Pissed off at work. The co accepted two deadlines for monday
when both were science experiments involving new hardware
or new settings. The new settings screwed everything up so
day 11-5:30 was a bust. Yesterday (Saturday) I did manage to
get a code variant cosmetically nice and running on 'safe'
hardware so the Pres of the Major Client can do a dog and pony
show on the road.

But today was a frustrating pain. I tried a couple of times on
different variants (the current one, and an older one sent to
UL). Figured out a simpler way to try it, by globally running
a certain sensor at the lower voltage we decided to run
(during certain operations) at lower voltage than normal.
That only revealed that everything was screwed with new settings
and it wasn't just a parameter change, it was a whole new
set of experiments, analysis, algorithms, and testing. Testing
takes a long time. But off the cuff changes aren't sufficient.
This requires systematic approach, understanding of the changes
induced, the engineering changes required.

Gotta give the kid his netbook. For now.