I'm not a pilot but a retired aerospace engineer with a great deal of experience in spacecraft control systems. Reading this thread and the one on the A320 lost at Perpignan, would it be fair to say that:
- human beings do a fair job of flying an airliner and are usually highly reliable
- and so does an automated system and is also highly reliable
- in fact either provides competent piloting, but
- the problem arises at the man/machine interface when either human beings or automated systems intrude or are called on to intervene in the other's territory.
Glaringly obvious maybe, but since we cannot reengineer human beings (yet), it's the automated systems interface to humans that need closer examination. Recent incidents show many failures in this area.