Man / RAT,
' flying as an art' , but art has no boundaries, no set format; we know what it is because we 'know what we like'.
Ian W,
expectation - 'the pilot can pick up the bag of bolts' . Yes.
The industry chooses to automate functions to mitigate less reliable human activity, yet when automation fails the industry then expects the 'less reliable human' to manage the situation and blames then when they cannot meet that expectation.
We are unlikely to progress safety until this line of thinking changes. Not that the industry is unsafe, but there is increasing need to maintain and improve the current standards as aviation expands, yet this expansion is driven by the same automation which the human is expected to manage in all circumstances. Round and round.
'On the loop' or 'out of the loop', like art depends on what you see. A backward, reminiscent view (I include my self in that age group) may see the flying loop, whereas today the issue is more a flight and systems management loop. The future has to include the human loop; not just the crew, but the interactions of design and certification ... the manufacturer, regulator, and investigator loop.
I fear that we will never catch up, never break out of the circular argument, because it will always be some loop or other.
The required safety activity could be described as 'loop management', but more practically, how to manage the unexpected. Yet this form of management is exactly what has been discussed under flight experience, and the current problems of low levels of experience, - manufacturer, regulator, operator, and finally the crew.
I still argue the need for experienced thinkers; starting with the regulators and management - top down. But as long as they believe that they can manage their 'loop' with bottom up regulatory constraint and more training the industry will continue to be surprised.
Danny, to add to the 'been there, done that, T shirt', etc; I probably occupied Padhist's house and job some 15 years later, but instead of evaluating automation the task was to map human capability if the autos failed. The human is more capable than many believed.
And then a further 15 years later certificating a highly reliable, but cheap autoland system, being surprised by unexpected in-service failures. Crews did not follow the abnormal procedure as landing below 50 ft in fog was judged safer than a GA - it may have been; and that the system failures were 'caused' by the crew's apprehension - 'white knuckling' the controls and inadvertently disconnecting the autos! Lack of confidence , low experience ... C'est la vie.