67
$\begingroup$

Edit Please note this is not me posting with my moderator hat on (I've signed the letter), and I didn't discuss this with the mod team before posting this. It's not within shouting distance of being official. Someone mentioned the old discussion in regards to the current events, and while there are no plans or hints of plans of moving (as noted below), trying to sort things out at the last minute under pressure is not the best time.


Given past and current turmoils, the fact that MO is not owned by Stack Exchange Inc and the ever-present option of moving to a new home is present in the minds of some people. However, as I mentioned to the other mods in a recent discussion, we are at risk of being in the position of the fictional 1980s UK Prime Minister Jim Hacker, with the power of the nuclear deterrent, but no clear idea of when, if ever, it could ever be actually used in response to an aggressor. Especially when said aggressor may just use "salami tactics", making things slowly difficult one small step at a time, none of which necessarily deserve responding with the nuclear option.

MO has the option of just leaving SE Inc, taking its data, and setting up elsewhere. But it is our "nuclear option". Certainly it would not be a good look if MO left SE in the middle of a network-wide dispute of this or that nature (I'm not saying it's what is under discussion, at present), but it would be a big deal to get everything going again elsewhere.

I really don't know even an approximation of the answer, but what do people think would be a line in the sand that means MO (the board, the organisation, the community) gives up on SE and goes elsewhere?


Update October 2023 There's been big movement in the company, please see Practical effects of the October 2023 layoff This does not make me more confident in SE Inc.

$\endgroup$
17
  • 8
    $\begingroup$ Please note Tim Campion's answer here: meta.mathoverflow.net/a/4373/4177 (written before he was elected moderator) and some of the general ideas it contains. The specific suggestions were in response to the biggest problems back then, but the idea of thinking hard about the migration agreement is a good one. $\endgroup$
    – David Roberts Mod
    Jun 6, 2023 at 0:26
  • 11
    $\begingroup$ This is not directly responsive to the question (hence a comment, not an answer) but I expect that there are many (including me) who would be happy to contribute financially to the acquisition and support of the necessary infrastructure if, after much discussion, there was something like a consensus that moving is a good idea. $\endgroup$ Jun 6, 2023 at 2:11
  • 10
    $\begingroup$ It would be good to know how "seamless" vs. "messy" divorcing from SE would be, from a purely technical point of view. I understand that MO owns its "data," but would the website and software largely continue to work? Would it go down for hours... days...? $\endgroup$ Jun 6, 2023 at 3:01
  • 7
    $\begingroup$ @Sam Hopkins asked about the software. There was a bit related question here on meta - it got basically no response, if we do not count a few comments: Is there a software solution if we do cut the cord from SE? Since one of the above comments mentioned financial contributions, I will include link to the previous post on this topic: Donations to MathOverflow, Inc. $\endgroup$ Jun 6, 2023 at 4:29
  • 2
    $\begingroup$ The first comment links to the migration agreement. For the benefit of people who cannot access it on Google Drive, here is a link to the post on this meta containing the same text: What is MathOverflow's "agreement" with Stack Exchange? $\endgroup$ Jun 6, 2023 at 4:57
  • 7
    $\begingroup$ If it were time to migrate away from StackExchange, I would be wondering about the prospect of migrating to Monica Cellio's Codidact platform. Would it be technically possible, and would that migration have the support of the community? Or would MO have to roll their own software and server infrastructure? $\endgroup$
    – ziggurism
    Jun 7, 2023 at 3:32
  • $\begingroup$ I totally lack any information about something crucial here: what are the SE trends that might be harmful (a) for MO and (b) maybe not for MO as an entity but still harmful in some global sense $\endgroup$ Jun 8, 2023 at 4:57
  • 7
    $\begingroup$ @მამუკა The current management have not been making life easy for moderators, and imposing policy that moderators feel will degrade the quality of sites on th network, including MO. We are not thinking of leaving at this time. But the question is to slowly collate a consensus, if at all, what would happen to make us want to actually leave. That is, what do you think are things that SE might do that would be the line in the sand? $\endgroup$
    – David Roberts Mod
    Jun 8, 2023 at 7:33
  • 14
    $\begingroup$ Getting rid of the data dumps and doing so secretly with no discussion may well cross the line for me. meta.stackexchange.com/questions/389922/… $\endgroup$ Jun 10, 2023 at 2:26
  • 3
    $\begingroup$ Not super on topic, but I have database and website design experience and would be happy to help if it ever becomes necessary. $\endgroup$ Jun 10, 2023 at 16:27
  • 6
    $\begingroup$ To me, one of the thresholds triggering the "nuclear option" would be SE firing one or more MO mods (the current agreement between MO and SE does not make it clear if SE can do so). $\endgroup$ Jun 11, 2023 at 4:02
  • 5
    $\begingroup$ For the average user (like myself) I have found it difficult to get a clear overview of all issues, especially things along the lines of those mentioned by @DavidRoberts (e.g. "the current management have not been making life easy for moderators"; this is knowledge difficult to convey to the average user). I understand that there are many layers of this not visible to me, and so I feel somewhat hesitant towards the idea of a general vote for an MOExit. However, I have total faith in our current moderators, having directly voted for them in the past, and I will support them and their decision. $\endgroup$ Jun 13, 2023 at 19:52
  • 2
    $\begingroup$ @Carl-FredrikNybergBrodda let this comment give you a hint of the tone of the room right now meta.stackexchange.com/questions/389922/… $\endgroup$
    – David Roberts Mod
    Jun 14, 2023 at 3:31
  • 2
    $\begingroup$ @DavidRoberts If even the hint is that atrocious… Thank you for sending that. (To be clear: in no way do I attribute my lack of knowledge to you moderators not communicating properly, or anything like that. From what I have seen, you have been doing a very good job) $\endgroup$ Jun 14, 2023 at 7:26
  • 2
    $\begingroup$ FYI, see Philippe's updated answer to the MSE post that Noah linked above. $\endgroup$
    – V2Blast
    Jun 15, 2023 at 18:51

7 Answers 7

46
$\begingroup$

In my opinion, if it were easy for MathOverflow to leave, then we should have done it already. There's no bright line, but it's become clear over the last few years that Stack Exchange Inc is not an honest partner; far more than any specific dispute, their constant lying is the biggest red flag. This is easy for me to say, since a split would be a lot of work that wouldn't be done by me. What it would take to leave in practice, is a different question. The answer to the question that you actually asked is: now.

$\endgroup$
1
  • $\begingroup$ While I'm not particularly active on MO, I do read a lot of the questions and answers here. I concur with Toby's take on the current question. $\endgroup$
    – Clayton
    Jun 12, 2023 at 19:43
27
$\begingroup$

For one thing, I would consider SE changing drastically the licensing on MO data, or making restricted commercial profit out of it, to be a red line. In my opinion, MO data should remain free to access, to use and to cite (even for LLM training if it comes to that, I don't want to pinpoint a particular use-case as long as the licensing is respected). I know that SE management has already performed in the past years a re-licensing (An Update On Creative Commons Licensing) and are trying now to get money for the use of the network data as a training dataset (Is SE [going to be] selling our content for AI model training? And what exactly does "reinvest back into our communities" mean?).

I have very suspicious feelings regarding the latter action, since a lot of recent behaviors of SE top-brass management are far from being to my liking. But that's maybe another point altogether.


EDIT: as has pointed out by Emil in the comments, SE is now holding back the automated data dump: June 2023 Data Dump is missing. As a consequence, I revoked all my SE accounts. I keep my MO account since it is not a complete part of the network, but I'd like now to actively advocate for a referendum (as suggested by Timothy) regarding a SExit (if I may say so). I'm willing to go, I don't see why we should stay in a place where the makers of knowledge are bullied and trod upon by money sharks. There seems to be other networks run by old allies that could be a new home, although the "where" is partly besides the point.

I'm willing to contribute, financially and technically, in any reasonable measure that I can, but first we have to make our collective mind.

$\endgroup$
2
26
$\begingroup$

To me, the answer comes down to "what makes life easiest for the mods". I don't feel any loyalty or goodwill to the SE corporation, but I also don't feel goodwill to the deans and provosts of my university, and I recognize that it is still better for my department to be hosted at a university than to strike out on our own.

As a non-moderator, the site is still working fine for me. In particular, I see very little GPT spam. But I support our moderators in keeping it that way, and if the SE overlords are making our moderators' lives hard, then I support our moderators in moving.

$\endgroup$
1
  • 2
    $\begingroup$ I understand your main point but the comparison is unfair in many ways. For example, your university does give you a regular paycheck, which is intended to compensate you for various efforts toward your university/employer. Nobody on MO is getting a paycheck from SE as a result of their work as moderators nor as board members. SE offer plenty of moderation tools for free but I don't think that's comparable in any reasonable sense. $\endgroup$ Jun 23, 2023 at 6:55
19
$\begingroup$

I suspect that the only way, in practice, to engineer a "MOexit" is the same way Brexit happened: have a referendum. A MOexit that isn't backed by an election result is surely going to be perceived by many as an overreach of authority by those in charge. Conversely, relying on the results of a referendum means that one doesn't have to deal with the messy business of drawing bright lines through gray areas; just state the pros and cons, and put the decision to a vote.

$\endgroup$
9
  • 11
    $\begingroup$ You still need to know when to call the vote though! $\endgroup$ Jun 10, 2023 at 21:29
  • 1
    $\begingroup$ @NoahSnyder True, but the threshold for calling a vote can be much lower, since merely calling a vote is not exercising a "nuclear option." $\endgroup$ Jun 11, 2023 at 2:11
  • 2
    $\begingroup$ Nevertheless, it would be counterproductive to keep calling a (failed) referendum frequently. $\endgroup$ Jun 11, 2023 at 7:20
  • 4
    $\begingroup$ I agree that it shouldn't happen without a vote, but how to make sure that the vote outcome is an accurate representation of of what community feels is best for the future, is another issue. $\endgroup$ Jun 11, 2023 at 19:16
  • 2
    $\begingroup$ @NikeDattani: well, that's the really old question about who "representative democracy" actually represents. Not sure that anybody came up with a complete answer, whatever the chosen ballot type. $\endgroup$ Jun 12, 2023 at 11:55
  • 1
    $\begingroup$ @LoïcTeyssier who it represents is one thing, but more specifically, how they are chosen is another. Elections on SE don't give the community much time to do research about candidates, and even if it did, most people don't know when the elections are and have 0 experience with seeing how the candidates act in important situations, so they vote for people for incorrect reasons. $\endgroup$ Jun 12, 2023 at 13:29
  • 2
    $\begingroup$ I expect that a referendum would be decided by voters who have far less information about both costs and benefits, and far less invested in the outcome, than the moderators do. I'd therefore prefer to see this decision made by the moderators (probably after considerable public and/or semipublic discussion). If we had different moderators, I might feel differently. $\endgroup$ Jun 13, 2023 at 23:21
  • 4
    $\begingroup$ @StevenLandsburg At the risk of sounding elitist, perhaps we could revive John Stuart Mill's notorious suggestion of plural voting, and weight the votes of high-rep users more heavily? $\endgroup$ Jun 14, 2023 at 0:52
  • $\begingroup$ @TimothyChow Some safety measure of that kind will be necessary, since it is easy to create new accounts or sockpuppets to skew the vote. $\endgroup$ Jun 15, 2023 at 10:15
13
$\begingroup$

I love MO, and have always been a big fan of 'sticking it to the man' whenever possible. That being said, as I get older I'm beginning to feel the effects of constantly cutting off my nose to spite my face.

What we have here is a wonderful, beautiful thing; an action like splitting from the network is extreme enough that it could (theoretically) compromise the site's long-term viability, opening up a position in 'the marketplace' (cringe) for a competing site on the SE-network.

Accordingly, any 'redline actions' would have to possess a probability of critical site compromise greater than (or at least equal to) the probability that separating from the SE network would compromise MO in some essential way -- off the top of my head things like extreme monetization of the site come to mind, and it doesn't seem like these new AI moderation policies are entirely detached from questions of money.

For my money, I think MO has (for some time now) had the legs necessary for walking solo and should do so as a matter of principle, to avoid having to dignify nonsense like this in the future by posting about it. But principles rarely keep the lights on (without hard work).

$\endgroup$
2
  • $\begingroup$ There is already a competing site on SE! $\endgroup$ Jun 23, 2023 at 7:05
  • 1
    $\begingroup$ PS: The nose comment is excellent! I will steal it! $\endgroup$ Jun 23, 2023 at 7:08
11
$\begingroup$

Converted from a comment by another user:

To me, one of the thresholds triggering the "nuclear option" would be SE firing one or more MO mods (the current agreement between MO and SE does not make it clear if SE can do so).

$\endgroup$
2
$\begingroup$

There's no such thing as a "line in the sand." It implies that one can define a precise boundary between acceptable and unacceptable behavior, which I think is absurd on its face. The fact that we have to ask the question at all probably means that splitting off is something we actually do want, if we could pull it off. I'm not knowledgeable on the technical side of things, but I'm guessing it would be pretty difficult. So maybe start working on it now, to reduce the chances of needing to sort it out under pressure at the last minute, as you said.

Also, regarding this comment:

Certainly it would not be a good look if MO left SE in the middle of a network-wide dispute of this or that nature

MathOverflow was founded for and by career mathematicians. I think there's no other single group of people on the planet for whom it would be clearer that correlation does not imply causation. I don't think this is an issue.

$\endgroup$
5
  • 4
    $\begingroup$ It would not look good for SE, I meant. And optics are not about causation. It doesn't matter if there is an actual causal relationship, people will suspect there is one. And I don't mean a sharp line, a threshold; I mean a condition sufficient to clearly indicate it's time to go. $\endgroup$
    – David Roberts Mod
    Jun 11, 2023 at 8:09
  • 1
    $\begingroup$ I guess I'm saying that it's not possible to predefine that sufficient condition in any reasonable way. $\endgroup$ Jun 11, 2023 at 8:14
  • 1
    $\begingroup$ To elaborate a bit more: I think that if we draw that line, every time SE makes a questionable decision (which seems likely to keep happening), we'll have to ask two questions: (1) did it cross our line, and (2) was our line drawn correctly, which means having the line at all isn't really that useful. I would think the correct path is just not to subject ourselves to that entity that makes those questionable decisions in the first place. $\endgroup$ Jun 11, 2023 at 8:38
  • 1
    $\begingroup$ Point taken. At the very last, people are thinking and talking about the idea, rather than just feeling grumpy in private about things SE does $\endgroup$
    – David Roberts Mod
    Jun 11, 2023 at 11:37
  • 9
    $\begingroup$ A line in the sand is never the whole definition of “acceptable” vs “unacceptable”, but it can be a very useful part of negotiating a relationship. Laying down a red line isn’t saying “everything up to there is OK”, nor is it saying “until you reach there, I’ll do nothing”. It’s saying explicitly: here is point which would be 100% unacceptable, and if you go there, I will be willing to walk away from the relationship and burn the bridges. (And implicitly: if you approach that point, I’ll be increasingly, unhappy, and will negotiate with whatever intermediate options are available.) $\endgroup$ Jun 12, 2023 at 13:42

You must log in to answer this question.

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