Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.


    UPDATE: There's now a blog post explaining what the options are for SE 1.0 sites. We won't be expected to make a decision for at least three months, but the decision will be final once we make it: "Migration of sites will occur all at once. This is a one-time event, not an on-going process." Basically, the options are

    1. Remain an SE 1.0 site indefinitely at zero cost. There will be no updates to the software except urgent bug fixes.
    2. Migrate to the SE 2.0 software under the following conditions: "Sites are owned by Stack Overflow Inc. There is no co-ownership of sites, commercial relationships, or revenue sharing. SE 2.0 sites are run by the community. We will make every effort to accommodate former site owners’ wishes to moderate the early site but no special relationships, like appointing someone Administrator of the site for life, will be considered."

    They also say "These issues will all be discussed with site owners and the specifics spelled out on a case-by-case basis." I'll post here if I get any more news. Right now, remaining an SE 1.0 site seems like the only reasonable option since migrating doesn't fulfill any of the requirements I had in mind for migration. It doesn't allow us to retain control of the domain name (in the event that we want to change platform), it doesn't give us access to full database dumps (ditto), it doesn't guarantee that MO remain ad-free (or even have control over ads), and it doesn't give me any control over custom javascript (or css or static pages).

    • CommentAuthorHarry Gindi
    • CommentTimeJun 1st 2010 edited


    1.) Maybe they could make MO a special case where we could do a one-time payment to upgrade to the new SE2.0? This would be optimal, because it would solve fog-creek's problem of incentives, and it allows us to maintain autonomy. Raising the money to do it would be incredibly easy, given the number of generous offers for funding from the users here.

    2.) The thing is, this site is essentially community-run, and I can't think of anybody who doesn't want you to have the administrator powers that you currently have. What would be an alright compromise is if they made you and Scott full administrators with backend access (to dumps, the database, the javascript and CSS, etc.) for MO but hosted it themselves. Someone should suggest to fog-creek that a new type of administrator (something like an "overlord") with these priviliges should be created for all SE2.0 sites.

    Part of the reason that MO is so good is that we have a very strong administration team that handles both high and low level problems. We should appeal to the fact that administrators can only administer a site effectively if they participate in it. That said, having low-level administration done by fog-creek would be frustrating to them and to us, because they don't understand the issues of the community, and we won't have administrators who can competently deal with problems.

    3.) Perhaps we could do something like administrator elections (which Anton and Scott would certainly win), which would mitigate the problem of having SO appoint someone administrator for life.

    Yeah, that's too bad. But you're certainly right Anton that those issues are non-negotiable.
    • CommentAuthorMariano
    • CommentTimeJun 1st 2010

    It turned out worse than what a sensible pessimist would have imagined, sadly :/


    Yeah, there is no way we should give up full database dumps and the ability to install outside scripts.

    I think we should be planning for the possibility that StackExchange goes bankrupt. I have no understanding of where their income stream is going to come from, especially after imposing conditions like this. The strange thing is that we were willing to pay them for this service, but they don't want the money!


    We could always try bribing them!


    I think we should be planning for the possibility that StackExchange goes bankrupt.

    Fortunately, I don't think we have to worry about that seriously (at least in the short term). Stack Overflow Inc has something to the tune of $6M in venture capital. From what I understand (I have no inside knowledge), they're deliberately putting off deciding exactly what the business model is. They want to make a giant network and then worry about how to monetize it. I actually think this makes sense for them.

    It turned out worse than what a sensible pessimist would have imagined, sadly :/

    I disagree. The sensible pessimist would have imagined that SE 2.0 migration terms are unfavorable (and cannot be negotiated) and that SE 1.0 will completely disappear. Having the option of leaving things running exactly as they are is a little annoying (I was really looking forward to getting some tasty new features), but overall is pretty decent. Not only that, but there may be some wiggle room since they've said they'll handle things on a case-by-case basis.

    @Harry (ideas post): The problem is that they're not doing SE 2.0 out of greed or spite (I think). Given that they're trying to build a huge network, it makes sense for them to impose a lot of control on SE 2.0 sites. If this is correct, then it doesn't make sense for them to go out of their way to code in special exceptions for us. If they can easily accommodate MO as a special case, maybe they will, but it might turn out that it's too much work.

    Some negative aspects of migration would be tolerable if absolutely necessary. If SE 2.0 sites don't have any notion of "administrator", it will mostly just be a big pain to update custom javascript, html, or css because I'll have to do it by email. If additionally they insist on "reviewing" and sometimes rejecting such changes, I would be pretty annoyed (I'd probably count that as intolerable). I think moderator elections are a good idea (I've been meaning to post about that ... give me another week). However, some things are absolutely non-negotiable: no ads, full database dumps, and they cannot have the domain. These things don't sound like they should be a huge hassle for them to accommodate. I'll certainly keep everybody here posted when it comes time to negotiate what the terms of migration would be specifically for MO.

    • CommentAuthorHarry Gindi
    • CommentTimeJun 1st 2010 edited

    Well, if there are going to be moderator elections, I nominate Andrew Stacey, Yemon Choi, François G. Dorais, Mariano Suárez-Alvarez, Pete L. Clark, Scott Carnahan, and David Speyer.

    I nominate Andrew and Yemon more than the rest of the gang because Andrew lives in the cold arctic north, and Yemon lives in jolly old England. That is, it's better if we have moderators in more diverse time-zones.

    At the moment, we have Anton, Scott, and David in PST (GMT - 8), Ben Webster in EST (GMT - 5). If we add Yemon and Andrew, we'll have moderators in roughly (GMT + 0) and (GMT + 2). In addition to them, we could use a moderator from Australia GMT+8, somewhere in east asia (GMT + 4-5), and probably somebody in New Zealand.


    Phew, slow down Harry :-)


    So what does it mean if we stay on the 1.0? Is there a possibility that we (and by we I mean not me, since I know nothing about computers) could update the site ourselves? That is to say, will we have access to the code? In that case it could be good to use some of the money we were willing to pay StackExchange to pay someone to work on updates and such.


    Although SE 2.0 is technically free, I think there is still potential for the NSF (or equivalent) to support MO. I think it would be reasonable to pressure SE people for a "research exception" to grant certain privileges that might be necessary to secure public funding.

    • Full database dumps for research purposes. (Otherwise it would be hard to accurately measure the broader impact of MO.)
    • The PI(s) should be granted administrative privileges without elections and such.
    • Other useful stuff...

    Just a quick point of correction: although I am currently visiting England on a 4-week research visit, I actually live and work in Canada. So you might want to think about other suggestions for mods in GMT-ish zones...


    While I appreciate the thought, I am going to preemptively decline my nomination as a moderator. I probably spend too much time on MO as it is.


    @fgdorias The censored dumps would probably be fine for research; the reason we need full database dumps is for our own protection, if we need to switch to a different provider.


    I got an email today from Robert Cartaino (SE community coordinator) saying that there shouldn't be a problem migrating MO with only fairly superficial changes. He didn't give me any details (the details won't exist until the first SE 2.0 sites are up and they start figuring out how to do migration), but I have renewed hope that we'll get favorable terms for migration.

    There is one thing that I should point out that would be a major change if we migrate to SE 2.0: we will be part of the SE network. To ease the frustrations of using a new SE site, I think they're going to start people off with 100 rep if they have a reputable account on another site. This means that we might get a surge of rep 100 users, which could be a bit of a pain, but I don't think it will be a big deal. If people think this is likely to be a problem, please say so.

    I agree with you that a surge of rep 100 users will likely be an annoyance (especially for the first week) but is not a major problem. And on the flipside I'll be happy to start with rep 100 on other SE sites.

    What??? It's taken me weeks to get any decent rep on stackoverflow and now all you lot will get it for free? I want my money back.


    Meta enthusiasts may want to check out, the proposal site for new StackExchange 2.0s. Mathematicians may be especially interested in the LaTeX site proposal.


    Another proposed site on area51 that we should keep an eye on (if only to redirect people there once it's up and running) is Statistical Analysis.


    I like the LaTeX site proposal, though I'm finding it difficult to come up with a truly awful question for it (apart from the obvious one, which has already been asked)![1]

    Others to keep an eye on: and (both justify their existence by contrast to MO!).

    The full list of proposals is rather ... interesting!

    [1] For those who don't know, a site has to garner a short list of "good example" and "bad example" questions before it can go to The Next Level. The LaTeX proposal has plenty of good ones, but not many bad ones yet.

    My first reaction is that having these additional sites on mathematics. calculus and statistics will improve our site.

    I noted however that the mathematics site (it really needs to be called "MathUnderflow", doesn't it?) also wishes to discourage blatant homework questions. This is a continuing issue for us and I can only predict it being a much bigger issue for MU: the best way to be sure that a question is not homework is that it's too hard / advanced to make a reasonable homework question. I wonder what will happen with this.