Hacker News new | ask | show | jobs
by midoridensha 1100 days ago
>We have laborious and very expensive approval processes for airplanes, ask the 737 MAX inspector from the FAA how well that guarded against bad engineering.

Those approval processes don't work when the government regulators allow the company being regulated to "self-regulate". There was no real FAA approval process for the 737MAX: FAA just let Boeing do it themselves. Of course that's going to lead to bad engineering when the sales team and executives make engineering decisions, and no one outside the company knows anything about it.

1 comments

that’s my point; approval processes are always gamed around because of the cost such processes impose. You can’t just say oh the rules ought to be perfect and perfectly implemented for it to work. That’s not a world in which we live in. In fact, the more thorough the process, the more to gain in working around it.

In the meantime such processes add much to the expense such that only larger well-funded and well-connected people can participate. The ones most likely to game it, btw.

It’s a reason planes basically look the same for many decades too; newer stuff is harder to get through. Also part of why Boeing thought it best to revise an existing design once more, besides designing a new one. The process led to bad design decisions quite directly, though of course everyone blames Boeing (rightly) but not the process the FAA imposed.

>It’s a reason planes basically look the same for many decades too; newer stuff is harder to get through.

Not true. The reason planes look the same is because that's simply the optimal shape and layout, aerodynamically, for carrying passengers in the most fuel-efficient manner. Ask any aerospace engineer.

But there's been some big differences if you look more closely. Engine nacelles are a lot larger than they were 50 years ago, since bigger high-bypass turbofans are more efficient. And winglets are basically standard now. Under the skin there's huge differences: fly-by-wire, composite wings, etc.

>Also part of why Boeing thought it best to revise an existing design once more

Wrong. It's because some customers (cough Southwest cough) would only fly 737s, and because the FAA's broken rules allowed anything called a "737" with a 737 airframe to be flown by any pilot rated for that aircraft, even if they've only flown one from 1969, even though there's big differences between the generations, and because the FAA didn't mandate a more thorough process for this loophole. Boeing was afraid that these 737 users would buy the A320neo instead of a different Boeing plane, if forced into a choice of something new. In reality, the 737 airframe is old and obsolete, and should have been retired ages ago, but is only kept alive because of bad FAA regulations.

>The process led to bad design decisions quite directly, though of course everyone blames Boeing (rightly) but not the process the FAA imposed.

This is correct. The FAA is the root of the problem here. But the problem isn't that "processes can be gamed around", the problem is that this particular regulatory agency was corrupt and failed in its primary duty. Of course regulations can be gamed around; that's why the regulators are supposed to stay on top of that, and continually revise regulations to deal with this. It's a cat-and-mouse game, but here the cat just gave up and let the mouse tell it how to do its job as a cat. The answer to this problem isn't to get rid of cats; it's to euthanize this particular cat and get a better cat.

> the problem is that this particular regulatory agency was corrupt and failed in its primary duty.

What regulatory agency isn't captured?

If you make a regulatory body, you want to staff it with those who 'know' about the subject. That's going to come from the most well-known groups at the time. These are people, and they had lives and friends before in the groups they came from. So, naturally, a lighter touch will be given to some from the same groups, but not to others who are unknown; and that's best case.

Also, I have seen no evidence that any regulatory body has done much to help safety, and they will definitely discourage risk-taking. Planes for instance, were getting better and safer before the FAA arrived. Concern for the environment grew before the EPA, workplace safety increased before OSHA, and so on. All these agencies (because they are staffed again by people, who want to keep their jobs - ask a doctor, he'll say you need a doctor, etc.) will point to the improvements after their inception, and say, "look at the good job we do!". It does not hold though, that these improvements would not have occurred without the agency, and it could very well be the case that improvements come sooner.

For instance, the current basic airframe might be best, but few try other designs, like a lift body, or a flying wing, which might be more efficient. I know the B-2 bomber was based on a flying wing design and does need sophisticated computer control of the control surfaces though, to work without a tail, and that's costly enough to get right. Add the FAA rules on top, and, as has happened, that idea dies early on the vine. You might say that's a good thing, and maybe in this case it was, but it also clearly wasn't tried for very long either, and the attempts would draw greater scrutiny from the FAA.

And that's a shame.

>Also, I have seen no evidence that any regulatory body has done much to help safety

Stockton Rush, is that you? Aren't you dead? Or are you a big fan of his?

>Concern for the environment grew before the EPA

That sounds like "thoughts and prayers". "Concern" doesn't protect the environment; laws and enforcement does. The EPA was made for a good reason.

>workplace safety increased before OSHA

Some, but not in other places, just like the EPA. OSHA made rules and enforced them, so everyone got workplace safety, not just some places where management cared enough.

>ask a doctor, he'll say you need a doctor

Yes, because everyone eventually gets some illness that requires the services of a doctor. Many illnesses are only detectable with tests.

>For instance, the current basic airframe might be best, but few try other designs, like a lift body, or a flying wing, which might be more efficient.

You can read stuff all over the internet explaining why flying works don't work for passengers; it's not because of "regulation", it's because it's just a bad design for that. It's only used for bombers because it gives them stealth (combined with other design choices), and bombers don't need pressurization. Pressurizing a flying wing adds lots of weight, and to get equivalent internal volume, the design requires a huge wingspan that airports can't handle. It has nothing to do with FAA rules, and everything to do with physics.