AF 447 report out

[areobat]

I wanted to post this a while back, but didn't have a chance to . . .

But as long as this thread is seeing activity again, there are two things I wanted suggest as things I see as factors in AF447 - and both of them are design related.

The first has been discussed many times (including just recently) before, and that is I think there needs to be some form of mechanical feedback between the control inputs (whether actual or simulated). People are inherently tactile beings and they very much - despite training to overcome it - depend on haptic feedback when the brain is processing higher level intellectual tasks. Look at touch keyboards, musicians, even surgeons using robotic instruments. While they can function without it, people work better with it.

The second may have been mentioned, but I might have missed it in the thousands of posts on this thread. And that is the way that the stall annunciator logic operates. It is pretty clear that the audible "stall, stall" means just that - you are stalling. But in the Airbus, the stall annunciation can be silenced by either recovering from the stall, or in the case of AF447, when the system loses key data inputs (the pitot tubes clogged), yet the system does indicate which.

While the actual cause for the silencing of the warning could have been deduced from various other indicators, in a rapidly changing and unusual system, I think it would be easy for the mind to get into the wrong feedback loop. Warning=on, I'm stalled. Warning=silent, I have recovered from stall. I think that is exactly what happened on AF447 (read the transcript of CVR).

If the system was designed something like, warning="stall", you have stalled; warning=silent, you have recovered and the system knows it because it is fully functional; warning="stall warning suppressed", you can no longer rely on the stall annuciator for the status of your aircraft, I think the outcome may have been different.

Just my thoughts.