The post by Jeff seems to point in the same direction -- it's not hard to just code, it's hard to code something that improves the user experience.

I don't think that the authors of the SE software will ever do it completely right. "Completely right" for them probably means something completely different to "completely right" for us.

Not really, because the solution people have repeatedly asked them, and the thing I had in mind, is to allow writing plugins. This will be a safe and simple way to extend Stack Exchange functionality and then it will be possible to add better jsMath/talk/interaction abilities using just volunteer resources.

Absolutely right. I'd better make all my theorems closed source from now on. I don't want to get hassled when I do "upgrades". If you want to construct a Dirac operator on an infinite dimensional manifold then you have to ask me to do it for you - I don't want just anybody doing it and then suing me because they didn't do it correctly.

Well, if you were in business of selling people the right to read papers about Dirac operators at $100/month with full refund guarantee :) ...

http://www.codinghorror.com/blog/archives/001284.html

The bit that I really don't understand is that given what he writes there, why doesn't he make the source open? He states that what makes SO a success is almost nothing to do with the software and everything to do with the experience and community. At the moment, MO has the community but I have to say that the experience is irritating. jsMath just looks awful, the lack of even a possibility of interaction is severely limiting (yes, yes, I know that this is by design but I say that for mathematicians this is bad design), ...

I don't think that the authors of the SE software will ever do it completely right. "Completely right" for them probably means something completely different to "completely right" for us.

Finally, thanks due to Ilya for posting the funniest thing so far:

Either way, I don't think they have any incentive to allow modifying the source code of Stack Exchange -- things would be so much likely to break across upgrades and its easier to prohibit modifications than be on call for lots of dissatisfied customers.

Absolutely right. I'd better make all my theorems closed source from now on. I don't want to get hassled when I do "upgrades". If you want to construct a Dirac operator on an infinite dimensional manifold then you have to ask me to do it for you - I don't want just anybody doing it and then suing me because they didn't do it correctly.

Either way, I don't think they have any incentive to allow modifying the source code of Stack Exchange -- things would be so much likely to break across upgrades and its easier to prohibit modifications than be on call for lots of dissatisfied customers.

But! At least, it's not their flagship product which runs on custom language invented by Joel.

The other interesting thing I learnt from that blog was that the Stackexchange software runs on an almost entirely Microsoft stack!

I've been involved a bit with Wikimania'06, and I think meetups can help the community a lot under right circumstances.

$260/month price bracket

This seems to be totally within a computer budget of most math departments. I've heard that public universities in California currently have funding problems because of a crisis though.

1. there's so little money involved (MO has zero revenue and tiny expenses),
2. Fog Creek hasn't actually decided what kinds of educational/non-profit discounts they will offer, if any, and
3. I feel like I'd actually have to put in a fair amount of time into figuring out how to make MO into some kind of official business entity.

So far, the cost of MO has been $0 since SE is still in beta (it will be at least two more months before the first bill comes). I think Ravi said he'd be willing to fund MO at $130/month (the lowest price bracket) for 5 years, which is basically forever, and he agreed to do that before we had anything concrete to show him, so I imagine it should be easy to get additional funding if we need to. Over the last month, we've gotten about 800,000 pageviews, so it's fairly likely that some time in the next year we'll break 1,000,000 pageviews per month, at which point we'll move from the $130/month price bracket to the $260/month price bracket. Given how level traffic has been over the last five weeks, I think we'll have a few months warning before that happens.

Since we'll probably eventually have to find some non-Ravi source of funding, let's make this thread a list of possible ways to reduce the gap between hosting costs and funding sources, should one arise. Here are some options, more or less in order of descending desirability:

1. We can apply for support from CMI, MSRI, or the NSF. Does somebody know how I would do this?
2. We can seek out more people like Ravi who can support MO from their research funds.
3. If Fog Creek offers educational/non-profit discounts, we might be eligible. I might have to do some paperwork.
4. We can set up a system for individual donations.
5. We can sell various MO gear, like the t-shirt I'm wearing in my profile picture. I'm not very excited about this option since it would probably end up eating into my time more than I'd like. By the way, you're welcome to make your own MO gear; I've got a handful of $6 t-shirts if you want one, but don't want to make your own.
6. We can sell advertisements. I'm not very keen on this option since I like the current purity of MO. I've already turned down a couple of offers to buy ad space (they weren't the sorts of ads mathematicians would want to see anyway).

But if that cannot continue, a way must be found to cover the cost. There are certain legal requirements to become a non-profit organization. Not making a profit is insufficient.