Use of software to make a plane airworthy when it isn't natively airworthy is a problem. Civilian aircraft should have a different standard than fighter aircraft: one that allows graceful degradation under failure.
Being a total layman in terms of aircraft design -- I've always wondered why they weren't built more akin to distributed systems. Instead of 1-3 large fuel tanks, have hundreds of small ones, so losing one is not a disaster. Similar with hydraulics and electronic wiring -- instead of a single cable running along wings, why not a redundant network that can survive everything short of a wing being sheared off?
The A-10 Close Air Support aircraft manages to survive in exceptionally stressful circumstances. The flight control systems are triply redundant: two independent hydraulic systems, and a set of cables if both hydraulic systems fail.
Having a second angle of attack sensor on the Boeing 737 MAX (which is critical to aircraft safety) be an extra cost option is unconscionable.
They are built with extensive redundancy. Airbus A320s have 5 separate computer systems, programmed by 4 different teams, for aircraft control, of which you only need one to fly the plane.
That's pretty much the case. Fuel tanks are split into multiple smaller tanks in the wings, and most systems have one or multiple redundancy (multiple hydraulics systems, multiple flight computers, multiple black boxes, graceful failure protocols, etc.)
The answer is economics - complex systems are expensive to build and maintain, and the air travel industry is notoriously low-margin. I don't see travelers accepting higher ticket prices because the airline bought (costlier) airframes with more redundancies. Any airline that attempts this will be undercut by those that do not.
You can link them with one-way seals, make them fireproof, etc. But I don't think you need hundreds. The A-10 example is great, that's an example of a plan that's designed to be shot at from close range and still fly.
Same way, with more pressure. They have to be connected anyway. (Honeycomb pattern, probably.)
The real question is how to seal them on failure without huge weight cost.
Use of software to make a plane airworthy when it isn't natively airworthy is a problem. Civilian aircraft should have a different standard than fighter aircraft: one that allows graceful degradation under failure.