My question is: Why, having done your research (and seen for yourself the enormous fuckup he made of the Pandora and to an extent the iCP1) did you back him?
My assessment stands, I believe.
I didn't order an iCP2, but I don't think the people who did were foolish at all. Just because Pandora had a lot of problems didn't mean iCP2 was going to be an automatic failure, and research and analysis of the situation didn't make it look that unreasonable. If a fair project management style risk assessment was made it'd look very different. Think about it:
a) iCP2 was much simpler than Pandora, especially from a software perspective
B) A LOT of problems that happened to Pandora wouldn't have applied to iCP2. KS funding prevented the many funding problems Pandora had. They were going with proven (as in, people he had experience with) and relatively local manufacturers, so no problems with vendors like CC or the headaches communicating with the Chinese case manufacturers. The nubs were already developed and there were no LED cables to worry about. Hardware-wise, there was nothing on the board that was new to them if you don't count the lighting.
c) They'd already done a device that wasn't that much different
d) The stuff shown in the video at least looked more mature than Pandora was when preorders were first made
e) Pandoras were actually eventually delivered to a majority of the preorderers, if quite late - some would want to back iCP2 even if it was very late, there's nothing intrinsically foolish about that if they're okay with it
f) Much of what we know now about Craig's questionable practices were not as widely discussed then
g) None of the LED nonsense started until after the funding was completed
The only huge red flag at the time of the KS was the ridiculous estimated completion time. Honestly I'm pretty surprised, you could even say impressed at how badly this failed. I would certainly find it foolish to back anything Craig does now
since he's shown how much he can screw up what would appear to be a pretty straightforward project.