8 ms·

Agree on basically all accounts. To elaborate a bit on this point: >Because building a plane is so tightly regulated that a startup that can satisfy the regula

Agree on basically all accounts. To elaborate a bit on this point: >Because building a plane is so tightly regulated that a startup that can satisfy the regulations would basically look like Boeing, have roughly the same cost structure, and move as slowly. Highly capital intensive industries like aerospace are extremely difficult to innovate in. The timeline for return is measured in decades, so you need full commitment from all involved. So it's not just that the regulatory environment is extremely onerous, it's that the aerospace industry itself, like other industries that index extremely high on complex engineering, is already so difficult that no one really wants to try (except Anduril, I guess). This is where everyone brings up Skunkworks and all the innovation that happened back in the 60s and 70s and so on, but I think we can agree that that time period really was different and we were much earlier in the curve, so to speak.

nostrademons 2y ago
Yes. I pick on regulation because it's the most general instance of this, but the actual problem is overconstraint. When you have too many requirements - when things are too optimized in certain directions - it becomes impossible to add new constraints. Gall's Law kicks in: "A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over with a working simple system." In aviation, the issue is that we've very tightly optimized along the dimensions of safety, pilot training, and ground operations - to the extent that nearly every detail of the plane is specified, and starting over is cost prohibitive on a systems-level because it requires changes at the airline level - and so when airlines introduce the new constraint of "optimize fuel economy", the only way to satisfy it is to take a 737, bolt on better engines, and compensate for the aerodynamic changes with software. This actually makes the plane less safe (though in unknown ways), because of potential bugs in the software and aerodynamic instability, but the alternative of going back to the drawing board and optimizing for all of these constraints in actual priority order is untenable. Things like pilot training and ground operations are externally imposed by the airlines, safety by regulators, and there is no room to go backward on these dimensions. If you go to an aviation museum and look at fighter jets from the 1950s, sometimes very high-performing jets, you'll notice that they are much simpler than 4th generation fighters from the 1970s onward. Simple cylindrical airframes, turbojets that are mechanically simple but guzzle fuel, analog avionics, usually straight lines. They also perform worse in several dimensions - they have shit fuel economy, range was short, pilots lack the situational awareness that modern-day ones do, they lack the maneuverability that today's jets have. But the simplicity allowed aerospace engineers to radically optimize in certain dimensions (eg. the SR-71 or U-2) and to try new things, something that doesn't really happen now.
- remarkEon 2y ago
  If one were to start from scratch on designing a passenger jet, imagining away the constraints of current airport design, passenger safety rules, everything else ... I do wonder what the planes would look like. Perhaps like B2s, with engines in the rear.
  - nostrademons 2y ago
    Boom XB-1, which is funded by YCombinator: https://boomsupersonic.com/xb-1 https://boomsupersonic.com/xb-1 Note that they're targeting a very different market segment from most commercial aviation - they're going for supersonic overseas business travel at a premium price point. Hence, relaxing some constraints. Even then it's taken 10 years to go from seed funding to first flight, and they're still far away from a commercial flight.

nostrademons 2y ago

Yes. I pick on regulation because it's the most general instance of this, but the actual problem is overconstraint. When you have too many requirements - when things are too optimized in certain directions - it becomes impossible to add new constraints. Gall's Law kicks in: "A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over with a working simple system." In aviation, the issue is that we've very tightly optimized along the dimensions of safety, pilot training, and ground operations - to the extent that nearly every detail of the plane is specified, and starting over is cost prohibitive on a systems-level because it requires changes at the airline level - and so when airlines introduce the new constraint of "optimize fuel economy", the only way to satisfy it is to take a 737, bolt on better engines, and compensate for the aerodynamic changes with software. This actually makes the plane less safe (though in unknown ways), because of potential bugs in the software and aerodynamic instability, but the alternative of going back to the drawing board and optimizing for all of these constraints in actual priority order is untenable. Things like pilot training and ground operations are externally imposed by the airlines, safety by regulators, and there is no room to go backward on these dimensions. If you go to an aviation museum and look at fighter jets from the 1950s, sometimes very high-performing jets, you'll notice that they are much simpler than 4th generation fighters from the 1970s onward. Simple cylindrical airframes, turbojets that are mechanically simple but guzzle fuel, analog avionics, usually straight lines. They also perform worse in several dimensions - they have shit fuel economy, range was short, pilots lack the situational awareness that modern-day ones do, they lack the maneuverability that today's jets have. But the simplicity allowed aerospace engineers to radically optimize in certain dimensions (eg. the SR-71 or U-2) and to try new things, something that doesn't really happen now.

remarkEon 2y ago
If one were to start from scratch on designing a passenger jet, imagining away the constraints of current airport design, passenger safety rules, everything else ... I do wonder what the planes would look like. Perhaps like B2s, with engines in the rear.
- nostrademons 2y ago
  Boom XB-1, which is funded by YCombinator: https://boomsupersonic.com/xb-1 https://boomsupersonic.com/xb-1 Note that they're targeting a very different market segment from most commercial aviation - they're going for supersonic overseas business travel at a premium price point. Hence, relaxing some constraints. Even then it's taken 10 years to go from seed funding to first flight, and they're still far away from a commercial flight.