# How much is MathOverflow exposed to recent events on StackExchange?

[Long-time 10k+ MO user here, asking this question anonymously for reasons that should be obvious. Added Oct 16: My access to this account will self-destruct in 3, 2, 1 .... (sorry, won't be able to accept an official answer. For what it's worth, I strongly support Tim Campion's suggestion.)]

This question is not meant to be a place to discuss the pros and cons of the dumpster fire on meta.SE, or the underlying issues that started it. Go there if you want to get depressed. Also, I think that only moderators are in a position to answer this question.

The short background is that StackExchange Inc. 1) is trying to implement a new Code of Conduct (CoC), and it's proving difficult, and 2) has retroactively relicensed all existing content, questioned by many as being of dubious legality. These are separate issues, but have both contributed to a wide feeling (expressed through community voting and extensive commenting) that SE Inc. is making a mess of things, one way or another.

Since MathOverflow uses the SE platform, but is actually a different organisation it is not clear what exposure (in the sense of risk) MO has to the recent controversies. Are our moderators bound by the same agreement as the SE network moderators? Will the CoC apply here, when it is finally sorted out? Were contributions to MO retroactively relicensed (making this question outdated)?

Again, the intent of this question is not the content of the CoC or the relicensing issue, but to get in writing the facts of where MO stands in relation to these from a contractual/legal/etc way.

Edit (To clarify some things) I asked this question in a slightly vague way, partly because I didn't want to pre-empt any position or bring up debates here. But I've read a bit more, and Johannes pointed out some relevant facts in an answer. So here are some more focussed questions. Again, this is not to debate the merit or otherwise of the below points, I'm looking for official answers from mods or MO owners/board members on policy about changes that StackExchange Inc makes.

• StackExchange relicensed all contributions users had made to their websites (from CC-By-SA 3.0 to CC-By-SA 4.0, which change the CC people say can't be made without permission), but this has also relicensed all MathOverflow content too. Is this compatible with the agreement between the MathOverflow corporation and StackExchange Inc?

• I presume that there is very little objection to the principles of the new Code of Conduct (CoC), and find it nigh-impossible to forsee any circumstances in which MathOverflow as an organisation would object to the now in-position wording, and adopting the CoC as official here (were it a choice). However, the official FAQ on the CoC, which has given rise to concerns among many StackExchange users, is getting rather distant from any document that I can see referred to under the legal agreement between MathOverflow and StackExchange Inc. The CoC is linked to from MathOverflow pages, and so is, it seems, in effect and binding. I haven't seen, after some cursory clicking about, explicit links from MathOverflow to the contentious FAQ. Does this latter nonetheless carry force on MathOverflow to the extent it does on the StackExchange network? Will all future policy changes in the same manner also immediately and automatically apply on MathOverflow? (Added: I note that this type of question was raised nearly six years ago, to which it seems the official answer was, to paraphrase, MathOverflow has an informal understanding with StackExchange Inc that SE Inc employees shouldn't interfere with our moderation as we can choose to leave the network. However, this seems to me to be a bit like the nuclear deterrent...)

Perhaps all of this is moot. But I hope MathOverflow is not hit with the same ugly stick terrible optics that StackExchange Inc has encountered in trying to create an environment that once was expressed through 'be nice'.

Update: Yesterday a new FAQ regarding the pronoun discussion has been released. This new FAQ is based on previous input from the community and it seems that the acceptance of these new rules is (so far) much higher.

• I'm pretty sure that so long as MO stays on the platform we're subject to the same CoC as everyone else. We could of course try to take our ball and go home. But although my heart breaks for Monica (who's an old friend from another site, who I think very highly of, and who was really terribly treated), as far as I can tell there's nothing actually bad in the CoC, so I can't imagine leaving being seriously considered. Oct 12 '19 at 1:51
• @NoahSnyder I still don't fully understand what's going on even after several hours of reading over the last two days. Most of the uproar that isn't tied to Monica's firing but to the new CoC itself seems to come from the FAQ post which reads very different than the CoC itself. It is easy to read the new code charitable and as an expansion of the general "don't be a dick" theme of the CoC. It is much harder to read the FAQ in this way. Especially the questions about what is considered a violation and how violations will be enforced are causing worries. Oct 12 '19 at 2:05
• I hope this pronoun madness doesn't start disrupting MO. Oct 12 '19 at 18:33
• I think the meta.SE post about pronouns with the best ratio of light to heat is this one which has a more understandable version of what the FAQ is hopefully trying to say and an insightful answer with a somewhat different viewpoint raising a few objections. I think it's really unfortunate that SE rolled this out in a way that would maximize distrust of them (since Monica is widely beloved), not least because it's harmful for SE's trans and non-binary users that their reasonable requests are now associated with SE's unreasonable behavior. Oct 12 '19 at 23:04
• To make things clear for casual readers, the link in @NoahSnyder's comment does not discuss the official FAQ. Rather, it proposes a replacement for the FAQ. It is basically irrelevant, as the company has made it clear over the course of the last year or two that they have zero interest in taking suggestions from the community into account, they just want everybody to shut up and do what they are told. Oct 13 '19 at 7:47
• Also, at this point it’s clear that the company does not care one iota about members of the LGBTQ+ community, or of any of their other users for that matter. Users of the free service are not customers, but a product; SE treats them (that is, us) as an expendable commodity it owns. The only point of the CoC update is that they can show it to their investors and real customers (advertisers, etc.): “here, look, we care about inclusiveness and diversity”. The actual effects of the change mean nothing to them. Oct 13 '19 at 7:59
• Concerning the licensing issue, it might be a good idea if representatives of the MathOverflow corporation in their official capacity ask SE for explanation what is the legal basis for this unilateral retroactive licence change. They staunchly refuse to say anything to users, but they just might treat a request from the MO corp. more seriously. Oct 14 '19 at 7:20
• @FrancescoPolizzi There is some discussion of “the OP” in the FAQ thread, but it seems inconclusive. The only thing that’s clear is that if a CM determines it might offend someone, the punishment will be swift and hard. Oct 14 '19 at 10:17
• That is, the official “rule” is apparently that if you know someone’s pronouns, and call them “the OP”, it’s ok if this is something you would do anyway, but it is a punishable violation of the CoC if you’ve done it in order to avoid using the pronouns. Since SE failed to issue moderators with mind-reading technology, it’s anyone’s guess how they are supposed to judge your intent. Oct 14 '19 at 12:03
• Am I particularly susceptible, or this at the very least seems irrational, and at worst borders on lunatic? Oct 14 '19 at 12:40
• My opinion is that requiring the use of requested pronouns and punishing people for using a completely innocent and neutral form as "the OP" is coercing speech. Oct 14 '19 at 12:49
• You judge intent in this case the same way that you judge it in other cases: by examining the visible evidence. If somebody's posting history makes it clear that they always say "the OP", then the intent in using it now is probably not to avoid pronouns. But if their posting history makes it clear that they regularly use pronouns and have never before said "the OP", then the intent behind its use now is much more questionable. Mens rea goes back at least to the 13th century (says Wikipedia) and has never required mind-reading, just examination of the available evidence. Oct 14 '19 at 21:16
• For the sake of some misinterpreted "inclusiveness", we are making people afraid to speak and we are coercing their language, expecting to judge their intent. We are using the expression mens rea when talking about "the OP"! We suddendly are in "1984", or what? Oct 14 '19 at 22:33
• @Francesco Polizzi: I basically agree, but what if a user "explicitly asks" to use neopronouns like ze/zir? I do not know how to conjugate them nor do I think this is real/correct English language. Shouldn't a user have the right to avoid such doubtful constructions? Or, in the inverse direction: Why should a minority of users have the right to dictate other users which words they have to use?
– tj_
Oct 15 '19 at 6:29
• Having said that, SE now scrapped the original CoC FAQ and replaced it with a much better one, taking the most important complaints of the community into account, see meta.stackexchange.com/q/336364 and meta.stackexchange.com/q/336366 . There is some hope the company finally started to communicate with its users again. However, the situation of Monica (that involved highly unethical behaviour on the part of SE) as well as other issues such as illegal relicensing remain unresolved as yet. Oct 23 '19 at 11:46

I think that given the sort of turmoil experienced of late at Stack Exchange, it might be wise for us here at MO to take certain measures to reinforce our status as an entity independent from Stack Exchange. I'm no legal expert, but I think our migration agreement is rather vague on many points concerning exactly what our status is. And if we establish too much precedent of deferring to Stack Exchange, that could come back to bite us down the road. For example, we might contemplate

• Writing our own Code of Conduct. It could even be verbatim the same as Stack Exchange's, but it would be ours to interpret as we see fit.

• Agreeing on our own licensing convention. That might mean we use the same license as Stack Exchange uses, but it would be our decision.

• Negotiating some written agreement with Stack Exchange over when they can and cannot unilaterally "fire" one of our moderators.

I don't know what kind of standing we have to do any of these things, but I think steps like this might be worth discussing.

• Good idea, IMHO. Here's a virtual +1 as this account can't upvote. Oct 15 '19 at 5:44
• It has been > 2 years. Where are the steps that MO has taken with respect to these three points you stated? Jan 11 at 18:17
• @user21820 I'll point out that at the time I wrote this answer, I was not a moderator, so this answer shouldn't be taken as "official MO policy". As far as I'm aware, not much has been done along these lines, so thanks for the nudge.
– Tim Campion Mod
Jan 11 at 18:21
• Oh oops I didn't realize your moderatorship started later! Thanks for responding, and I would be glad to see an update on this in the future. =) Jan 11 at 18:25

Will the CoC apply here, when it is finally sorted out?

seems to be "there is no 'will' and 'finally'". The blog entry posted on october 10th shows (some?) of the controversial changes to the CoC. The two changes mentioned in the blog post are already part of MO's CoC. I cannot say whether those lines were there before a couple of days ago, but they are there now.

There are very few MathOverflow posts in which anyone might feel a need to use a third person pronoun. If such a situation arose, it is most likely that the writer would naturally use a pronoun that everyone involved was happy with. In the unlikely event that a pronoun was used and someone was unhappy with it, by far the most likely scenario is that the question would be resolved amicably with an apology freely offered and graciously accepted. So my estimate of our "exposure to recent events" is zero.

• One would hope amicable solutions can be speedily arrived at in all cases. But this is not really an answer to the question, which is more after a more-or-less official answer about the relationship between SE Inc and MathOverflow, and if and how changes the former makes on the SE platform do or don't apply automatically to MO, given MO's special status. Oct 14 '19 at 10:39
• "it is most likely that the writer would naturally use a pronoun that everyone involved was happy with" - The CoC is not about being happy, it's about inclusiveness. For example the usage of "she/he" in mathoverflow.net/questions/92750/… excludes trans persons. To my understanding, in the specific case of the linked question, under the new CoC, one needs to use singular they in place of "she/he".
– tj_
Oct 14 '19 at 11:22
• The usage of she/he in that question is clearly meant to be inclusive... Also, it doesn't necessarily exclude transgender people, since many of those identify with a specific gender... It could be interpreted as excluding non-binary people, if one is so inclined, but perhaps we could salvage this with a topological interpretation? she---he is a continuum, not necessarily one-dimensional, in a larger dimensional manifold (humanity) that cannot be split continuously. Oct 16 '19 at 11:12
• @FrankaWaaldijk Interesting. We should write it as [she,he], then? Oct 16 '19 at 13:11
• I’ll answer myself: no, this would imply she${}\le{}$he. Oct 16 '19 at 13:55
• This analysis is reasonable under the assumption that no one provokes a conflict intentionally (e.g., by changing their pronouns and asking for retroactive changes in existing posts, or declaring their pronouns in an ambiguous way, or otherwise fishing for reasons to be outraged). Unfortunately, this is far from guaranteed on an open forum. (This kind of behavior has happened on the political scene in Canada, for example: google Jessica/Jonathan Yaniv.) The question then becomes whether administration will take the side of the provocateur or that of the community. Oct 16 '19 at 18:21
• I don't agree that there are very few MO posts in which one might want to use a third person pronoun. Looking through my own previous answers, I almost immediately found something of the form "When Kunen proved..., his result was hailed as..." I can easily imagine saying, "In so-and-so's MO answer to a related question, she invokes such-and-such a theorem." It's true that sometimes these can be rewritten to avoid using a third-person pronoun, but my understanding is that choosing a different phrasing from what you would normally use, solely to avoid a pronoun, is a potential CoC violation. Oct 16 '19 at 20:35
• @Frank'a Waaldijk: I think that in my example above (under the new CoC) Q6 of the official FAQ (meta.stackexchange.com/questions/334900/…) applies: Q6: What should I do if I don't know someone's pronouns? A: When in doubt, use gender-neutral language or refer to the user by name. -- In the example, the pronoun (she/he) is used in the question and doesn't address a particular user. Hence, no pronoun is known and by Q6 gender-neutral language should be used.
– tj_
Oct 16 '19 at 21:20
• Neither the "controversy" nor the amendment to the CoC are due to such incidents actually occuring in reality; that's not what it's about. This is a power play, and MO will be affected by it. Oct 20 '19 at 22:12
• The CoC amendments are certainly about real incidents, mostly on sites like The Workplace, Interpersonal Skills, Mi Yodeya etc where there is obvious scope for such incidents to happen. You have offered no argument at all for why MathOverflow will be affected. Oct 21 '19 at 5:30
• @NeilStrickland "Certainly"? Did you ever come across such incidents in real life? I didn't. My reason to belief that MO will not be unaffected---despite the apolitical and asexual nature of mathematics---is that I see this as yet another example of an all-to familiar script that has crippled other communities in the past (e.g. atheists, linux kernel devs). The first step is always the introduction of a "code of conduct", written in a way that turns perceived victimhood into a currency yielding the power to silence and deplatform others. Oct 21 '19 at 13:34
• @NicolasSchmidt I wasn't aware that there was a worldwide code of conduct for atheists. Seems kind of paradoxical. In any case nobody notified me. Oct 22 '19 at 15:07
• @NajibIdrissi There isn't, nor did I claim there was. Look up "New Atheism" and "Atheism+" for context. Oct 22 '19 at 15:30
• @NicolasSchmidt I have heard no one disagree that the LKML had a lot of open abuse of a kind that has never been tolerated by community or moderators on MO. They introduced a CoC, there was a brief burst of drama about it, then everything quickly calmed down. The loudest complaints were about Linus Torvalds; he stepped away from the project for six weeks, but was hardly silenced or deplatformed. A look at lwn.net/Articles/798505 does not support the claim that the Linux kernel community was crippled. Oct 22 '19 at 15:40

It is more likely a question for StackExchange Community Managers than for moderators. I believe the moderators can provide an interesting but non definitive perspective. If the comment linked in your edit does not suffice as an answer, I cannot imagine more satisfaction coming from the MathOverflow moderators. I think (read in my opinion and not StackExchanges) that the less complaints this forum generates, the more the community will be left to manage itself.

Toward the notion of moderating ourselves, I am finding recent instances where the community participating in MathOverflow is not being inclusive. Of course we do not intend to expand the scope to answering standard questions in high school courses (or pick your other line in the sand), but there are other areas where I believe we as community fall short. (For some history, pursue some discussions on tea.MathOverflow, the old meta forum.) Again in my opinion, we the community need to adapt our behaviour toward a model that will support a future audience, not just the present one.

The topic of MathOverflow's current situation is grave enough to me to omit a thought-provoking signature.

• Yes, MO has had (witness blog posts from Izabella Laba, for instance) and I assume still has issues with being welcoming to congressive mathematicians, let alone people from marginalised backgrounds. Oct 13 '19 at 20:13
• There is nontrivial overlap between mods and MO board members as far as I can tell, so I would expect them to use that hat. But, I'm feeling increasingly like I maybe should have let this whole issue lie until if and when it becomes more pressing. Oct 13 '19 at 20:23
• Regarding the discussions over at StackExchange, I think the MathOverflow community has avoided a lot of the angst, even if they are not immune from the underlying issues. There are issues to confront however, and I think your question provides an oblique but effective introduction to these. Gerhard "Hopefully Good Talk Will Result" Paseman, 2019.10.13. Oct 13 '19 at 23:00
• The link to congressive mathematicians in the first comment is very worthwhile, imnsho. Thanx! Oct 16 '19 at 11:20