Please. 250 paid employees is nothing, considering the number of projects they are working on, the number of platforms and localizations they target, and the speed of innovation they're putting out. Also, they have to do a lot of their own toolchain and infrastructure, which some of those larger companies have entire teams for.
I visited the main offices in 2008 - it was a pretty spartan affair, nice but not luxurious. I was really impressed with everyone I met. They take this "we're here to improve the web" stuff seriously. I'm not going to say that there's no bloat, that might be expected given the comfortable income they make from the partnership with Google. But every Mozilla employee I know is pretty awesome and working on something that you probably care about.
Huh? They don't have that many projects, they only officially support 3 platforms, most of the localizations are done by volunteers, and there isn't a whole lot of innovation going on.
The only real project is Firefox. There's some support projects like addons.mozilla.org, but it's fairly badly run and slow moving. Most Mozilla Labs projects are dead, Jetpack seems to be only one with legs. Weave is taking forever and nowhere near production ready, and Ubiquity is a power user feature than 95% of Firefox users won't use.
Most of the new stuff they're talking about is catching up to Chrome, catching up in the mobile space, or already speced out HTML5 features. Hardly any substantial innovation.
They've got a lot of fat really. The actual browser engineers on the ground do want to improve the web, but they've got a lot of ancillary folks who don't really do much, and a bunch of cultural baggage which prevents them from being an effective open source project. It's still insanely hard to get any feedback on feature suggestions, they still mostly do their own thing and largely ignore the external developer community.
How about a conjecture based one zero evidence or reasoning at all: If you're on top of an open-source project that has started making millions of dollars, wouldn't it be a good thing to do it hire a whole bunch of your programmer friends to hack on open-source projects that interest them, doing the OS community good?