Yeah, my dad just retired from United and flew 737, 767/77 and ended up with the 747-400 (among others). I don't think most traditionally trained pilots are too concerned about fly-by-wire (maybe I'm wrong), but my software experience certainly taints things for me.
Well, I
was concerned, too, when I read that Avidyne's MFD OS was WinNT-based
My thoughts tend to drift into wondering how close to the edge of the envelope will the computer let me go? The answer is obviously a moving target as technologies and, indeed, testing environments and modeling systems change.
Agreed
Also, even in the traditional "all hardware" flying model my life is hanging by some engineers' idea what is an acceptable failure rate anyway. To quote JP, "Oh God, our lives are in the hands of engineers". As we saw in Sioux City, even redundant hardware systems fail.
True, who knows how many screws, hoses and flanges are non-redundant points of failure in your average light plane? Actually, it amazes me how often trivial mechanical components feature in those accident reports (right behind judgment errors or just plain lack of skill in low-time amateur pilots).
It's always an interesting discussion. I just hate adding potential software failure points to a systems that already has enough points of failure in the hardware. Even though I know that we're trading in some hardware failures for software (i.e., no long wires/pulleys or hydraulic lines/valves to fail).
Again, agreed: in principle, simpler system should be inherently more reliable. But, let me repeat: the human component is often the weakest link...
Previous Topic
Index
