27
$\begingroup$

Context for this question: I am writing an article on MathOverflow for the Notices of the AMS. One thing I try to address is the apparent gender skew of our website. One of the readers of my draft asked what procedures MathOverflow has in place to address harassment or similar misbehavior. Since I'm not a mod, I don't know, so I figured I'd ask over here.

Unless there is an objection, I'll probably cite the answer to this in my article.

$\endgroup$
14
  • 2
    $\begingroup$ I just found this, by Sune Kristian Jakobsen, posted here: "David Speyer once made a similar experiment to test if he got more upvotes on answers when he posted [on MO] with his own name, than when he posted as an anonymous." Might you include this in your article? $\endgroup$ Dec 23, 2020 at 2:00
  • 2
    $\begingroup$ MathOverflow specifically, or StackExchange in general? I presume the latter. And that is something that, to my knowledge, has been slightly in flux over the past 18 months due to various events. $\endgroup$
    – David Roberts Mod
    Dec 23, 2020 at 2:40
  • 3
    $\begingroup$ MathOverflow or SE? It makes a big difference. A user could be active on 50+ SE sites. $\endgroup$
    – smci
    Dec 23, 2020 at 12:39
  • 2
    $\begingroup$ Mathoverflow please. $\endgroup$ Dec 23, 2020 at 13:06
  • 8
    $\begingroup$ @JosephO'Rourke I can't get to my old post about this to read the results, because tea is down, but my recollection is that the difference was very small and not significant. (IIRC, for 20 answers I wrote them and then flipped a coin to decided whether or not to sign my name.) I remembered this but was planning not to include it because (1) it was in 2010 and (2) I think the conclusion is not that reputation doesn't matter but that writing like someone who has been on the site for a while matters much more than the number next to your name. $\endgroup$ Dec 23, 2020 at 13:10
  • 2
    $\begingroup$ The old post is msleziak.com/mathoverflow/tea/discussion/385/… $\endgroup$ Dec 23, 2020 at 13:13
  • 6
    $\begingroup$ "(1) it was in 2010": MathOverflow has changed quite a bit in the last decade. Like a different site now than it was then. So it makes sense to not mention that old experiment. $\endgroup$ Dec 23, 2020 at 13:43
  • $\begingroup$ @DavidESpeyer (and smci) I meant policies, not behaviour. As far as I know the harassment/abuse policy is meant to be uniform network-wide. $\endgroup$
    – David Roberts Mod
    Dec 24, 2020 at 5:05
  • 2
    $\begingroup$ Since moderators have to implement these procedures, it seems relevant also to ask what training, if any, moderators are asked to do on gender discrimination. $\endgroup$ Dec 24, 2020 at 20:12
  • 8
    $\begingroup$ It might be worth noting that last year there was a huge debate on SE that essentially stemmed from a disagreement on what is considered discrimination/harassment. In particular, @MarkWildon, "training" moderators on what discrimination is may be a thorny issue. $\endgroup$ Dec 25, 2020 at 17:12
  • 8
    $\begingroup$ There's Code of Conduct that is general for all SE sites. $\endgroup$
    – Andrew T.
    Dec 26, 2020 at 6:45
  • 6
    $\begingroup$ One answer was upvoted 9 times and downvoted 20 times, and deleted, moved to chat as well as the ongoing comments. The link is: chat.stackexchange.com/rooms/117841/… $\endgroup$
    – YCor
    Jan 3, 2021 at 13:41
  • 4
    $\begingroup$ Please don't vote to undelete my answer. There was a somewhat coordinated offsite effort to crater it from +2 to -9, and I'm not interested in having any more attention from twitter. Thank you in advance. $\endgroup$ Jan 5, 2021 at 11:27
  • 9
    $\begingroup$ I understand Harry's concern. However, I find quite worrying that has been a "coordinate effort offsite" to downvoting a completely polite answer, especially because the starting point was given by some user inside MO. Online shaming aimed to shut-up dissent is typical of Twitter, and I am afraid that this kind of disruptive and intimidating strategies could become usual here, too. $\endgroup$ Jan 5, 2021 at 19:27

3 Answers 3

21
$\begingroup$

My apologies for being slow to respond, but I felt there should also be a response from the moderators here. MO is covered by Stack Exchange's Conduct Policy (https://meta.stackexchange.com/conduct) and has no other separate conduct policy. This policy specifically bans harassment, and empowers moderators to impose penalties, up to banning users, in response to such behavior. The moderators support this policy, and endeavor to enforce to the best of their ability.

$\endgroup$
14
$\begingroup$

The obvious answer:

  • Questions and answers containing harassment can get downvoted. Posts with sufficiently many downvotes can be deleted.
  • Questions, answers and comments containing harassment can be flagged as "rude or abusive" or for moderator attention. Then moderators can delete them.
  • Moderators can suspend repeated offenders.
$\endgroup$
2
  • 17
    $\begingroup$ It maybe worth adding that, unlike most other "community" websites, mathoverflow does not have a private message feature. Depending on the background of the reader, they may have assumed that there was such a feature. $\endgroup$
    – dhy
    Dec 25, 2020 at 17:02
  • 7
    $\begingroup$ As corroboration of this answer, note that the Code of Conduct lists three stages of "Enforcement": (1) Warning, (2) Account suspension, (3) Account expulsion. $\endgroup$ Dec 27, 2020 at 13:43
13
$\begingroup$

At least from the user side, there is a procedure that hasn't been listed yet: following the link to https://mathoverflow.net/contact and contacting support with "I want to report a Code of Conduct violation". This is, as far as I can tell, the procedure most clearly dedicated to preventing harassment.

I don't know how these "contacts" are handled on the moderation side, and would appreciate elaboration from someone who does.

$\endgroup$
1
  • 1
    $\begingroup$ Since the "contact us" link was mentioned, perhaps it worth adding that there is also an email address specifically for MO moderators: Who are the MathOverflow moderators? (Probably it's up to them to say whether this contact would be suitable for the purposes discussed here.) $\endgroup$ Jan 6, 2021 at 5:58

You must log in to answer this question.

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