They didn't capriciously shitcan the OS X port, there isn't anyone in their community that wants to wrestle with the Mac port. That kind of thing isn't what causes a software project to fork. Forks happen when someone says "macs are for weenies" and removes support for a platform while there are still people working on that code. When nobody wants to work on it, nobody wants to work on it.
And if you were willing to buy a couple developers a Mac, they might switch. It doesn't really matter what OS the users are running, they're not being supported financially by the project and have no incentive to scratch an itch that isn't theirs. My guess is that it will languish for a while and then someone will decide to come along and update it in a while. But it won't be a forking event.
Here you go. A port-o-rotary for $200. They provide full source and schematics. You can even buy a 6000mAh battery to run the thing for weeks and you don't have to deal with any PDA functionality. Any more complaints?
Radios are expensive. The only reason phones are cheap is because they're heavily subsidized or because they're a simple little phone produced a million at a time from a small handful of highly-integrated mixed analog/digital ASICs. "Open-source" devices are small-run devices with hopelessly obsolete radio hardware because it's all they can get documentation for and manufacturers aren't looking to release their secret sauce to just anybody.
And on top of all of this, most of the open-source types are desktop or server programmers. On the desktop, you don't have to think about low-power code. Everything changes when you're running off a battery. There just isn't the expertise there (yet). Having said all this, I love my rooted T-mobile G1. I built a scratchbox environment for it and ported a few important CLI tools and it's now perfectly capable of being all the pocket Linux machine I need and it's not very difficult getting Debian running on top of the Android environment.
The bogosity meter just pegged.