11
$\begingroup$

The question Assessing effectiveness of (epsilon, delta) definitions has been recently reopened. In the history, I see that it has been "left closed in review" two times in the past 10 hours, and then reopened at the third attempt.

I imagine someone has nominated this post for reopening three times in a very short timespan; is this considered acceptable? I suppose that any controversial post will eventually be reopened if it gets checked by the right five people, but nominating questions for reopening multiple times seems like a way to waste the moderators' time and push against the consensus.

$\endgroup$
38
  • 5
    $\begingroup$ According to the help center: "Questions can go through multiple close and reopen cycles, but each individual user may cast at most one close vote and one reopen vote per question." So I wonder whether a bug manifest here or whether each review was actually initiated by somebody else. $\endgroup$ Commented Apr 21 at 8:40
  • 5
    $\begingroup$ However, if it got into a review through editing, there is this feature request: Limit the number of times an author can choose "Submit for review" when editing a closed question. So I guess that can happen more than once. (I thought that such edits are indicated in the revision history and in the timeline, but I do not see anything there. Moreover, it seem there were no edits between the recent reopen reviews.) $\endgroup$ Commented Apr 21 at 8:41
  • 6
    $\begingroup$ "it has been "left closed in review" two times in the past 10 hours, and then reopened at the third attempt" - and then closed again after only 1h20min, and then "left closed in review" 2 days later, and now (11 hours later) I have encountered it again in the "reopen" queue. This looks like a game of hide and seek, it leaves an impression of using MO in bad faith, and is getting annoying. Could a moderator lock that post, please? $\endgroup$ Commented Apr 23 at 18:05
  • 3
    $\begingroup$ I’m voting to close this question because it's based on a false premise that a user can nominate a question for reopening more than once. That is, if you see a question reopened 20 times, it means that 100 people wanted it reopened. A 100 people, not 5 people. $\endgroup$ Commented Apr 26 at 21:03
  • 3
    $\begingroup$ @DimaPasechnik: I'm confused. If a question was reopened 20 times, then it was closed 20 times, at least. So at least 100 people want it closed. No? Also, the question here is about the adding of a question to the review queue for reopening a question, not quite reopening the question. $\endgroup$ Commented Apr 26 at 21:42
  • 4
    $\begingroup$ @AsafKaragila There is a caveat though: one user can vote to close a question only once, but they can vote it to leave it closed from the review queue multiple times. And, indeed, in this question the 12 "leave closed" votes come from 5 distinct users in total. (On the other hand, there were 4 nominations to reopen from 4 different users, I think, but we don't know which users, and the suspicious timing might suggest sockpuppets.) $\endgroup$ Commented Apr 27 at 8:34
  • 3
    $\begingroup$ @AsafKaragila If the question entered reopen queue via flagging, then I suppose moderators can check who flagged the post (and whether it was the same person or not). Do moderators have some tools to see how the question entered the reopen review queue? (I.e. by reopen vote, by editing the question, or by flagging - I am not sure whether there are also some other options.) $\endgroup$ Commented Apr 27 at 16:12
  • 2
    $\begingroup$ it can be that the author of the closed question asks colleagues to vote to reopen. It's IMHO a legitimate way to get a re-open. One should not assume that illegitimate "spare" accounts are used. $\endgroup$ Commented Apr 27 at 20:35
  • 6
    $\begingroup$ @Dima: Asking your buddy to vote on your behalf is not, will not, and had never been a legitimate way to do anything. (Note that this is inherently different from having your colleagues do something on their own accord.) $\endgroup$ Commented Apr 27 at 21:02
  • 3
    $\begingroup$ @AsafKaragila - as Federico said, reopening can be stopped by 5 active users, against wishes of, say, a 100 users. 5 active users can effectively bully someone they have something against out of asking questions. $\endgroup$ Commented Apr 28 at 1:05
  • 5
    $\begingroup$ @Dima: We all see terrible questions that get incredible amount of "drive-by voting" once they get public enough (e.g., get on the Hot Network Questions list or posted on some other platform). So we all have a vague understanding that voting is not necessarily representative of quality or suitability for the platform. And if my memory serves me (possibly wrong), there were discussions in the past about people voting to close on subjects they aren't familiar with, and there are arguments to both sides. But this shouldn't happen "in the shadows". $\endgroup$ Commented Apr 28 at 11:06
  • 3
    $\begingroup$ Makes me wonder what the real NSA thinks about this conversation. Surely, repeatedly mentioning their name in proximity to things like conspiracy and plotting must raise a bunch of red flags in their network monitoring software. $\endgroup$ Commented Apr 29 at 10:12
  • 3
    $\begingroup$ @Dima: I vehemently disagree with your for sentence. And yes, it is impossible to police these kind of things. Which is why we trust users to behave in a kind and trustworthy way, and this is also why when users are found to not behave in a trustworthy way they are removed (temporarily and to varying degrees of time, or less temporarily) from this community. $\endgroup$ Commented Apr 30 at 7:41
  • 3
    $\begingroup$ @Dima: I disagree that "ask it on MO" is a call for votes. Also, unlike classical mathematics, real life doesn't have formal and rigid rules. There are rules, some clear yes and clear no, and a lot of gray zones. In this community, the moderators act as human exception handlers. When something is problematic and the rules aren't clear, we deal with, manage discussions on it should the need arise, and help to enforce rules. For example, there's no rule against using the letters F, u, c, or k in a comment. But using all of them together is considered a violation of "be nice". $\endgroup$ Commented Apr 30 at 15:15
  • 6
    $\begingroup$ @Dima I suspect your comments have ceased to be constructive for the OP. As Asaf mentioned earlier, MO is not a wholesale discussion forum by design and implementation, we on MO (being mostly academics) are just a bit more casual with that, certainly compared to some of the other sites. Continuing to try to probe and see where the discussion leads is not helping to clarify or answer the OP's question. I personally have nothing further to add. I voted zero times either way on the question at hand, and have nothing against non-standard analysis. I dislike people disrupting the mood in MO, tho. $\endgroup$ Commented Apr 30 at 21:32

3 Answers 3

Reset to default
6
$\begingroup$

It would be nice to see clarification from somebody who know the processes around closing and re-opening well. (I am not sure what happened here and how the same user can get the question into the reopen review queue repeatedly in such a short timespan.)

My understanding is that one user has only one re-open (or close) vote - but the votes age away. So if the process of re-opening (closing) is not finished, the same user might cast re-open (close) vote again. (But only after the vote aged away, which takes some time - so this is not the situation in the particular case you pointed out.)

One situation where I think it is legitimate to cast a re-open (close) vote repeatedly: The question gets into the review, but not enough users actually review it and the review is invalidated.

This happens quite often on meta. From the stats for close and reopen review queues, you can see, that only a few people actually do reviews on meta.

  • Here is a SEDE query showing questions which have been through reopen review queue several times: main, meta.
  • Here is a similar SEDE query for close votes: main, meta.
  • List of review of a specific post - main 19152 and 23478; meta 3486 and 672. (I have chosen some posts withe several reviews as examples.)
  • Users with sufficient reputation can see the history in the review queues on main and on meta. Here are SEDE queries - accessible to any user: main and meta. And in the following queries you can change which types of reviews you want to display: main and meta. You can see that on meta the review often ends as "Invalidated".

When using those queries, keep in mind that SEDE is only updated once a week.

I will freely admit that several of the questions on meta with many close votes reviews got there through my close votes. I consider them to be obvious duplicates. (Or at least most of them.) As I explained, if there weren't enough reviewers and a review was invalidated, it seems reasonable to try the close process again. (In the hope that actually some users might look at the review this time.)

$\endgroup$
0
$\begingroup$

It is specious to assume this is a single user; the default assumption should be that it is multiple users.

This specious assumption backs your later claim about 'consensus' and 'wasting time' -- if it was multiple users, it would appear that 'consensus' is on the side of reopening the question.

Assuming that consensus goes for reopening, it is actually stuff like this that is 'trying to go against consensus' and 'wasting time' to use your phraseology. I would call it 'good and healthy democratic discourse', but to each their own.

$\endgroup$
5
  • 6
    $\begingroup$ Occam's razor: it seems very unlikely that three different users, independently, in the span of ten hours, decided to flag for reopening a question that wasn't on the front page and had seen no interaction, not even a comment, in about six years. $\endgroup$ Commented Apr 22 at 17:49
  • 3
    $\begingroup$ But even if it was multiple users, there is still something wrong with the system: reviewing the same question three times in the same day is not a good use of the time of volunteer moderators. $\endgroup$ Commented Apr 22 at 17:53
  • 7
    $\begingroup$ Having consistently received "leave closed" reviews, I'd find it hard to argue for consensus for reopening. Quorum? Perhaps, but then again, also a quorum for leaving closed. $\endgroup$ Commented Apr 22 at 18:23
  • 10
    $\begingroup$ Good healthy discussion in the comments has resulted in multiple closures/leave closed decisions by the community, even without the moderator involvement. That feels like a fairly strong signal to me. That this is posted on three different sites is a huge red flag to me. We tend to not accept such behaviour from new users who don't know how the system works (ideally, being gentle about it), to be consistent, accepting it from an established user doesn't set a good example. Also, how much additional insight is going to land ten years later? It really should go at matheducators, if there's any. $\endgroup$ Commented Apr 23 at 6:54
  • 3
    $\begingroup$ @David, I agree with your position that the community should be left to decide such cases. I hope you would also respect the community's decision if it is to keep a question open. The question is concerned with attitudes of both mathematicians and educators with regard to the effectiveness of epsilon-delta teaching, and of course the differences between them. Therefore this question arguably is appropriate both and education and math SEs. I appreciate your focusing on the relevant issue here, namely multiple sites. ... $\endgroup$ Commented Apr 27 at 10:21
-2
$\begingroup$

Let me point out that we are in the rather delicate field of figuring out how to agglomerate votes to achieve "the best" outcome. This is far from trivial. Democratic systems often have to be set up in a way that let a sizeable minority have a voice, as you would like to allow a degree of plurality, and you want the minority opinion to be represented. (Otherwise you get a very unhappy minority: indeed one sees a bias against non-standard analysis by a number of prominent members, especially in the area where comparison of standard and non-standard methods enters the picture, and this is not healthy when a pro-non-standard analysis minority gets shut down due to this bias).

I think a question where there is a lot of closing/opening has been done falls into a category of questions which are better left open, perhaps with a special flag.

$\endgroup$
19
  • 5
    $\begingroup$ This question aside, is there really any significant bias against non-standard analysis? Like, some actual evidence to back that claim? Are NSA questions more likely to receive downvotes or be closed, compared to, say, number theory or differential equations or geometry of various types? $\endgroup$ Commented Apr 26 at 10:44
  • 6
    $\begingroup$ "I think a question where there is a lot of closing/opening has been done falls into a category of questions which are better left open": hmm, interesting, can you elaborate why? I would think that if a question attracts a lot of close and open votes, this is indicative that it is controversial, and I believe MO should avoid controversy, extended debate, etc. when possible, so I would take these votes as evidence the question should be closed. $\endgroup$ Commented Apr 26 at 14:22
  • 3
    $\begingroup$ @SamHopkins - a healthy mathematical debate should not be suppressed here on MO. "If not here, then where?" Besides, what's wrong in extended debate? How can you avoid such a debate, say, if you discuss directions for a topic to develop? Or discussing merits of different proofs? $\endgroup$ Commented Apr 26 at 18:32
  • 3
    $\begingroup$ @AsafKaragila - I am not an NSA expert, but there was a dismissive tone on NSA set up long ago by a number of prominent mathematicians, and it sometimes shows here, IMHO. $\endgroup$ Commented Apr 26 at 18:35
  • 7
    $\begingroup$ Dima, about your comment to Sam, this is not a forum. It's a Q&A website. The platform is built to not have discussions. In response to your reply to me, most of the supposed dismissive behaviour I see is actually a small number of users who claim that any closure or downvote is an act of persecution against NSA topics. You don't need to be an expert to collect evidence for discrimination, too, just saying. $\endgroup$ Commented Apr 26 at 19:22
  • 4
    $\begingroup$ @AsafKaragila - "this is not a forum" is an incredibly elitist point of view, sorry. For many, MO is the only place to get in touch with colleagues on a regular basis. Not everyone works in a maths department, not everyone can go to conferences or seminars easily. Please do not tone police users for what you think are "discussions". $\endgroup$ Commented Apr 26 at 20:58
  • 3
    $\begingroup$ I've voted to have this, meta, question, to be closed, for the following reason: " because it's based on a false premise that a user can nominate a question for reopening more than once. That is, if you see a question reopened 20 times, it means that 100 people wanted it reopened. A 100 people, not 5 people." Thus, moderators go against opinion of a lot of users when they repeatedly close re-opened questions. This is not healthy for this forum (sorry, a Q&A site :)) $\endgroup$ Commented Apr 26 at 21:06
  • 7
    $\begingroup$ Dima, this is not elitist. It's about the software design and the meaning of forum vs. Q&A website. I suspect that we're not speaking the same language here. $\endgroup$ Commented Apr 26 at 21:29
  • 3
    $\begingroup$ Asaf, well, software enforces the elitist setup. Are you saying you're just the servant of the software? :-) $\endgroup$ Commented Apr 26 at 22:28
  • 3
    $\begingroup$ Dima, it's true that you can cut bread with a screwdriver if you really try and want to. $\endgroup$ Commented Apr 27 at 7:54
  • 7
    $\begingroup$ Let me reinforce Asaf's question: Is there any evidence that there is bias against NSA on MO? Martin Sleziak's SEDE query lists the posts with most nominations to reopen, and I don't see any NSA on the list apart from this question. Some users (including me) listed their motivations to vote to close this question, and they did not mention NSA. $\endgroup$ Commented Apr 27 at 8:46
  • 6
    $\begingroup$ I've decided to spend a few minutes looking into the NSA tag. Currently there are 124 undeleted questions (of which, 24 come from a single user). There are also another 31 questions which are deleted, but a random sampling of a few from the past 15 years indicated things like objectively bad questions (e.g. "ChatGPT said so and so") as well as questions from users which have a long history of garnering downvotes. In comparison, the real analysis tag has 5758 undeleted questions and 3563 deleted questions. The number theory tag, 7176 deleted vs 17007 undeleted. $\endgroup$ Commented Apr 27 at 11:34
  • 5
    $\begingroup$ In short, I don't see any particular evidence to "NSA is unwelcomed here". $\endgroup$ Commented Apr 27 at 11:34
  • 2
    $\begingroup$ @Kostya_I At least one commenter under the post read it as a thinly disguised [attempt] to push an agenda about infinitesimals, even though the post does not mention NSA, as you said. $\endgroup$ Commented Apr 30 at 8:42
  • 4
    $\begingroup$ Frankly, that comment has a broader context than what it appears to be. But as a moderator, I'd feel uncomfortable discussing this (even though I was not a moderator at the time of that comment). $\endgroup$ Commented Apr 30 at 10:47

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.