PJ2
No disagreement here, only a philosophical muse relative to the topic you bring to the fore. Appreciating how dynamic and challenging flight can be, I see a trend toward solutions that involve further refinement of problems not solved by automation, but instead caused by them. Reliability is the fulcrum of any solution to an aerodynamic challenge.
Historically, extremely odd and cataclysmic events have shown up to shake the confidence one develops in systems in a lifetime of relying upon them. In BA038, we see one of these (actually two). I deplore the fascination with the flight crew's responses to these two simultaneous events. Their handling was superb, by any measure.
We are left with equipment failure, and evidently having a second engine is not enough. Who'd have predicted dual failure of these two powerplants in simultaneous fashion?
Delta survived a similar (?) occurrence, though the failure happened at altitude, affording a human solution to an engineering (?) problem. At some point, and to the relief of most of us, I would say, a better understanding of systems and piloting relating to how the two work together is needed.
447? Who needs an old fashioned horizon when there are three computers? Pitots? BA038? Who can entertain the "maybe" relative to thrust loss further away from the runway? Was Delta lucky in recovering the one instead of losing the second? As safe as it is, I don't believe stuffing the computer with solutions to oddball scenarios is the way to go. I'd say the pilot part of the process is just fine. What of the mechanical?