I see a number of statements here - late in the thread - that suggest both a lack of review of previous posts, but also a number of scenarios not really supported by the information we have available.
The latest publicly available reports seem to implicate the right thrust lever not responding to an A/T-commanded reduction of thrust resulting in a highly assymetric thrust condition. The role of weather, traffic and such seem to factor mainly as crew distractions rather than causal factors.
The A/P system (which operates in concert but with, but independently of, the A/T system) fuctioned in a manner consistent with its expected operation; it attempted to fly the aircraft along the commanded flight path until it reached its limits, at which point the software recognized it could not properly manage the flight path and disconnected. Unfortunately, if the A/P system has masked the malfunction of the A/T system as seems to have happened here, the disconnect immediately results in a rapid, chaotic departure from controlled flight.
When we watch a horror movie where the camera angle allows us to see the monster approaching an unaware victim, we do not blame the victim for not predicting that the monster would hide at THAT particular location, or did not figure out that "that particular noise" would result in a direct threat to their life. Similarly, we should be more aware of the human factors that are becoming more apparent with this accident.
Just like the horror-movie victim that could have avoided being killed by any number of different decisions, any flight crew can avoid the vast majority of these types of situations through strict discipline to procedure which, in turn, results in a high level of awareness and good decision-making. Our realization that this apparently did not happen, while instructive to rest of us, should be tempered with the reality that operating complex machines in highly dynamic situations on a routine basis results in "hours of boredom punctuated by moments of terror".
Thank you for your great post.
From a safety point of view, I am wondering how much mitigation exists against the probable A/T malfunction scenario you describe. In this safety context the accident can be maybe be split in multiple parts like this:
1) Design decisions that can produce a fast and dangerous event in case of a single failure. The safety analysis of that failure and the resulting risk mitigation that was produced.
2) The information concerning that failure risk and the training of the appropriate mitigation.
3) The mitigation of the failure in flight before it produce the fast and dangerous event.
4) The mitigation of the failure in flight after if produce the fast and dangerous event.
While the point 1 and 2 can span a over long period of time, the pilots faced point 3 and 4 in a very limited time frame. My understanding so far is that point 3 time frame was a few minutes under near normal workload. The mitigation at that point was expected by scanning the available information and then implementing the right procedure. For some yet unknown reason that was sadly not what happened. The final report is expected to find what contributed to this. But humans are not 100% reliable, especially in case of rare event. This is why critical safety require multiple lines of mitigation. So what mitigation exists in that failure scenario is case the first mitigation did not success and that the malfunction finally produce a rare fast and dangerous event of the point 4? Was that possibility analysed at the design stage? Was an appropriate training defined and implemented to survive a such fast and dangerous event?