Gerhard "Ask Me About System Design" Paseman, 2011.10.28

I would start posting a mix of questions, some of which I knew and some of which were related to the ones I knew, but had not worked through, whether for laziness or lack of cleverness or what have you. Those who paid attention would see some of the struggles and successes with my current dabblings on Jacobsthal's function, whether they be in the literature or not. Then I might resurrect my work on the Hadamard matrix conjecture, then on Frankl's union closed sets conjecture, followed by hyperidentities and other work related to Murskii on finite basis problems. I figure each of these topics to be good for 10 posts at least; depending on my strategy, I might let half or more of them be questions to which I know the answer.

My intent would partly be to circumvent (or totally alter) the process for doing graduate level research; I tackle the stuff I know I can handle, and the stuff "I don't feel like doing" I post on MathOverflow. If I don't get shut down by the moderators, I amass enough material for a dissertation or two and present it to my advisor.

The discussion of how ethical this is (or not), how it changes the role of the advisor (who might be compelled to check the references and attributions made or not made in the submitted draft of the dissertation), and how it affects doing research can be saved for other threads; I think this is a lousy way to make a mathematician. It takes away some of the struggle I feel is necessary to build one's abilities. (Supporting anecdote: I recall the time I proved to myself that the real numbers were separable and how this could be used in forming sequences of functions that were used in regularity results of Leray on Navier-Stokes; never mind that I did not really understand the PDE course I was taking nor that I had not yet taken topology; I'll always remember the flash of "there is a rational number between any two distinct real numbers", and how from that followed many of the claims Leray made in his paper. The struggle to reach that insight as much as anything from the claim of certain properties of countable sequences of functions was formative for me.)

It's possible a version of such a policy might be useful to MathOverflow, but I am not seeing that yet.

Gerhard "Is Feeling Somewhat Expansive Today" Paseman, 2011.10.28

http://blog.stackoverflow.com/2011/07/its-ok-to-ask-and-answer-your-own-questions/

To be crystal clear, it is not merely OK to ask and answer your own question, it is explicitly encouraged. (Joel Atwood, emphasis in the original)

This leads to questions like the following: http://math.stackexchange.com/questions/76683/what-is-the-millionth-decimal-digit-of-the-10101010th-prime

My impression is that on Math Overflow there has been a consensus against this kind of thing, especially when the original poster is not forthcoming about the fact that it is a "quiz" rather than a real question. Thus if we plan to join with the SE network, we should consider to what degree this policy will or should apply to Math Overflow.

Something to consider for the case of migrating MO: a few hours ago, there was a hiccup in the SE 2.0 engine that caused answers to be disassociated from their owners. The bug was fixed in short order, but a bothering aspect of this was that bug reports filed on the meta site were deleted by a moderator (not the elected ones, I might add) rather quickly with nary an explanation/comment.

If we migrate to SE and we get a question like that then we'd just migrate it over to math.SE. There's just no problem there.

In general, I am also not so worried about such extreme cases. (Whether they are really windfall for mathematics, in a positive sense, well I don't know, at least I don't think they are overly harmful.) However, as documented for example by Nilima's remarks in another thread, highly voted questions actually can shape the image of a site (at least locally). And, I assume in this vein, SO even has edited-in a disclaimer for some old superhighly voted questions (that however are essentially off-topic) to the extent that these should not be taken as representative.

What I am slightly more worried about are almost invisible effects: this softish question gets, say 27 instead of 15, that simplictic but nice one 19 instead of 12, yet the real on-topic one still only gets 4.

It is not a big deal and already present now, but I am quite sure it will get worse. If it can be avoided a bit, by not allowing these automatic-voting-powers for almost everyone on the network, I'd too appreciate it. Yet, I understand that this might be difficult to achieve.

This is the currently most viewed, and highest voted, question on Math.StackExchange. You may be interested in some figures.

But Memes tend to be short lived and sporadic, so over all I don't think they can cause that much of a problem. Afterall, no one is seriously using question vote counts for any statistical analysis. In regards to the image a highly voted question may create for a site, in hindsight, I agree with G Edgar's admonition in the Meta.Math.Stackexchange thread I linked to: those type of popularity are windfalls for mathematics as a whole.

And yes, the migration problem. I remember reading here and there (something on Meta.SO) discussions about whether it is kosher to migrate something that has been open for quite some time. One time when this discussion happened was during the infancy of TeX when a lot of questions with ridiculously high number of votes were mass-migrated from StackOverflow. I don't recall whether any thing ever came out of that discussion, so I don't know what the current SE policy about that is.

I think both voting and commenting by non-mathematicians can be problematic, particularly when SO has half a million users. Voting might be broken but does not justify making it even more broken. Note that SE does not allow other new users to comment or vote, only those from the SE network who have at least 200 on some SE site, I understand their philosophy but IMHO it is not a "feature".

Btw, here is a new thing I have noticed recently, when a question is migrated from another SE site the votes also get transfered, e.g. if someone posts a math question on SO and it gets 20 up votes and then it is migrated to MSE the question retains the 20 votes it got on SO and the owner of the post gets 200 rep for it on MSE even if no user on MSE has upvoted it.

None of these is a big deal in my experience on cstheory, but still it would be nice if they didn't force these "features" on all sites. :)

• If all you care about is knowing what someone's "real reputation" is, you can figure out how likely it is that they got the boost based on their network profile.
• The privileges you get from 101 starting reputation are of no particular importance with the possible exception of CW editing (and that can be abused by users who "legitimately" have 100 reputation just as easily). Voting on MO is already broken.
• Anyone who is "not suited" to MO will probably find that out very quickly and leave anyway regardless of whether or not they get a +100 boost.
• Anyone who is "not suited" to MO and won't want to leave won't be particularly discouraged by not having the +100 boost.

Giving users from elsewhere on the network the ability to leave comments is a boon to moderation; without that ability users will frequently leave answers instead, which will then need to be cleaned up.

somewhere in this thread (the end of the third/beginning of the fourth page) there is a small discussion over the automatic +100. To reiterate my stand on the topic, if MO wants to separate itself from the rest of the SE network it is important that this sort of feature will be gone, or at best diminished.

I'm interested in the possibility that with a number of linked SE accounts, one could garner even 500+ rep by each boost in rep pushing one over the minimum in other subcritical accounts. As was discussed above, community norms vary between sites. What makes a barely ok 2 or 3 questions asked on SO/food/scifi/sceptic/english/TeX/etc poster might not make an ok MO poster.

But as I said, I don't want to raise old discussions.

Perhaps I should keep my mouth shut :-)

For the sake of completeness, these are the privileges you have at 100 rep:

15 Vote up

15 Flag for moderator attention

50 Leave comments

100 Edit community wiki posts

They're all the kinds of things that a similar site might automatically let you do as a new user.

I just happened to look at scifi.SE, and got the following pop-up messages:

Your associated account on Mathematics has passed 200 reputation: +100 reputation

Congrats, you've gained the privilege – create chat rooms and 4 other privileges learn more

so merely by one account on one SE site getting a certain rep count, another account on another SE site got a massive boost, giving me privileges before unthought of. I could imagine this snowballing.

Retroactive automatic editing of question punctuation affecting correctness

The 101 rep problem is something to think about, sure, but I think the structure of meta as an effective tool for community organization/community discussion is totally destroyed by the SE meta approach.

Just look at this thread we're talking in right now.

Now imagine reading it as a comment thread on SE.

and then the recent "Chat" activity

Necessary evil. The reason chat exists is so people don't ask chatty questions and/or clog up comments with long conversations. It's not actually desirable behavior except in that it provides an outlet that keeps the main Q&A part of the site cleaner.

1. Scott Morrison's post from about 5 days ago, beginning to put together some reasonable proposals for migration, is spot-on.

2. My only pet peeve with the current front-end at the SE sites is that they're a little too social-network-y for my taste. For example, the right hand column of Physics.SE starts with a "Welcome!" box, then a "visit Meta" box, then an ad for Area 51, none of which I mind too much (although the logo on the Area 51 ad is probably perfect for attracting random netizens to some site, but not something you'd see in a math journal), and then the recent "Chat" activity, and then the recently posted-in Tags, and then a long list of badges and what people won them. On CSTheory.SE, they have things like "66 people chatting" and "Love this site? Get the weekly newsletter!". I would, of course, get used to this kind of right-hand column fluff (I almost never actually see the ads on gmail, for example), but if possible, I hope we can retain firm control over the front-end displayed content when we migrate.

3. With many other people here, I do wish that we could pay Fog Creek for their services, and retain more control over the look and feel of the site. I don't think we should roll our own --- there is a culture in mathematics of doing all computing things in-house, and sometimes it's great, but often it means mathematicians using outdated or bad technology. But I understand that Fog Creek may not want to create a set-up where some sites pay for services (even if I don't fully understand why they would not want this). Anyway, I think all of the worries people have expressed about migrating will quickly dissipate, and it's certainly true that SE is pretty awesome software.

4. Here's one final thought, which for me speaks in favor of migrating. Every once in a while, I wander over the physics.SE or math.SE, and think, oh, I'd like to leave a comment here, or upvote there. I don't want to be as active in those communities as I am on MO, and I haven't made it over the energy barrier of setting up accounts there. But if it's easy for my MO account to wander to those sites, then great.

It seems with Scott Morrison's list and the earlier suggestion of an SE Academy/ SE Omega set of sites among the SE family, the ideal time to migrate that people have in mind is the day that Fog Creek agrees to accept money for premium services.

I don't exactly disagree with this notion. I think advertisements (even for SE network partners) for example would make the site look noticeably less professional. Worse, they could possibly scare off the sort of mathematician who still hasn't switched to LaTeX and only came here because the site was mentioned in the Notices. That said, as many have noted there's a danger here as that day may not be coming any time soon (maybe it's a weird condition of their venture capital? maybe they feel like we'd be more likely to move on if we were both paying AND had access to database dumps?). I don't feel like I'm enough of a part of the community to make a choice either way but it seems like it would be an easier decision to make if that day were on the horizon.

I can very well imagine situations of the form: Question on MO (about alright but on the simple/general end could well also be on M.SE), some answers/comments, next day somebody suggest on the question or on meta 'wouldn't this better on M.SE...sufficiently many people agree', question gets moved. Whether this happens or not is of course a pure community decision, but what I hope is that we collectively will resist the temptation to transfer too much (both directions) just becasue it is easy, since I think I would find this (mildly) annoying.

I'm not sure why this is an issue. I think a little mild annoyance is well worth being consistent about the division between the two sites to prevent confusion among new users in the future.

First, let me rephrase it. In case too many questions should be moved around between MO and M.SE it seems to me I could eventually come to the conclusion that I'd find it more convenient to have just one site; rather than to be implicitly forced to visit two sites. What I am worried about regarding the moving of questions is that now a suggestion to move to M.SE either happens quickly (almost instantly) or not at all. Perhaps, I am misunderstanding this transfering process, in which case my point is less severe, but if a question with its answers can fairly simply be transferred, then I can very well imagine situations of the form: Question on MO (about alright but on the simple/general end could well also be on M.SE), some answers/comments, next day somebody suggest on the question or on meta 'wouldn't this better on M.SE...sufficiently many people agree', question gets moved. Whether this happens or not is of course a pure community decision, but what I hope is that we collectively will resist the temptation to transfer too much (both directions) just becasue it is easy, since I think I would find this (mildly) annoying.

Second, regarding SE Inc. (not sure if this is the right name, or is it Fog Creek, well I think you'll know what I mean). It is really not my intention to suggest explictly or implictly that they do not have good intentions for MO. And, I find it reassuring that those who know the people their (mainly, you and Anton) have a high opinion about them. Just, as far as I know, not too long ago, they had an altogether different idea how to make use of their software (roughly, offer the software and the servers for a fee to whomever is willing to pay the fee; and as far as I know this was how MO initially envisioned the relation). Now, they have an altogether different idea and scrapped the other one (it is nice and reassuring that they seem to treat their old customers very well); yet one is told what excatly this idea is unclear (eg, how revenue will eventually be create is unclear). So, to me and perhaps to truly everybody it is genuinely unclear how this SE network will look like in one, two, three years. Perhaps millions of people will be there to ask about cooking, gardening, home improvement, parenting, poker, and what not, and this will be great for SE as all these people will be susceptible to quite targeted adds and might buy all kinds of things directly from the site, and it will also be great for the people using these sites because they will get useful information. It just might not be so great for the couple hundred or thousand research mathematicians that use MO on a regular basis, because they simply don't fit into this environment.

And, let me add reseacrh mathematicians in their role as research mathematicians. Of course, most mathetmaticians have interests besides math and might enjoy these sites too. But, just because I enjoy some venue in some context, I do not have to think it is a good place for discussing mathematics.

Its say in matters of what should or should not be with other communities seems to me as quite a peculiar way to run a network of communities.

You might think that. I couldn't possibly comment.

Seriously, I can only talk in terms of the impression I've gotten from hanging around the SE network for a bit. My impression gleaned from that is that SO is The Model and the other SE sites are meant to work in a similar fashion. That's why seasoned users get their "head start" of 100 extra rep: the assumption is that their experience on other SE sites will stand them in good stead on a new site. When the SE2.0 stuff started, then this 100 extra really was for people coming from SO to encourage them to get in to the new communities to show them how it was done. The original system on Area51 (the proving ground for new SE sites) was that your voting capability was tied to your SO reputation. The whole idea of Area51 and the new SE sites was that it was meant to grow organically from the tried-and-tested soil of SO.

Now, with a site like TeX-SX, we haven't had a problem with a mass influx of SO users. This, I think, is due to the nature of the topic. TeX is one of those things that if you know what it is, you probably know enough to participate on the site and have something to say there. The only real annoyances have been those that don't know anything about the SE sites at all and try to start discussions. But on a mathematics site, the problem is that there are people on SO who think that they know what mathematics is but really have no clue whatsoever.

So it's a balance: will the ease of access from SO balance against the ease of migration to MSE? Actually, I think it might balance in our favour. If you search on SO for "MathOverflow" then you'll see that we're already regarded as a bunch of elitist snobs who close any "reasonable question" in no-time flat. So I doubt that there will be a large influx of people, and if we can keep our "hot questions" off that list then there will be little opportunity for people to "stumble upon" MO by chance.

To respond on you response to Qiaochu. The SO community might be a democracy but it is a democracy within the confines of SO. The maintainers of SE may bow or may not bow to the whim of their largest community, but as a whole it seems to me that SE is the whole network while SO is just one big community. Its say in matters of what should or should not be with other communities seems to me as quite a peculiar way to run a network of communities.

I am not familiar with how things run, but from the impression I have so far (both from math.SE as well from Anton's and SE's owners in this discussion) this is not how things are done.

to ask and have answered certain questions of interest to research mathematicians.

That's what I originally thought, but now I would phrase it more in the form:

to ask and have answered certain questions of that research mathematicians could encounter when doing their mathematics.

It's a bit more awkward, but the point is that the mathematics has to be of a certain type and level. If I've phrased it correctly, it also doesn't mean that the person asking or answering has to be a "professional mathematician". Just that they play the role whilst on the site.

I also believe that this is more in line with Anton's vision. I don't want to put words in to his mouth, but that's certainly the impression I have after debating this issue with him many times on this site.

To Qiaochu:

SE has no intention to ever merge the two sites. Everyone, please stop bringing this up. It is really, really, really, really not an issue.

SE might not, but there is a sizeable (or at least very vocal) segment of the SO community that does not like "two level" sites. Since SE is meant to be a democracy, the fact that the "overlords" don't want to merge the sites doesn't mean that it wouldn't happen. I, personally, doubt that it would happen, but I'd like safeguards to ensure that even the question of it is ridiculous. As you are a moderator on an SE site, you can look in the moderator chat room and search for "mathoverflow", "mathematics", and "mathematicians" to get a sense of the comments that I've seen that lead me to this opinion.

I'd also strengthen Noah Snyder's comment:

This is very much a side point, but if we end up migrating I do not think that a 2/3rds vote among the moderators is an acceptable option for how to "break the glass." The reason being that there's no way to stop a future bearded EvilSE from just changing all the moderators one day.

I think that I'd go for 2/3rds vote amongst the current moderators, or a successor team appointed by them for precisely this purpose. After all, if things deteriorate to such an extent that we're worried about SE replacing moderators to swing the vote their way, they could just replace the lot!

What I mean in particular is: if a year down the road, say, 90 percent of regular MO users also have an M.SE acount since there was so much transfer and all the soft-questions got moved there (as suggested above), won't it then be hard to argue why the sites need to stay independent, eg, in case in the course of some general stackexchange network cleanup a merge should be suggested?

SE has no intention to ever merge the two sites. Everyone, please stop bringing this up. It is really, really, really, really not an issue.

Questions like common misbeliefs, counter examples, urban legends and refereeing papers all seem much more at home compared to math.SE.

I would say that misbeliefs and counterexamples (unless made more specific) are both fine for math.SE. Urban legends could go either way. Refereeing papers is definitely an MO question.

Questions like common misbeliefs, counter examples, urban legends and refereeing papers all seem much more at home compared to math.SE.

I do share Qiaochu's expectation to see more MO users joining MSE. I think that this site has a lot to offer from the point of view of writing answers.

(2) We crack down on soft questions, and migrate anything that could plausibly fit in MSE to MSE, so visitors below a certain level of knowledge never see anything in MO that they understand or enjoy. This would work, but it would make me sad, since it would remove some of the content I understand and enjoy.

I don't see the problem here. Migrated content is not removed content: you can easily associate your account with math.SE and read soft questions there! I fully expect this to start happening if we migrate.

1) Empowering moderators to block questions from appearing on the multicollider (see http://meta.stackoverflow.com/questions/99765/mods-should-be-able-to-suppress-questions-from-the-multicollider)

2) Automatically stopping all MO questions tagged "soft question" from appearing on the multicollider.

Unlike the 100 point bonus, this is entirely behind the scenes and would not interfere with user experience. As such, it seems like the sort of thing that SE would be more likely to agree to.

An influx of votes seems bizarre to me. Most people's reaction to a typical good MO question is to chuckle at its absurdity. It would take a very weird sort of person to show up and start randomly voting up questions or answers that they don't understand. I just don't think we're going to get that many people like that.

My impression from following SE stuff is that this sort of basic user experience issue is exactly the sort of thing that they'd be least likely to want to budge on. They really really want the experience of people coming to a new SE site to be uniform. They don't want different thresholds for powers at different sites. They want people to be able to move to a new site and know how the mechanics work. (Of course, this is because they want users to be able to move from site to site, while we want to stop users from moving to our site...)

I feel that the 101 thing is more of a red herring. If we, the regular users, are aware of the issue and make sure that we keep an eye on the "new posts by new users" list (which is a feature of SE2.0) then we can leap on these people (gently) and make them aware that MO is only for snobs. Sorry, I mean "mathematicians".

I really don't want us to try rolling our own site. Alex's point that big programming projects run by mathematicians don't do well in the long run is very important. (Of course, I'm not objective here, as it's very much against my own interest to have Scott involved in large programming projects.)

In terms of negotiating points, the one thing I would really want is a top of the screen popup (like the "we just linked your accounts" popup) for users coming from the SE network who haven't used MO specifically before which says something close to "This site is for research-level mathematics only. For other math questions please go to math.SE."

I think a message like the above would be more effective, less annoying, less confusing, and more acceptable to the SE network than trying to restrict the initial rep of outside SE network users.

(In the long-run, what I really think would be nice would be for there to be a "SE academy" subnetwork containing the sites like MO and cstheory. I think SE is right not to want too much variation in how the individual sites work, but I think that having a well-marked subnetwork for research-level sites might be more acceptable for everyone.)

Especially in light of how protecting a question, a feature of SE2 allowing only users with "enough" reputation to post, therefore blocking the comments and answers by newcomers which may be problematic, is useless against other SE users (since "enough" is more than 10).

1. As Pete L. Clark worried before, and I very much agree with this, the SE engine changes the text automatically. This is not a good thing. If MO joins the SE network, I believe that we must insist that automatic edits will be turned off. I have to say that I am also upset about the fact the changes are not public. I agree that the best thing is to let users run along, and if they don't fall and scrape their knees they will never know better.
   Personally, I feel cheated every time the SE team does something like that. Even if the change is for the better, I believe that at least hosting some tracker that we can know what issues they have updated/working on is a welcomed thing. I do not think, however, that the lack of transparency is a reason to vote again this migration. Just something that I find disturbing; on the other hand the automatic edits are much more severe as a whole.

This has gone for long enough, I just have to say one last thing (and if you read this far, you're a braver person than I am), I am not against this idea. I am even somewhat in favour (although not 100% sure it is necessary right now). Change is good if done properly. I have a firm belief that MO (and by extension math.SE) is a rather unique community that does not fit into the "usual" Q&A mold of hobbyist sites, and if we (by which I mean you) decide to migrate I think that it as important to consider the above points as much as it is important to ensure Mr. Atwood and others will not interfere with MO and meta.MO.

To give a bit about my "street cred", I am a grad student, I joined MO slightly over a year ago (and mostly sat quietly in the corner) and joined math.SE during the public beta stage (almost a year ago), as well the meta.m.SE about six months (or so?) since I became an active participant there as well.

Some issues to consider, that have been brought up before this post but seemed to have been neglected compared to other issues:

1. The incompleteness theorem can be abused in the notion of bugs and security holes: every big enough software has some huge bug that occurs once every five years when the planets align and the moon is in the scorpio. All network-interfaced software which are large enough have immense security holes which, when found, will let the smarter user go through them like lye through fat - in the most corrosive manner you can imagine.
   By this, of course, I mean that unattended software is bound to fail eventually, and the bigger they are they harder they fall. This is due to server updates, or possible leaks in underlying software (php? perl? common lisp?) or even bugs in the code that is the SE engine. While this does not pose an immediate threat, eventually it will become impossible to fix, and one disgruntled hothead user with the right kind of knowledge can in fact wreak havoc (at least for a while). Of course this is a very catastrophic scenario that is not very likely to occur, but isn't the probably of a tsunami washing parts of a city just as unlikely - but just as devastating as we saw twice during the past decade?
   Furthermore, as Anton said, usenet is still perfectly fine. Most of the users nowadays don't go there anymore. It is a dying medium. Platforms that stop attracting users whither away, and if this community wants to thrive for an indefinite amount of time, it seems like a good idea to upgrade at some point.

2. While having different reputation for questions and answers is perfectly fine in sites like SO, it seems to me that MO is a whole other thing. Mathematicians appreciate questions, good questions, questions that are phrased in an interesting way and leading you to think. This makes good questions seem like an excellent source, equal (to some extent) to good answers. While it is clear that answers are the goal, I would hate to see a site which is based on questions of high level being dropped of this status.

3. The connection with the rest of the SE network seems somewhat troublesome to me, it seems that no clear answer about whether or not it is possible to have the site seem disconnected from the SE network. That means no ads (and perhaps as importantly no ads of MO in other SE sites), and even more so ditch the 101 reputation for newcomers from the SE-land.
   The latter seems to me to be important for a main reason, not the comments nor the voting and not the CW editing. It is because a reputation of 101 points gives you a voice in a community that you did nothing to earn its approval for voicing. I still remember how hard it was for me to get my first 50 reputation points on this site (something that some might have shared and other might not). I was just starting my way in my grad studying. I would hate to think about people from other SE sites, which effortlessly given the permission to speak where they might not be wanted. It was pointed out that this community is run by mathematicians. Not by hobbyists, nor students which may or may not be qualified in either moderating or mathematics. This is a website with a very strict framework, in which you can do pretty much everything you want as long as you remember "to put on some pants" (i.e. play nice with others, and stay within a reasonable epsilon from the norms of the website).

(ctd)

The meaning of "internet thing" escapes me. Do you just mean it appears on a website? If so, I don't see how moving to another site makes it "more" or "less" of an internet thing.

Well, take Twitter, for example. My impression (and it is only an impression) is that senior mathematicians consider such things faddish, a product of a modern obsession with social applications, and overall a waste of time. It is possible (and I am pretty sure I saw a related point raised earlier, which is why I mention it) that SE 2.0 will seem more like such things. Perhaps this is a non-issue, which would be fine with me.

The only concern I can think of at the moment that wouldn't be alleviated by keeping this meta is the concern that senior mathematicians might be put off by the association with the SE 2.0 network. But this seems like more of a mild temporary setback to me than anything else. MO was an "internet thing" before it became important, and MO 2.0 will perhaps be slightly more of an "internet thing," but as long as everything keeps running smoothly it shouldn't take much more effort to get used to. Perhaps some brave volunteer should poll some such mathematicians? I don't think the people I have in mind frequent meta.

We're mathematicians busily writing papers, and trying to get jobs or tenure. Please bear with us, and try to see some of the downsides of joining the StackExchange network in this context.

We (the MO community at large) certainly shouldn't take lightly the commitment of those of you who have created MO and keep it running. If moving to SE2.0 eases that burden, and you would like to do it, then that reason alone would be enough to convince me.

I do also find Tyler Lawson's first point very persuasive.

That said, I think Scott's list of requirements makes a great starting position for a negotiation.

• As Qiaochu says above, we are already hosted by StackExchange, and nearly every worry about calamity in the future because of a change in their business plan applies at least as much to our current situation.

• the StackExchange software is really good, far better than we can expect to find elsewhere or produce ourselves.

• Fog Creek does a fantastic job of handling all the heavy lifting, running webservers, backing up data, dealing with network outages, fixing critical bugs, etc. This is a gigantic advantage for us, and we should be very very pleased to be getting all this for free.

As such, if we eventually decide to migrate, I (and I think the other moderators) will be willing to cash in some of the community's goodwill, saying something along the lines of:

"Running mathoverflow without this support from Stack Exchange might not be viable. We're mathematicians busily writing papers, and trying to get jobs or tenure. Please bear with us, and try to see some of the downsides of joining the StackExchange network in this context."

I'm very happy to help write code that opens the door to future migration away from SE, and happy to put in effort finding funding for or managing any development work that we decide is necessary for such a migration. Nevertheless, it's important to remember the scale of the advantage we gain by piggybacking on Stack Exchange.

I don't think comment everywhere is that big an issue. If someone misbehaves in a comment, you can always flag it for moderator attention. It is pretty hard for moderators to not see that in the SE2 tools.

For votes: I agree partly. Insofar as meta is concerned, if we keep meta on this platform, that would not be a problem. Perhaps we can arrange to have a different rep threshold for voting? But I don't think that would be that much of an issue.

Community wiki edits: that could be a big issue.

If we identify specific things that we don't like, we can try to negotiate with SE to see if they can be changed to our liking, right?

@Qiaochu Yuan, Your argument about the necessity of real-time intervention by SE employees is not convincing for me. As Joel noted, SE employees can't monitor the sites all the time either. I am sure at least one of the MO moderators will visit the site in a reasonable time and waiting a few hours is not a big deal. If it is a very urgent matter then MO moderators can be contacted using other methods like email (there is also flagging by ordinary users can solve many cases without even MO moderator intervention). Discussing rare situations is not helpful, it is not difficult to give similar "what if" arguments in the other direction. The point here is to make it clear that intervention by SE employees without approval of MO moderators is not acceptable and that is the rule, to break the rule they should have very convincing reasons that it was necessary to break the rule in that particular case.

@WillieWong, The main problem I see with the initial 101 rep for users coming from other SE sites is that it gives them a say (i.e. vote, comment, ...) without being part of the MO community, it can bias the votes on the main site, but more importantly it can bias the votes on the meta (and this does happen, someone links to a discussion on the meta in another site and suddenly we see a huge increase of views and votes by users who have never visited the site).

One point that has still not been addressed is whether being attached to the SE network would make the website less attractive in the eyes of mathematicians. As long as the current policies on meta.SO that many of us here appear not to agree with (i.e. closure/deletion/locking of complaining posts) are not extended to meta.MO (which seems a safe bet since Joel Spolsky has assured us that MO would remain self-moderating), I myself don't have any objections.

Incidentally, at some point or another, I have looked at the MO profiles at most (all?) of the mathematician participants in this thread, and it seems that, on average, we tend to be much younger than Jane Q. Mathematician (or even Jane Q. MOwer). I think it would be very interesting to hear, in addition, the views of those who have been in mathematics for, say, 20 years or longer.

He thus has a financial interest in defending the corporation and convincing us to migrate. I usually find him pretty level-headed, and I was wondering why he was so vehement about defending SE on this thread. I thus did a little google search and learned this.

My internship is for a fixed period of time, and my pay is also fixed and does not depend on the success of the SE network.

Let me also clarify this "vehemence" issue. I believe I am the only MO user in this thread who has actually spoken to an SE employee in person (though perhaps Anton has as well). It's therefore easier for me to see them as actual people as opposed to faceless corporate automatons. The "vehemence" you observe is a result of me trying to inject a strong opposing voice into this discussion because it is all too easy to demonize people you have never met. (Some cognitive biases I think are relevant to this discussion: the fundamental attribution error, confirmation bias, and the availability heuristic.)

I'll also repeat that I am not 100% in favor of migration. There are still things I don't understand about the long-term viability of doing nothing, as well as the viability of OSQA, that I would like to understand before I "vote" for one thing or another. Again, if we reject migration I just want to make sure we're doing it because we actually think it would be a bad idea and not because we're so scared of non-mathematicians that we would rather do anything than be forced to work with them.