help anyone. I'm not sure how much this is an issue network-wide, but being able to help people fix their question is key. Understanding the number of people who down vote and/or close vote without comment and why they are doing so would be a welcome next step.

I don't often review as a moderator, since all of my votes are binding. I do sometimes, but I'd prefer if I had the option to toggle my binding vote or if my vote wasn't binding in the context of a review queue. I'll leave the discussion of the review queues to others.

8
  • 1
    I notice that about half of the questions that are closed are closed with your community-specific close reasons - have y'all considered updating those close reasons to make use of the new fields we've made available? This allows you to guide close voters, the asker, viewers, and those with enough rep to help improve the question with specific guidance on how to help questions that are closed with that reason. Here's a guide I wrote up on MSE - meta.stackexchange.com/a/362584/284336
    – Catija
    Jun 4, 2021 at 19:37
  • 2
    I'm not saying this is a perfect solution but it might help give more specific guidance to askers as well as informing users with the ability to close who see the question about how they can help guide the asker or even directly edit the question to improve it. We can also help you customize the first time asker help info box and the right side how to ask box for all users asking a question. :)
    – Catija
    Jun 4, 2021 at 19:38
  • 2
    @Catija That is on my list to look at and understand. I was busy when it came out and then forgot about it.
    – Thomas Owens Mod
    Jun 4, 2021 at 20:06
  • 4
    @Catija Although for questions that are closed with the custom reasons, most of those are not salvageable. They can't be made to be on-topic here. Even if they were made better, the best choice would be to migrate them.
    – Thomas Owens Mod
    Jun 4, 2021 at 20:08
  • 1
    @ThomasOwens I would add that according to stats many (most?) questions would be blocked at destination even if improved to migrate-worthy state. Guess even Meta SE would have better success in migrations if they have chosen to follow this approach
    – gnat
    Jun 6, 2021 at 12:07
  • 2
    @gnat I'd have to dig into it. I suspect you're right for coding/debugging questions that belong on Stack Overflow. I'm less convinced about career questions (for The Workplace) and legal (for Law or Open Source). I'd probably never migrate to one of the recommendation sites because of their strict rules on what is required that I don't keep track of, so those are off the table.
    – Thomas Owens Mod
    Jun 6, 2021 at 14:24
  • 1
    @Catija I think that if you could customize first time asker info for those who recently hit a block at Stack Overflow, this would be quite useful (especially if moderators could somehow check how askers handled this info). Such customisation would be useful at Meta SE as well, but maybe to a lesser extent, see references to stats in my previous comment
    – gnat
    Jun 6, 2021 at 14:32
  • 2
    @ThomasOwens per my observations career questions are typically very poor fit for migration. This is probably because Workplace has very good visibility already and most of those willing to properly ask their topics are aware of the correct destination site. Legal and licensing questions seem to be different matter, with better chances for migration
    – gnat
    Jun 6, 2021 at 14:37

You must log in to answer this question.

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