Hacker News new | past | comments | ask | show | jobs | submit login

In 2016, I think, there were 0 airplane crashes and 0 air travel fatalities. Across 200 countries, hundreds of airlines and several airplane manufacturers, probably thousands of airports, millions of flights.

If you think that was an easy feat, then I don't know what to tell you.

Do you know how it was achieved? With finely tuned and ruthlessly efficient bureaucracy.

When people really care, bureaucracy works wonders.

Bureaucracy is basically formalizing social interactions for a specific topic. Formalizing something ossifies it, but it also prevents your pilot telling your copilot "Shut up!" just as the plane is about to crash into the mountain.




I can't find it now, but I once read a fascinating history of train braking in the US. Before modern air brakes became mandatory, trains had brake men who had to jump from on car to the other in order to fasten brakes. That had a high risk of the brake operator fallng off the moving train - and the whole train becoming uncontrollable. It was an extremely dangerous job and at the time train companies thought it would be too costly to change the system, so they resisted change.

> Bureaucracy is basically formalizing social interactions for a specific topic.

One could even say that bureaucracy is one form of organized collective action. Which in general is quite necessary for humans as a social species.


> 0 airplane crashes

I presume you meant commercial aviation, rather than aviation in general. Unfortunately not quite true even there. Your broader point stands though.

https://en.wikipedia.org/wiki/Category:Aviation_accidents_an...


> Your broader point stands though.

Well no, you've neatly disproved it.

The reason it works for commercial aviation and not anything else -- not even aviation in general -- is that the bureaucratic processes used for commercial aviation incur a massive overhead. When you have a product which costs a hundred million dollars a unit anyway and can kill 300 people in one shot if it fails, you pay the cost. For anything else it's too expensive, but spending less money causes the bureaucracy to be ineffective.

And even in commercial aviation, the overhead is still there, it's just capable of eating the loss. (Or maybe it isn't, given the miserable lack of competition in that industry now. And then where does that lead us on safety, Boeing?)


> Do you know how it was achieved? With finely tuned and ruthlessly efficient bureaucracy.

Not to dispute the effectiveness of a finely tuned and ruthlessly efficient bureaucracy, but pilots and airlines have a strong incentive to not have fatal accidents; websites however have a strong incentive to track their users.

To use an analogy, enforcing this will be less like mandatory driving exams and more like net-zero carbon emissions.


Airlines and airline manufacturers also have vast financial incentives to risk crashes by taking minor shortcuts. That doesn’t exist for your drivers test.


That's not in the airline's interest because then the plane crashes and they lose a hundred million dollar plane and get sued and go bankrupt.

It could be in an individual manager's interest because then they get a bonus and are working for some other company by the time the plane crashes, but the airline itself has the aforementioned incentive to put processes into place all on their own to prevent that from happening.


Losing a brand new hundred million dollar aircraft is a huge hit. Losing a 50 year old aircraft is a different story, especially when the options are to retire it or keep flying. Further risking a crash when your airline is facing bankruptcy suddenly looks like a reasonable trade off.

It’s not like a major airline is going to intentionally crash an airplane, but if they can trade 1 billion dollars for an extra crash every 20 years that’s a net financial benefit.


> Losing a brand new hundred million dollar aircraft is a huge hit. Losing a 50 year old aircraft is a different story

A 50 year old aircraft still costs tens of millions, and the biggest cost of a crash is the lawsuits anyway.

> especially when the options are to retire it or keep flying.

50 year old planes fly all the time. The options aren't retire it or keep flying, they're maintain it properly or don't.

> Further risking a crash when your airline is facing bankruptcy suddenly looks like a reasonable trade off.

Which is where the insurance company comes in, and we're back to having an existing bureaucracy with an incentive to prevent that from happening.

> It’s not like a major airline is going to intentionally crash an airplane, but if they can trade 1 billion dollars for an extra crash every 20 years that’s a net financial benefit.

The value of a statistical life is generally regarded as being about ten million dollars. Times 300 passengers that's $3 billion. So that's how much they can expect to get sued for when the plane crashes, in addition to whatever the plane was worth. If they're "only" saving a billion dollars, they're losing money.

And if they could somehow save more than 3 billion dollars then that's what they're supposed to do -- at some point safety measures cost more than the value they provide and VSL calculations tell you where that is. (And if you don't think so then I assume you never travel by automobile or buy anything that has ever been in a truck.)


Insurance on older aircraft much like older cars drops because their worth less.

Your lawsuit numbers are also wildly off. Ex: “The US aviation giant has settled the first in a series of lawsuits filed by families of 737 Max crash victims. Boeing will reportedly pay $1.2 million to 11 families of victims killed in the 2018 Lion Air crash.“ https://www.dw.com/en/boeing-settles-first-lawsuit-with-737-...


No they're not. That's what they're supposed to get. If courts don't actually give families that then the problem is in the court system, not the company.


It doesn't matter where the problem lies. The value courts assign to life informs the airline when they estimate how much they would lose due to a crash.


It matters where the problem lies if you want to fix it.


Your argument is all "logical" but it does not agree with reality, the boeing 737 Max is a prime example.


Only then you're proving the other point just as much, because the 737 Max is also subject to the bureaucratic regulatory system being held up as an exemplar of something that works.


For 2 crashes out of 387 aircraft and several hundred thousand flights to be considered unacceptably dangerous suggests the system is largely working as intended. The aircraft had 0 fatalities in US or EU, globally an order of magnitude safer than driving, and it was still grounded.

Imagine if we took car safety so seriously.


The thing is, there isn't a huge commercial vested interest in having plane crashes happen.


"but it also prevents your pilot telling your copilot "Shut up!" just as the plane is about to crash into the mountain. "

I am curious? How does regulation prevents that?

(and how does it prevent a suicidal pilot, from locking out the co pilot and crashing in a mountain on purpose?

https://en.m.wikipedia.org/wiki/Suicide_by_pilot

)



And the aviation industry is not exactly a massive innovator at this point. The most commonly used commercial aircraft today were developed before the world wide web was even a thing.

Why don't we just skip a few steps ahead - delete the internet and go back to cable TV? That's where we're headed for anyway.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: