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

At least Airbus seems to be competent in building safe aircraft.

I would not be surprised though if Boeing's sales drop if the US government brings in tariffs or etc to try and force companies to buy their flying coffins.




> At least Airbus seems to be competent in building safe aircraft.

Prior to new models released in the last ~10yrs-ish, Boeing made the safest planes in the sky (as measured by passenger miles). Many of those planes are still flying and still doing great.

For me the interesting question is what changed in Boeing's design, testing and/or manufacturing processes which is apparently resulting in worse safety performance.


787 program:

- designed to outsource most manufacturing to the lowest bidder, many program management problems, overruns and delays

- to bust unions new factory was created in South Carolina and this has very poor QC, there are rumors that a certain large middle eastern airline refuses to accept any planes assembled there

737 program:

- they decided that they are not going to do a clean sheet 737NG and use the existing platform to put on new engines and do other modernization. They did it the cheap way and tried to paper over problems in software to make sure that their biggest customers would not need to send their pilots through extra training. Killed hundreds already.


It's safer to have a new airplane design behave like the old one. There have been many crashes due a pilot being stressed and automatically doing something that would have been right on a previous airplane he was familiar with, but which was wrong for the current airplane he is flying.

All jet airliners are aerodynamically unstable and use active controls to "paper over" that. There is nothing inherently wrong with doing that.

The MCAS problem was not due to its purpose. The problem was the software for it had too much authority, and did not shut off when the pilot countermanded it. A worse problem was the pilots did not follow emergency procedures for runaway trim.

For reference, the emergency procedures are:

1. restore normal trim with the electric trim switches (which overrides MCAS)

2. turn off the trim via a switch on the console


I don't think it's worth relitigating MCAS here but your analysis here is very generous to Boeing and harsh to the pilots who where unwitting test pilots in Boeing's mistake.


In the LA crash, the crew restored normal trim 25 times, and never thought to turn off the trim. Turning off the trim is a "memory item", meaning the pilots should not need to consult a checklist for it. The switch is right there on the center console within easy reach.

The FAA sent around an Emergency Airworthiness Directive reiterating the two step process before the EA crash:

Boeing Emergency Airworthiness Directive

"Initially, higher control forces may be needed to overcome any stabilizer nose down trim already applied. Electric stabilizer trim can be used to neutralize control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT. Manual stabilizer trim can be used before and after the STAB TRIM CUTOUT switches are moved to CUTOUT."

https://theaircurrent.com/wp-content/uploads/2018/11/B737-MA...

They also ignored the overspeed warning horn because they were apparently operating at full throttle. This was a large contributory factor to being unable to move the trim by hand.

So, yeah, it is harsh to the crews. I've also talked with MAX pilots who were quite harsh towards them. I'm amazed that one would not be harsh towards a pilot who did not bother to read/understand/remember and EMERGENCY AIRWORTHINESS DIRECTIVE about how not to crash.

Would you get on an airplane knowing it had such a pilot? Not me.

I assign 50% responsibility to the pilots and 50% to Boeing.


Maybe I'd buy that if the exact same thing didn't happen 5 months later with Ethiopian Airlines Flight 302. Most of the blame has to be on Boeing.


The EA crew did not follow the 2 step procedure in the Emergency Airworthiness Directive given to all MAX crews before the EA crash.

Also, before the LA crash, the MCAS failure happened on the previous flight of the same aircraft, and the crew just turned off the stab trim, and continued the flight normally and landed safely. That crew was unaware of MCAS, but they followed standard runaway trim emergency procedure.


The premise that MCAS makes pilot type training unnecessary, then under some circumstances the pilot should override MCAS, is dangerously flawed.


Pilots are already trained to stop runaway trim failures, and received an Emergency Airworthiness Directive explaining trivial procedure to counter MCAS failure.

Of course, the MCAS software was badly designed, and the single path failure, are squarely Boeing's fault.

Pilots must always be ready to deal with runaway stabilizer trim, on any aircraft type. It's a "memory item".


The whole shit happened because Boeing only used only one sensor for the system because then it wouldn't be seen as safety critical and wouldn't need to be mentioned extra in the manuals


> All jet airliners are aerodynamically unstable and use active controls to "paper over" that.

This is flatly wrong. The only aerodynamically unstable planes are experimental and some fighter jets.

No passenger plane can be certified if it can't be proven that it is aerodynamically stable.

EDIT: don't take my word, here is a short explanation by a 777 pilot: https://youtu.be/EuVlbVr_jrs?feature=shared&t=129


Aerodynamic stability refers to the desire of an aircraft to return to straight and level flight without any control inputs.

Stability is also commonly understood as the opposite of maneuverability, as a more stable aircraft is less maneuverable.

Combat aircraft such as fighters are designed to be aerodynamically unstable so their maneuverability is unhindered. Likewise stunt planes and other aircraft whose job is to not fly straight and level.

Aircraft such as passenger and cargo airliners are designed to be aerodynamically stable because their job is to fly straight and level with good fuel economy and minimal piloting, they do not need swift and nimble maneuverability.

The safety of an aircraft has no relation to that aircraft's aerodynamic stability.


Airliners also have to work at low speed in thick air, and high speed in thin air. Swept wings, for example, perform poorly at lower speeds.

An airliner has to perform well at both, which is why wings are swept, yet have flaps and slats to modify their aerodynamic profile. Swept wings have other problems, like dutch roll.

Airliners also become uncontrollable above a certain speed, which was a contributing factor to the EA crash (the crew ignored the overspeed warning).


> This is flatly wrong.

Look up dutch roll, the active yaw damper that suppresses it, and the FAA requirement that all of them have a yaw damper installed.

Besides, the MAX is not unstable. All it was supposed to do was just nudge the nose down a bit to match the profile of the previous 737 jets.


>All jet airliners are aerodynamically unstable

>Besides, the MAX is not unstable.

You might want to address your seeming hypocrisy there.


MCAS was not designed to address an instability issue.


It literally killed hundreds of people by pretending to be the same plane.


This is a good tabloid take but I don't believe it's accurate. It wasn't pretending to be the same plane - it was pretending to be any aircraft with certifiable flight characteristics, particularly control forces on approach to the stall.

Indeed there is no requirement for aircraft with the same training / type certificate to handle identically. For example the whole CitationJet class from the Mustang to the CJ4.


I believe that the 737NG is not the problem aircraft. The new engines and the problematic MCAS are on the 737 MAX, a different model.


Sure I apologize, there is NG and there is "New NG" or MAX.


> there are rumors that a certain large middle eastern airline

Thank god there are airlines with standards. I'm certainly glad my Middle Eastern long haul option for a particular cross-continent journey does not have any faulty Boeing models.


> what changed in Boeing's design, testing and/or manufacturing processes

My theory is that this is not limited to Boeing or even aircraft design, it's a much deeper and systemic problem affecting all kinds of fields. We've had a lot of industrial accidents lately.

When aircraft manufacturing was an emerging industry there were tons of undocumented safety margins and "slack" in the design and production pipeline.

Over time, the beancounters start optimizing stuff, so these undocumented safety margins are eroded in the name of efficiency/profit (and sometimes even documented safety margins too).

Furthermore, workers back in the day had a much better life when it comes to purchasing power (especially when it comes to property), and so could actually "give more fucks" about the job than they do now which is a compounding factor. You used to get a lot of implicit quality assurance back then which you don't get now.

We've now reached a stage where these undocumented safety margins have been eroded enough that it actually starts to cause issues, and the safeguards that are supposed to catch them aren't good enough, either due to 1) they've never been good enough but just weren't really needed before or 2) they too have been eroded in the same way for the same reason.

This also applies to software - software quality nowadays has gone down the drain for the same reasons, and even brands that were built on quality and polish (Apple) are now churning out shit (see the endless calls for another "Snow Leopard" bugfix-only macOS release).


I pretty much agree with this. The aggressive push towards optimisation everywhere in life is causing strain at the margins. I think this has always happened though, including without capitalism – it's basically the definition of growing pains. It's a natural process to some extent, and I'm not sure that corrections ever really happen, we just find new areas, new things that don't have the same limitations and grow into those.


Seems more parsimonious to explain aggressive optimisation everywhere as contraction pains. Efficiency and optimization as the obsession of a society aiming to rescue itself from encroaching constraints within a historically successful but fading paradigm- the facade of which may long outlast its public interest utility, propped up for the convenient credibility it lends private, perhaps mercenary, interests


I've heard that when they merged with McDonnell Douglas their new (McDonnell) management pushed out the good engineering culture and dropped their quality standards in a chase for more profit, leading to engineering experiencing the dead sea effect.


Is there an opposite of the Dead Sea Effect?

(I'm not talking about self-congratulatory conceits, like "raising the bar" and "fire fast", since the companies that first come to mind as saying those things tend to produce a large quantity of often poor quality, in very visible ways. Do we have concepts or terms lately for a place where most everyone does great work, and people who don't rise to that don't remain there?)


in other words: capitalism, the system we've designed our whole society around


Any system where management incentives are not aligned with end-user incentives. So all of them really.


Ah yes, the same capitalism that was around for the building of this industry that was incredibly safe up until it wasnt? Or was it safe because it was heavily regulated back then but the heavy regulations today we ignore because "capitalism"? You cant just throw these single word thought-terminators out there. When an actual cartoon is doing better than you it's time to recalibrate. "Think Mark! Think!"


Why do you think regulatory efforts have been undermined? Karl Marx showed us 150 years ago how a government under capitalism always tends toward domination by capitalists.


So you think that the failure of industry regulation is not due to capitalism as it actually exists? What then causes regulatory failure over time?


No, this isn't a well-functioning capitalist system. Competition is a core principle of capitalism. What occurred in the aerospace industry represents a government-sanctioned monopoly.


In the last 20 years nearly every industry has seen major consolidation between just a few large companies.

Is any part of the capitalist system "well functioning" anymore?


It’s always so funny to read this kind of answer when someone points out the evident flaws of capitalism! “Hey, wait a minute, this is not how capitalism is supposed to work, so you can’t say it’s capitalism”. Too bad that capitalism isn’t one monolithic thing and this is ABSOLUTELY how loosely regulated American capitalism works. It’s a form of capitalism where human life is an optimization problem that sits on the way to profits.


And Soviet built airliners are known for their safety?


Is there any kind of alternative that could be found, or have we reached the end of history, with our only two options being 2023 capitalism versus 1960s Soviet state capitalism?


Airplanes are designed by people, and are enormously complex. No system involving humans will be free of mistakes.


I'm less concerned about mistakes as I am about systemic failures and bad incentives.

Boeing seems to have created a political and regulatory environment for itself where its better for it to design and build planes poorly, than it is for it to design and build planes well.


Consider the incentives of the people at the FAA. Their incentive is to never approve a design, because if they approve a faulty design, they get the heat, too. It's much safer to just not approve anything, or at least delay demanding ever more documentation.

Hence there's always going to be a tug of war between the FAA and the industry. The FAA never wants to approve anything, and industry goes out of business if the FAA doesn't approve it.

You'll see the same forces in action with the FDA.

BTW, as is abundantly clear from history, a fatal design mistake can and has destroyed several airframe companies. Boeing's finances were punished severely after the MAX crashes. Boeing does not win by making an unsafe design. When I worked at Boeing, I didn't know any engineer who was willing to sign his name to a faulty design. Yes, the engineer responsible for a piece of work gets his name on the drawings. It's career suicide for him if he signed off on a bad design.


A working alternative has not been found yet. Even the Soviet system was not sustainable.


>Competition is a core principle of capitalism

Capitalism is literally defined by the ability to invest capital to accumulate more of it by way of profit. The logical end of this process is straightforwardly monopoly.


And if the world/environment/context of the business didn’t change then the monopolies might last, but because there is change there is room to innovate and outcompete the monopolies.


Yeah, no. You'll be bought in 95% of cases if you threaten a monopolistic position. That's why I'm really fond of signal btw.


Who doesn't love a fundamentally disastrous system justified by a sometimes possible exception?


Capitalism is not only defined by the accumulation of capital, but also by competition. The interplay between market forces, competition, innovation, and regulation in capitalism works against the formation of monopolies.

The aerospace industry is not a good example of capitalism. What we have with Boeing is basically a government sanctioned monopoly. It’s basically a weak form of nationalization, without the stigma.


>Capitalism is not only defined by the accumulation of capital, but also by competition

That's wrong. In reality, the mere theoretical potential for competition has always been more than enough to call it capitalism from any perspective. The facts are that actual competition is not a requirement.


Capitalism is simple: The capital rules supreme. As opposed to the previous system of aristocracy, where it was the land owners. Nobody would seriously claim that aristocracy requires any kind of competition between the aristocrats. Even the very first capitalist big enterprises, such as East India Company were created as _monopolies_


Monopolies are the end goal of capitalism


Monopolies are broken by capitalism just as much as they're created by capitalism. The whole "end-stage capitalism" schtick is wrong because a free market will lead to the ossification and then breakdown of a monopolist. You just have to finish spring semester of Econ101 to find out how.


Capitalism != free market. You think of liberalism here.

Capitalism is about private ownership of the means of production, no? Maybe the definition changed in the US?


In it's origins it was about the capital ruling, as opposed to the aristocracy. So political power would be in the hands of people with capital, not the landed aristocracy. The means of production only entered the equation with the industrial revolution. And capitalism is older than that, albeit not much older


>Monopolies are broken by capitalism just as much as they're created by capitalism.

Source?


There is also capitalism in Switzerland, Singapore, Australia and countless other countries.

America looks like it's speed running the soviet union and south africa.


There is a book recommendation in the other thread https://news.ycombinator.com/item?id=38894227

The McDonnell Douglas merger is also worth mentioning in this context.


I’ve heard it argued that the US government coerced that merger.


Read this: https://www.thebignewsletter.com/p/the-coming-boeing-bailout

One of the tenets of the article is that the breakdown of the once-impermeable wall between civilian and military divisions has resulted in disaster.



After the unfortunate incident in Japan this week, the Boeing 787, a plane designed in the 2000s and flying since 2013, is now the only passenger airliner without a hull loss. So they are clearly capable of producing modern safe airliners post merger.


There was an engineer who got fired for worrying publicly about the fire safety of the carbon fiber hulls during the 787 leadup.

It turned out I worked in the same building he had, and I found his old office. Word is after he left they locked it up, like a crime scene. I think they were worried his whistleblowing would turn to leaks and I guess they thought his office would have some sort of evidence? It was an interior office so no big loss real estate wise, but that was a super weird chapter.

He was painted as an aluminum bigot but I always wondered.

I used to talk to a coworker about how Mitsubishi - which built the 787 wings (something Boeing has never done before) - had introduced a regional jet and would be coming after a Boeing’s lunch. He was not worried. I’m a little shocked he’s been right so far. In fact that particular division of MHI seems to be defunct as of last February, which is news to me, so I suppose he was right. Maybe the 787 experience was as unpleasant for them as it was for Boeing.


> Word is after he left they locked it up, like a crime scene. I think they were worried his whistleblowing would turn to leaks and I guess they thought his office would have some sort of evidence?

If Boeing did clear it out, they'd be open to charges of doing a coverup. The smart thing to do is lock it up as is.


From the looks of it the composite fuselage of the A350 did its job splendidly, it held out against the fire long enough for everyone to evacuate, only flashing over after >20 minutes.


An emergency divert due to serious in-air fire could last far longer than 20 minutes.


And at speed it would act like a blast furnace.

That everyone survived the collision last week makes be feel better but still not great.


> It turned out I worked in the same building he had, and I found his old office. Word is after he left they locked it up, like a crime scene. I think they were worried his whistleblowing would turn to leaks and I guess they thought his office would have some sort of evidence? It was an interior office so no big loss real estate wise, but that was a super weird chapter.

If they were worried about a lawsuit from him they might want to preserve everything in case it was subpoenaed regardless of guilt or innocent - possibly especially if they thought his claims were wrong.


I believe I was the only one who found it weird or noteworthy. But it was locked for years.


I don’t feel like hull loss without context is a meaningful metric of anything. At the end of the day they are airplanes flown by pilots. You could have a less safe design flown by highly competent pilots and never lose a plane, or an incredibly sophisticated, technically advanced aircraft where the pilot makes a decision resulting the destruction of the aircraft.


Do you have a source for that? This lists a number of others: https://simpleflying.com/aircraft-types-zero-hull-losses/


Huh I never realized the 717 had no crashes either that’s even more impressive considering the era it was designed in tbh. The others on that list look like subtypes (ie the 747-8; there have been numerous 747 crashes).


To be fair, the 717 is a DC-9 subtype if we're being honest with ourselves. If the 717 counts as type without a hull loss, the 747-8 should count as well.


A380?


Producing, what about “designing safe aircraft post-merger”?


Back when their planes blew up on their own (TWA 800, Philippine 143, Thai Airways 114), and when the rudders liked to jam themselves to full-deflection (UA 525, USAir 427)?


Probably some upper-middle manager that needed to push something through to get a bonus.


Turns out when you care about R&D and not just profits you can build safe/reliable aircraft.

With how thin the airline industry is right now post covid though we knew stuff was going to start happening. FAA is short staffed, airline maintenance had a ton of people retire during covid, etc. Here is to hoping things will change before more people lose their lives.


The A380 is the best plane ever made in the entirety of human history. It’s a damn shame it’s not used more. I’m hoping new fuel efficient engines make it viable again because there simply is NOTHING like first class on an A380.


What makes it the best? I've flown on it, and it seems nice enough, though it isn't the most comfortable airliner I've flown on. It's certainly the biggest! But there are only what, 250 or so that were put into service? It's hard to have a strong opinion about a plane in such a small niche.


If the engineering of this plane wouldn't have been that great disaster would have loomed.[0]

No question about that.

[0] https://admiralcloudberg.medium.com/a-matter-of-millimeters-...


That was an excellent read. Thanks for that.


For me the A380 is the most comfortable. I'm slightly scared of flying, and turbulence making the entire plane jump around makes it that much worse. I'm only speculating, but I think it might be the sheer bulk of the A380 making it the smoothest rides I've ever been on.


Flying the A380 doesn't even feel like flying. Even the takeoff roll is sedate.


Agreed, I fly long haul a lot and the A380 is the best. A350 is pretty good. 787 is fine, and 777 is crap.


It’s resistant to turbulence. It has the nicest cabins for business class and up. It has many entrances which is nice because then economy doesn’t need to walk past you in contempt. It also has more safety features than I can list which makes it very hard to crash even intentionally.


It will inevitably come back in some form in the long term as air traffic keeps increasing. It's much harder to build a new runway and a new terminal in busy airports (ask Heathrow) than to add a floor to a plane.


It's interesting how so many people sounded the death knell on the A380 during the COVID-19 pandemic. Production of new A380s even ceased around this time. Now it seems demand for that size of aircraft is only increasing (at least for specific hub and spoke models).


A380s were scheduled to end production in 2021 even before COVID.

Airbus may have mistimed the A380 versus industry trends - though this is certainly up to debate and is still up in the air.

The first A380 deliveries were back in 2007, so the also had to eat the impact of the 09 financial crisis.


The control stick on the airbus is pretty bad, the fact that neither pilot knows what inputs the other is giving makes them far inferior to the system in boeing aircraft, even with warnings.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: