Craig has given us some mixed information about the GP2X's profitability. Part says the GP2X has not made GPH a profit. He has not said if each unit does. I'm sure GPH would have a lot more money to improve the GP2X if they weren't wasting money on this project though. That isn't to say they shouldn't produce a breakout device, but it should probably be executed differently.
First off, I don't see why it isn't just an attachment to the unit itself. Craig described something of this nature to me, with a keypad of sorts. Seems like a better idea-- a clip-on unit instead of some awkward box that needs to trail off the machine and have its own power! The device he described would have all the breakout dev board's capabilities incl. 3 USB ports. Seems like a pretty good idea to me. But I don't see that happening. I also heard talk of a replacement backplate to add BoB features. Not sure if it was to be from GPH or not, but that hasn't surfaced either.
The bigger issue I see is that GPH wasted all this money building a BoB that apparently still has engineering flaws (big ones) and now, instead of even fixing it, is rushing on to make a docking station of some sort. This is clearly a totally new PCB design and will cost even more money. Why doesn't GPH fix this negative pattern in their behavior by planning out one working device, testing a prototype, ensuring it works and then mass-producing it? This 'make something broken, then mass-produce it, then replace it later with something totally different' bit is not a step in the right direction.