In my opinion, with a couple of phase 0 devices sitting in their fancy lock-tite boxes gathering dust - while the dev workbench is littered with current-target

In my opinion, with a couple of phase 0 devices sitting in their fancy lock-tite boxes gathering dust - while the dev workbench is littered with current-target mobile devices far in advance - the only difference is that the hardware team behind OpenMoko ran out of gumption/funding/$$$. Hardware is much harder to do, the more you screw it up. Designs which don't work - i.e. basic power management - will cripple early-stage hardware development, and its one of the reasons hardware, is well .. hard. OpenMoko didn't screw up, open-source-wise, no sir. I think you are mistaken in this claim. Where they screwed up is not getting the hardware really nailed, such that it warranted actual production in the 10's of 1000's of units. It had to work, and you have to ship in the 5x'es before you gain any traction with a new hardware platform. OpenMoko was many things (I still have 3) .. but it wasn't a failure of open source software. Without that stack, the thing wouldn't have even gotten off the ground. It didn't reach orbit, because of that 5x traction ..