Has the timing and the sequence convinced me that nose down is an error no matter what else the computer is telling me?
gleaf, you don't need a computer to tell you that, while flying an assigned altitude in straight and level condition in cruise at flight level 350, establishing a nose up pitch of 12-15 degrees is an error.
Returning the nose to a different pitch should help you with the altitude change, and may (depending on how you do it ) assist in your energy management and performance correction.
The altitude change tied to pitch up and airspeed loss (changing kinetic energy to potential energy) seems to have been lost track of despite, or due to, various warnings and alarms.
Pilot not flying is recorded as trying to get the pilot flying to stop climbing.
Why THAT was a failed exercise remains less than certain to me, as I am not familiar enough with climate and culture both in Air France, the company, and that particular cockpit.
Non-digital issues.
Don't disagree that while trying to catch up, any number of stimuli were present that would aid and abet in the misdiagnosis of "what's it doing now" once the crew were behind the aircraft.
BEA's discussion of that in the final report is pretty good, though some would prefer some more depth and detail.