BitKeeper user here; AFAIK BitMover got $0 for hosting Linux though they did get a lot of fame. And their biggest problem apparently isn't that BitKeeper "turned into a LinuxKeeper", but rather free BitKeeper clones like git and hg (the first of which was originally created for Linux). So, just curious about your interpretation of your Cokecamp analogy; to me their problems (to the extent that they experience problems) do seem very much related to losing their biggest, most visible user, but in a different sort of way.
That's true, but it's something of a self-inflicted wound.
The release of git and hg was a direct response to McVoy yanking BitKeeper away (withdrawing the free-beer licensing terms that had previously allowed gratis use for free/open source projects); both were initially announced on the Linux-kernel mailing list as responses to McVoy's move, and each had had only a few weeks' worth of work at that point.
Before that, there had been a good deal of grumbling about the ever-shifting terms of the "don't piss off Larry license", whose increasingly draconian terms were designed to prevent the development of competitive software. (The last available version purported to bind all users to a non-compete preventing them from contributing to another SCM; I'm not sure California law would have allowed McVoy to enforce a non-compete on his own employees!) However, so long as these difficulties remained theoretical, there was not, in fact, much serious work going on to develop such competition. But when Larry revoked the license, effectively forcing the kernel developers to come up with an alternative damn quick, they did --- and damn quick.
So, the clones aren't something that the customers did when they got dissatisfied with BitMover; they're something that happened after BitMover chose, on their own, to end the relationship. Which makes them, as I said, a self-inflicted wound, that BitMover could have avoided through better customer relations management.
My impression is that BitKeeper was designed with an eye towards Linus's own habits, and my point was that BitKeeper suffered as a result. As best I can reconstruct I seem to have gotten this impression from this article:
"From the beginning, Bitkeeper has been aimed at making it easier to do development the way Linux kernel development is done. As Larry McVoy put it: 'BK makes it really easy to do what Linus is doing.'"
If you follow the links in that article, Larry McVoy doesn't really say what he's quoted as saying there, though he does make similar assertions.
BitKeeper certainly made no money on supporting Linux but I think the idea, on some level, must have been to get the software out there being used by serious programmers who would then love it, buy it, and evangelize it. In the driest sense, it was not a success. But in a wider sense in backfired so thoroughly that the idea of starting a commercial VCS project at this point is laughably absurd. Several things McVoy said turned out to be wrong, even though they must have sounded plausible at the time. I'm particularly pointing to this one:
"All you people trying to copy BK are just shooting yourself in the foot unless you can come up with a solution that Linus will use in the short term. And nobody but an idiot believes that is possible."
So, it probably isn't the best analogy, but on the other hand it was the first one that came to mind. :) I'd be curious to know what you think about this, and more importantly what your opinion of BK is today and how it compares after having a decade to mature.
