SuseJ772 wrote:I think something else should be considered as well. The premise is that at Flaps 1 the MCAS is always deactivated. But what if it doesn’t always deactivate it. What if that is part of the problem.
As a programmer you are always faced with how things are suppose to work versus how they actually do work. And it usually isn’t as simple as Flaps 1 = off. Because of course this doesn’t happen with every or even most flights. But you combine a previously unconsidered combinations of parameters (ie Flaps 1, AoA error, and [insert 3rd condition]), and it does something you don’t expect (ie not cutting off MCAS even with Flaps 1).
Just a thought
SuseJ722, I was a programmer too. And at one time it involved a flight computer. People don’t understand how complex the logic usually is. And we here don’t know the logic in the software or even all the inputs. I imagine the algorithm that is actually programmed is orders of magnitude more complex than what is alluded to here. All I know is we have to wait for an investigation to see what happened.
The comment by a previous poster about AI since the 80s or whatever it was reminds me of a comment I had once from an outside manager commenting on my software design during a review. I was praised later on on how I was able to keep my mouth shut.