-11

This is not a duplicate of Give 2k users a 'don't put this in the reopen queue' checkbox. That request has been declined; this is a request to reconsider that decision.

There is a known issue of users editing questions that get closed recently after they get closed, for grammar. While the edits are themselves valid, they push the question into the reopen queue, and since they do not make the question reopenable, users (rightfully) disagree with reopening them.

The earlier declined request mentions wasted reviewer time and potential errant reopenings as arguments for implementing this, but I don't think that those are the main issues, since users are clearly happy reviewing them, and there aren't any errant reopenings as far as I've seen. The main issue, on the other hand, is that such edits deprive the post author (or anyone else) of the ability to have the post reopened as a result of editing. In many cases, the question may already be reopenable (in other words, the closure was invalid), but it's been shown that users fail to reopen such questions just because the edit is minor.

Considering that we don't let users know if their post has already been reviewed and rejected for reopening, if the user just blindly follows the guidance of editing their question in an attempt to have it reopened, if someone else has already made minor edits, the user will just be wasting time waiting for a potential reopening (which probably won't happen since it won't go into reopen review again). This can leave users dissatisfied and pretty stuck, forced to make a question on meta.

There have been proposals asking to make only author edits add questions to the reopen queue, but I disagree with those, since others can also be interested in seeing the post reopened. In fact, this was how the system was configured initially; it was later changed (for good reasons).

To address the original reasons for declining it, in the answer there:

we ignore [edits] where the editor has given some indication that [they don't] want the post reopened (voting to delete and flagging are pretty strong indicators here, so that's what we go with).

The edit privilege is at 2,000 rep, with lower-rep users able to suggest edits. On the other hand, it isn't possible to vote to delete unless the user has 20,000+ rep during the first two days after closure, or 10,000+ rep during the last three days of the "on hold" period (where questions can be added to the queue if they're edited). Also, the only flagging options for closed questions are for spam or abuse, or moderator intervention (in most cases, none of those are necessary). I think it's safe to say that a large number of users who perform such minor edits simply can't otherwise express they don't want to in the current system.

If you're making a lot of edits on posts that are closed - nominated for deletion - but which you explicitly do not want to see reopened... Then perhaps consider spending a bit less time rearranging deck chairs on the Titanic and instead putting that time and effort into other, more constructive activities.

I'd agree with this, but on several sites, we have one specific user who goes around and makes minor edits to posts, without regard to closure dates or the fact that they're going to push a certain post into the reopen queue. I've tried repeatedly notifying them in comments and in chat about this, but they have continued to make such edits. In my opinion, they seem to be clearly aware of the fact that their edits are pushing questions into the reopen queue, but are choosing to do so anyway, thus violating the quote above.

And even if we didn't have such a user, I'm sure that this may still continue to be an issue, given that we don't let users know that they will push the question into the reopen queue. In that case, it's quite possible that simply letting them know would make them reconsider and follow the quote, but the reality is that we do have a user who (I'm certain) will ignore it.

Also, to address a point in a comment which states that this should be dealt with on a user-specific basis since these users are violating the guidelines: the guideline itself is controversial, as there are other users that believe that "we shouldn't stop users from contributing edits [even if they're just turd-polishing]"; implementing this request would prevent this from being an issue and still respect the wishes of said users. We also don't make said guidelines clear anywhere.

In summary, can the original request please be reconsidered? I'd argue that the checkbox should be disabled by default so that users (especially those that use semi-automated grammar and spelling correction scripts such as the one used by that user) will have to explicitly indicate that they're aware of it before they can do so.

  • 1
    So, why are you not flagging such a user for mod attention or bringing it up in the site meta? This checkbox is a huge amount of work for the developers, a load on the server, and takes learning and action on behalf of every other editor. The solution already exists - teach editors when and why to edit, and if they refuse to follow agreed guidelines, escalate to blocking their edits entirely. – Nij Apr 18 at 1:15
  • @Nij Because it's controversial. A lot of users don't want this user to stop editing, because they believe that all edits are valid, even if it's just turd polishing. This request would respect the wishes of users who want this user to continue editing, while reducing the "damage" it does. Additionally, said "agreed" guidelines aren't made clear anywhere. – Sonic the Anonymous WizHog Apr 18 at 1:16
  • 1
    It would be very nice if I got advice on how I can improve this request to address potential concerns. – Sonic the Anonymous WizHog Apr 18 at 1:50
  • I agree this is a problem. I would however instead suggest to not automatically put edited questions in the reopen queue, and instead enable the “reopen” vote link that already exists for the editor, after an edit was made. This button would put the question in the review queue (only those with the privilege would be able to actually vote to reopen, of course). Each question can then only be put into the queue once. There could be a popup warning of this. – Cris Luengo Apr 18 at 1:56
  • @CrisLuengo The idea of having a "recommend reopen" flag in addition to a recommend closure flag has already been requested and declined. – Sonic the Anonymous WizHog Apr 18 at 1:58
  • I don’t think that is the same. I’m suggesting only the person that edited the question can choose to put it in the review queue. Instead of always putting it in, the person needs to opt in. There’s only one chance, just like now. It’s not a free-for all, which is why the linked request was declined. Thus: instead of opting out as you suggest, I think it would be better to opt in. In part because it would be easier to repurpose existing UI elements for that, and in part because I think too many people would not remember to opt out. – Cris Luengo Apr 18 at 2:23
  • @CrisLuengo Which is what exactly is being suggested here. This would allow users to opt in to having their edit add this to the reopen queue, by means of a checkbox on the editing form. It would be disabled by default, so it would be opt-in (and not opt-out, like if it were enabled by default). – Sonic the Anonymous WizHog Apr 18 at 3:29
  • Ah, sorry, I missed that. That's the right approach. It would give people that fix their question a fighting chance to get it reopened. – Cris Luengo Apr 18 at 5:41
  • I agree that a checkbox would probably be too much work for the developers. It might be much easier to check whether the OP edited and only then push into the reopen queue. Why should a 3rd party edit push a post into the reopen queue in the first place, considering what edits are for? And even on stacks where more substantive edits are allowed and common (I only know one), they are usually done by high-rep users who then make a reopen request on the stack's meta. – Anne Daunted Apr 18 at 15:46
  • @AnneDaunted As I said in the post, that was the case in the past, but it was later changed. – Sonic the Anonymous WizHog Apr 18 at 15:47
  • Yes, but it shouldn't have been changed and is a much easier solution to the problem. If 3rd parties want to see a question reopened, there are better ways to raise awareness, like discussing it on meta or voting to reopen. Abusing edits is not a good idea (and what to do if there's nothing to edit?), only when missing information is given in by the OP in comments can edits improve it that much and maybe if it was wrongfully declared a duplicate (in these cases, I got already closed questions reopened without problems). – Anne Daunted Apr 18 at 16:01
2

TL;DR:

  • There is very little evidence of cases where a (third-party) edit actually ends up stealing an OP's chance of getting their question reopened in the queue.
  • Your problem seems to be with automated edits, so no people are actually wasting their time 'rearranging deck chairs on the Titanic'.
  • Leaving the checkbox unchecked feels counterintuitive for new users while checking it by default doesn't solve the problem you have with the edits made by the script.

All in all, this makes me feel like this feature request should still be declined.


but it's been shown that users fail to reopen such questions just because the edit is minor.

The feature request you link to here is declined, and the answer says this is because people need to review the question as appears in the queue and not the edit. Getting people to do that is a totally unrelated issue and not a change of behavior that's likely to be achieved with a checkbox for putting questions in the reopen queue.

The main issue, on the other hand, is that such edits deprive the post author (or anyone else) of the ability to have the post reopened as a result of editing.

I think you brought this argument to the table before, in another declined feature request. From the answer there, Shog9 clearly says these cases rarely happen. There's a few numbers mentioned in the comments, they can be found in this post on MSO. Essentially, very few questions get two edits within 5 days. And of those that even get 1 edit, this edit is most often made by the OP. They aren't numbers about MSE or any other SE site, but I'm fairly confident together with the answer your declined feature request got, they're representative for a lot (if not all) sites.

and makes minor edits to posts using a script

So, this user that's so annoying to you is in fact not really 'spending time and effort rearranging the deck chairs on the Titanic', it's an automated script. The effort is made once to create the script, then it just ... runs. The analogy there seems pretty flawed. The script probably does more good than harm by fixing grammar, as edits to a post are sometimes even named as a way to let a user know we care about them and their post, and to get them to engage in further improvements of their question themselves.

but the reality is that we do have a user who (I'm certain) will ignore it.

We don't have a user who will ignore it. As you said, it's likely an automated script. You're right that script is likely to ignore this visual cue, though.

Both this and your previous argument seem to be in reaction to 1 script making edits correcting grammar, instead of a very large group of people editing questions for proper grammar. I know you really, really, really dislike that script, but this seems a lot like pushing for 'your way or the highway', where you either want to stop the script from editing on-hold questions (which failed as you couldn't get a hold of the author), or change the entire system just to fix one perceived problem (as I said above, numbers suggest it's not a problem at all) with that script. But it's probably not worth the time and effort to counter the effects of 1 script if there's not a huge problem with a huge group of people doing the same.

I'd argue that the checkbox should be disabled by default so that existing scripts won't push questions into the reopen queue

And how is a (new) user going to know they should check a checkbox to get their question reopened? You'd have to have more clutter, like a message/reminder/notification/prompt/explanation to check the box. If anything, I feel it should be enabled by default. I have no experience with UX design, but it feels more intuitive to place the burden of knowledge about closing, editing and reopening (and thus when a checkbox should/shouldn't be ticked) on more experienced users. This won't fix your problem with automated edits though.

  • 2
    "There is very little evidence of cases where a (third-party) edit actually ends up stealing an OP's chance of getting their question reopened in the queue." I've seen many cases of 3rd parties fixing a few typos in on hold posts and pushing them into the queue when they weren't ready for reopening yet. One of them, when informed of the problem, was pretty open about not caring. Usually, the OP didn't come back to fix it so no actual harm was done (i. e. they didn't steal the OP's chance), but it maybe different on other stacks. – Anne Daunted Apr 18 at 13:20
  • @AnneDaunted Exactly. The numbers I found on MSO (from last year) seem to support that too: very little harm is done by these edits, as there are so few posts that actually get that second edit. So it's probably not worth the effort of implementing something to fix those few edge cases, like other declined feature requests say too. If there's no big problem, all this likely won't be needed. – Tinkeringbell Apr 18 at 13:27
  • If you're in favor of declining this request, how would you favor implementing the following two requests: 1. letting users editing posts know of the fact that their edit will push the question into the reopen queue (while retaining the current behavior), and 2. letting post authors know if their question has already been reviewed for reopening with a "Leave Closed" outcome? – Sonic the Anonymous WizHog Apr 18 at 13:34
  • Ignore whether or not they're using a script. (By the way, when I was writing this question, I was referring to semi-automated scripts - where the script would make the edits client-side, but the user would still have to choose questions and still submit the edit manually. I know this, because the user in question has published the program they use for this, and it's semi-automated.) – Sonic the Anonymous WizHog Apr 21 at 12:26

You must log in to answer this question.

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