Data communications over telegraph circuits 1910

You don't get my point at all. What I say is that I want systems that fail safe. When the 747 computer fails utterly I want a plane that it at least a good glider with manual controls.

When the nuclear power plant fails totally I want the control rods IN. Then the operators can read that error message in peace and quiet (and in an eerie backup light).

Monoculture of software is always a person. This problem is one of the hardest; when there is no steady state of operations to back down to.

No, I was describing a system that fails safe.

That is why you never let a single computer system handle all the control surfaces; and make a bailout for the pilot to get back an unsophisticated "powered glider" if this happens. The best solution is for the plane to this itself if the computer fails.

This is the obvious option for civilian aircraft.

If the correct heartbeat from the main system is not present, or commands outside the allowed envelope is performed, simple transistor logic and stepper motors insert the control rods. It will shut down the plant and possible make an blackout on the grid; but it will save lives.

Combat systems have the inverse priority BTW. Keep operating at all costs. An F16 does not have a stable fall-back flying mode; it must have the computer running. Probably a better example for your case.

-- mrr