14
\$\begingroup\$

I often feel that questions linger too long before being closed, collecting answers making it difficult to edit a closed question. I would like EE.SE to take part in the recent site-wide "three-vote close" experiment. The main meta post can explain it better than I can:

Testing three-vote close and reopen on 13 network sites

I believe that closing a "questionable question" quickly will lead to better quality over all. Closing a question is not a punishment, nor is it the same as deleting it. A question is closed because it has problems that should be fixed before answers start to arrive. As soon as an answer is posted, editing the question is more complicated.

As far as I understand it, it will also be easier to reopen a closed question — something I also approve of.

\$\endgroup\$
18
  • \$\begingroup\$ Could we get our site moved to a three close vote? \$\endgroup\$
    – Voltage Spike Mod
    May 6, 2021 at 18:05
  • 3
    \$\begingroup\$ Worth a try I would say. But there could be concerns of autocracy. \$\endgroup\$
    – Mitu Raj
    May 6, 2021 at 20:02
  • \$\begingroup\$ @mituraj please explain \$\endgroup\$
    – Voltage Spike Mod
    May 7, 2021 at 3:05
  • 5
    \$\begingroup\$ I mean it would go less democratic, with a fewer set of people now controlling closing and re-opening. @VoltageSpike \$\endgroup\$
    – Mitu Raj
    May 7, 2021 at 3:42
  • 1
    \$\begingroup\$ Yes, both processes would be faster. There is not a good reason for a close vote que if it doesn't work fast. Bad questions get answers and no correction \$\endgroup\$
    – Voltage Spike Mod
    May 7, 2021 at 3:59
  • \$\begingroup\$ So, is this going to be implemented here soon? \$\endgroup\$
    – Mitu Raj
    May 7, 2021 at 11:40
  • \$\begingroup\$ As you noted in the post, we conducted a test on 13 network sites — please stay tuned for the results of that step and next steps on this. \$\endgroup\$
    – JNat StaffMod
    Jul 9, 2021 at 10:41
  • 4
    \$\begingroup\$ Can you tell us all how the close queue currently works? I never really understood it. Does a "leave open" vote cancel a close vote, or does the question close with the first 5 close votes? \$\endgroup\$ Jul 9, 2021 at 17:03
  • \$\begingroup\$ The latter. The theory is that if five high-rep users thing a question should be closed, that's enough initiative, I think. \$\endgroup\$
    – mmmm
    Jul 13, 2021 at 17:33
  • \$\begingroup\$ Are there site statistics about the number of questions that remain at 3 or 4 course votes for say 4 days or longer? Feels like the number of questions that hit three, but never five, close votes is so negligible that if makes little sense to worry \$\endgroup\$
    – mmmm
    Jul 13, 2021 at 17:35
  • 1
    \$\begingroup\$ Closing a "question", because it is a "bad" question (or "homework") is not "profitable" for anybody (it is a teacher that says that), especially for shy newbies, nor for this "site" because if too much "question" are "closed", I would ask my "question" "anywhere" else ... Is the "goal" of this site ... really "train", "help" and "learn" ... newbies? \$\endgroup\$
    – Antonio51
    Apr 26 at 8:54
  • 1
    \$\begingroup\$ After a "question" is closed, for the same reasons I made in a preceding comment, it would be "invaluable" that who's "close" ... should also edit the question to show "how" this question "should be" formulated ... Remember ... "train", "help" and "learn" ... newbies. \$\endgroup\$
    – Antonio51
    Apr 26 at 9:02
  • 3
    \$\begingroup\$ @Antonio51 Your comments are orthogonal to my feature-request, so I don't know what you hope to gain from writing it here. This website is not designed to teach newbies, it's designed to create a repository of quality answers to quality questions with a high signal-to-noise ratio. \$\endgroup\$
    – pipe
    Apr 26 at 10:37
  • 2
    \$\begingroup\$ @Antonio51 I think it's bad that a better explanation isn't given when a question is closed but, the way that explanations for closure might be read by a newbie may lead to him/her flagging the comment and that then could lead to the mods regarding the comment being what they call snarky. That, done enough times will lead to an innocent person being suspended. And that is why I don't leave comments that try to help the OP understand what was wrong with their post anymore. \$\endgroup\$
    – Andy aka
    Apr 29 at 17:29
  • 1
    \$\begingroup\$ @Andyaka Agree totally with your "strategy". Thanks. \$\endgroup\$
    – Antonio51
    Apr 29 at 18:35

4 Answers 4

9
\$\begingroup\$

This is live.

I've reviewed the request and your recent history and I can definitely see that this will likely benefit the site. Thanks for your patience while we got to the place we could make this change.

To determine whether to make this change, I used a few queries - firstly, I like to see what percentage of questions that get at least one flag or vote to close get handled - either closed or marked "leave open".

Graph showing percentage of questions with at least one flag/vote to close that get handled over the last 2.5 years. The numbers are somewhat unstable and percent handled is usually around 50% but can be as low as 35% and as high as 70%.

The first thing I like to note is that the graph vertical axis is 0-90%, so it can look a bit higher than it is. In general, y'all aren't usually able to get most of the questions handled. While this has changed over time, it seems that you're generally sitting around 50% handled, which means a lot of them are aging out of review. The end result of this is that the site can have a lot of content that should have been closed that sits around cluttering things up.

I'll also note that you have a pretty decent "leave open" result (~10% at times) - many sites don't seem to have such a high percentage (generally closer to 1-5%) - it's not bad but it could indicate there's some disagreement amongst voters about what's close-worthy. As such, it's possible that if more reviews were completed, those questions might actually be left open - it's difficult to know.

The other thing I check for is whether the moderators are doing an outsized volume of the close reviewing - this can indicate that they're doing more than they're necessarily expected to do. While there are many things mods need to handle, reviewing isn't one that we generally include in that. They're certainly welcome to participate but - because of their unilateral close privileges, it means that reopening questions can be significantly harder without moderator participation.

That said, it looks like, in general, the moderators aren't usually doing the bulk of the closures here, so it's not a concern. Over the last year, it looks like they're participating in about a quarter of the closures on average.

Graph of who is participating in close and reopens on the site. For the most part, the community is doing the bulk of the closures and the moderators are doing some (between 50-100) per month while the community is doing more (200-300). per month.

If you're interested in which votes are being cast by mods, you can see that below - I made it for a shorter period of time because it makes it a bit easier to see the lines.

Graph of which vote mods are casting. It's generally pretty evenly distributed between vote 1-5 with no consistent common vote being cast

As this shows, mods don't seem to be more likely to cast fifth votes over any other and, in some cases, have cast many first votes - which is understandable! All-in-all, it seems pretty even.

So, the end result is, we'd love to see the percentage of reviews handled increase here and it looks like reducing the votes needed to close and reopen from 5 to 3 will help with that. Because the tests on other sites went generally well, y'all can consider this permanent - at least until you ask us to change it again. I'll check back in a few months but, unless there's something huge to report, I probably won't update things. If you have any questions or want to see how things are going, feel free to let me know.

\$\endgroup\$
12
  • 1
    \$\begingroup\$ Thanks for the update, Catija :-) Just one question regarding mod closures, you said: "looks like [the moderators are] participating in about a quarter of the closures on average." While us mods do unilaterally close some questions, for sure, do those stats for "moderator closed" even include where we were vote number 5 i.e. we had no more effect (despite being a mod vote) than any other user with close vote privileges? || I will be interested to see how/if the stats change, after enough time to see results. (I recently found some long-term users didn't know they could vote to re-open!) \$\endgroup\$
    – SamGibson Mod
    Apr 22 at 17:04
  • \$\begingroup\$ It looks like it does. I did a breakdown of what vote number mods are casting as well and have added it to the answer. \$\endgroup\$
    – Catija StaffMod
    Apr 22 at 17:14
  • \$\begingroup\$ Thanks Catija - I'm glad to see we weren't doing mainly first closure votes (and I didn't feel that were - except perhaps on unattempted homework). Many of the times I have closed posts have been in response to flags - some Q have had a close vote or two before mine (when those who flagged can also vote to close) and some A have had NAA votes before mine (again, from people who can VTC), but some flags are from those who cannot yet vote to close themselves, so mine would be the first actual vote. So, all interesting - thanks again. \$\endgroup\$
    – SamGibson Mod
    Apr 22 at 17:31
  • 6
    \$\begingroup\$ I think the unusal high "leave open" result is caused by a strange culture on this particular site, where veteran users who should really know better like to aimlessly invent their own meaning to established close reasons, from day to day, picking very strange close reasons out of the blue or close voting things that shouldn't be closed, just because they don't like the question. I would say that this is a recurring culture problem here. \$\endgroup\$
    – Lundin
    Apr 28 at 8:56
  • 5
    \$\begingroup\$ I agree with @Lundin If I decide to review the close votes, I probably place more than 10% "Leave open" votes. That is the main concern I have with the 3 votes to close. Some random toxic close votes can shut down legit questions that are then forgotten because newbies get a feeling of being not welcome and tend not to edit their questions. While some standard can expected of askers, I guess one cannot expect the same question quality of Rep 1 users than of more experienced users. \$\endgroup\$
    – tobalt
    May 1 at 15:06
  • 4
    \$\begingroup\$ I consider that the percentage of questions reopened is a disgrace. It is frequently stated when this subject is discussed that users may have their questions reopened if they improve them, BUT tis almost never happens (as can be seen) and new users are frequently puzzled or discouraged by the often headlong rush to close a question which could be saved if more people were willing to help. People have often enough made comments about people not having responded to comments - but this may be less than 12 hours after the comments are made. On a site where users are located ... \$\endgroup\$
    – Russell McMahon Mod
    May 10 at 12:33
  • 4
    \$\begingroup\$ ... worldwide and at least some need to take time to work, sleep and get a life, closing a question in under a few days, when it nearly certain that it will never be reopened, is (as above :-) ) disgraceful. || I agree that closing a question which has answers is problematic - but it certainly does not stop many questions being closed with one or sometime 3 or more good answers that show that the question has been well understood. \$\endgroup\$
    – Russell McMahon Mod
    May 10 at 12:36
  • 2
    \$\begingroup\$ Closing and deleting should only be done if the question is illegal. The mythical goal of "good quality" questions is just that, mythical. A visitor does not want a perfect question, the search engines will find the appropriate question. The end user wants to see that somewhere on earth someone has the same problem they have and hopes for clues towards an answer, exact answer is a bonus. A vote amongst visitors would not find support for close or delete, they are your customer base, not grumpy veteran editors who can ignore bad questions with no public loss of face, why pander to them? \$\endgroup\$
    – KalleMP
    May 10 at 19:58
  • 1
    \$\begingroup\$ @RussellMcMahon I agree with you and it ties in with a question I aske in February electronics.meta.stackexchange.com/questions/9614/… And promptly got 3 down votes for my pains! I think that there are far too many down votes and closure votes and too few attempts to help the OPs, particularly the newbies. It's of little use having a nice cosy club of EEs with reps greater than 3000 who discourage new members for no good reason \$\endgroup\$ May 11 at 13:43
  • 1
    \$\begingroup\$ @RussellMcMahon agree completely, and there is also the problem that sometimes the closed question is already clear and does not need any change. It's just that the people closing it did not seem to have understood it. \$\endgroup\$ May 11 at 17:05
  • \$\begingroup\$ @PeterJennings You keep confusing "new members" with "people who write bad questions". A lot of great questions are written by new members, questions which are promptly upvoted and answered. It sounds like you want to keep the users who can't yet write a good question - a skill better taught somewhere else and then put to good use on this website. \$\endgroup\$
    – pipe
    May 14 at 4:39
  • \$\begingroup\$ @pipe "a skill better taught somewhere else" Where better than right here? Who knows better than us what we want to see in a question? Help comes in several forms. But let's not get into an argument in these comments. Just say that we hold different opinions as to the purpose of this site. Looking at the latest 100 questions, the vast majority are from OPs with a rep of 1. a few with a rep of around 500, one, repeated 3 times with variations and a rep of 1200 and a single question from someone with a rep of over 47K. \$\endgroup\$ May 14 at 9:21
15
\$\begingroup\$

I support moving to a three close vote (if SE will allow it). The close review que has been hovering from 100-200 questions that need to be reviewed for quite sometime (at least 6 months) which means we have too many question that need to be reviewing or not enough people that review. Moving to three votes would make reviewing much more effective. I believe it is more important that questions get reviewed so that these actions can happen quickly.

There are a few other comments that I have:

If you haven't read this now would be a good time: https://electronics.stackexchange.com/help/closed-questions

If we do move to a three close vote I have a reminder for those who do review.

If the question can be fixed, then edit it and make it presentable.

Help new users understand the process, many people think that closing is bad and take offense. Its good to say something on the lines of: "We need to see X and Y improvements from this question, edit it and reopen it" and help them understand the process (obviously this does not apply to questions that can't be made on topic because of their subject matter, not everyone will take the time to fix their question).

If we do move to a three close vote we will also need to step up helping people understand the process (and not being condescending) because the closing process will happen much more rapidly. Also helping new users understand the voting system is a good thing.

In addition I believe users need 3000 rep to get their question reopened (https://meta.stackexchange.com/help/reopen-questions) so we would also need people helping to get questions reopened (or flagged for reopeneing if you don't have the rep) after the question has been fixed and help people understand this process.

(EDIT: now that I think about it edits will move the question to the reopen que)

Having more good questions benefits everyone, those who want to gain rep and those who want their question answered and those who visit the site for years to come to look at the question.

\$\endgroup\$
5
  • 2
    \$\begingroup\$ This system has been live on SO for quite some time now and it works just fine. The main benefit being that crap questions get closed in a much more timely fashion. \$\endgroup\$
    – Lundin
    May 6, 2021 at 14:18
  • 1
    \$\begingroup\$ I think " I believe users need 3000 rep to get their question reopened" is too harsh. Surely if the OP edits their question sufficiently, no matter what their rep, it should stand a chance of being reopened. If you just close a newbie's question with no chance of reopening it then you will put them off, possibly for good and that's not a good thing. It takes a long time to reach 3000, particularly if you are a learner rather than an experienced contributor. Few if any newbies will make it if they are treated like that. \$\endgroup\$ May 11 at 12:51
  • \$\begingroup\$ In the case of closing, all the user needs to do is edit the question. At that point it will be voted on. Usually a mod will open them back up. The flagging system can also be used to get the attention of a mod for reopening. Questions will not be reopened if they are not on topic. The whole point of closure is to get people to edit their questions. Closing a question should not be considered a negative thing, it is part of the process \$\endgroup\$
    – Voltage Spike Mod
    May 11 at 13:33
  • \$\begingroup\$ @Lundin - there are a whole lot more people with >3k rep on SO than here. \$\endgroup\$
    – TLW
    May 15 at 4:08
  • \$\begingroup\$ @VoltageSpike - this assumes that questions are never closed in error. Mods are good; no-one is perfect. \$\endgroup\$
    – TLW
    May 15 at 4:09
8
\$\begingroup\$

I agree with OP, I think OP is the best. Let's do it!

But seriously, this answer is here so that people who agree can upvote, and those who disagree can downvote.

\$\endgroup\$
1
  • \$\begingroup\$ I need this to stay so the CM's can see it \$\endgroup\$
    – Voltage Spike Mod
    May 11, 2021 at 16:09
1
\$\begingroup\$

I don't know whether a 3-vote close is a good thing or not, but don't see a real upside to participating in the test. I'd rather wait and see how the test works out and any additional tweaks are made until adopting the policy.

\$\endgroup\$

You must log in to answer this question.

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