Hacker News Reader

Comment by rob74 | Hacker News Reader

Let's not get too carried away, shall we? At least the 346 people that died in the two 737-MAX crashes would probably question your "unqualified success" statement. But yes, no one can deny that the 737 has been a cash cow for Boeing (for far too long, some would say).

View on HN

737 MAX MCAS issue pale in comparison to 737 classic rudder issues (two crashes confirmed because of the issue, three crashes suspected, and a few more cases where pilots were able to overcome the plane). Yet it happened in the era without social networks and Internet news, so few people remember about it.

>Yet it happened in the era without social networks and Internet news, so few people remember about it.

And the public's opinion of the air travel industry was different so hand wringing over it wasn't so fashionable.

I’m old and I remember it being all over the news and Johnny Carson making jokes about Boeing “BOING” in his monologue. There was no social network but we were ALL watching the same damn 3 channels.

That's true.

Then again the 737s with rudder issues were not brand spanking new planes.

Without Boeing's negligence and corner cutting this never would have happened.

The 737 and 707 are the exemplar for airplanes. MCDs reverse takeover of Boeing has been a disaster / but safety wise even with the 737 Max debacle the 737 is the lane that took the jet age from only the elites to everyone.

To be fair, the 737 NG fiasco that preceded it isn't well known and it's unknowable how many excess deaths it caused or will cause because there are planes flying around with substandard structural components.

https://christinenegroni.com/boeing-workers-warn-of-737-ng-s...

It's hard to take the article seriously, when it claims pictures of 737s breaking from same points as evidence of substandard components. There will be always weakest points in the fuselage, which give away first when you crash the plane. I think the Turkish airlines in particular stayed intact surprisingly well, considering how it was stalled and fell like a rock.

The 737 Max has almost nothing to do with the 737. Don't conflate the two.

Do pilots need a different type rating to fly the 737 Max? No? Then it's a bit strange to say that it has nothing to do with the 737...

IANAP, but a lot of the root cause analysis around the 737 max crashes was that they were sufficiently different and so should have required re-certification of pilots before being allowed to fly them. Due to the costs involved, Boeing made the ultimately fatal mistake of minimising these changes to airlines so that pilots didnt know what they were flying (insofar as some of the subsystems).

IMO this is a bad take. MCAS was a fine system, the different aero charicteristics were fine. The only problem is that Boeing cheaped out and made MCAS depend on a single sensor for critical decision making. If they just had 3 sensors like any other saftey critical system none of these problems would have happened.

This. How many 10’s of thousands of flight hours were logged on the Max and no pilot even noticed MCAS.

They did, before the two crashes there were other incidents (e.g. a Lion Air flight just the days prior to the one that crashed where a third pilot was in the cabin by accident and his quick thinking saved the plane: https://www.cnbc.com/2019/03/20/lion-air-boeing-737-saved-by... )

I've tried to explain this before but it's been pretty negatively received. The assumption that safety critical systems on jet aircraft have triple redundancy is a misconception. At least in so far as any kind of automated, transparent to the pilot, monitoring and switching.

For example, the front of the 737 does have 3 pitot tubes. However, these drive the air data computers for the pilot and the copilot positions. The 3rd one drives a small set of backup instruments. A fourth at the rear of the plane is an input to the hydraulic pitch and feel computer. The Airbus is closer to what some might imagine but the switch between the 3 data sources is still manual (AF447 might have reached Paris if this wasn't the case).

There is another argument that adding more AoA sensors would have had a negligible impact on safety given they are exposed to the same environmental conditions. The main outcome was limiting the authority of the system, it would have prevented both accidents.

No, it would have been fine if 1) they had 3 sensors and 2) the system was in the flight manual and the pilots were trained on the system. If this means needing a new type certification, then so be it.

Which would have been a hard sell to mgmt and maybe stopped the whole project. But it should have been stopped if it wasn't profitable with new training included.

> The only problem…

No. Pilots weren’t even aware of MCAS’ existence until after the second fatal crash.

That’s a huge problem, compounded by

> Boeing cheaped out and made MCAS depend on a single AoA sensor.

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

There were also optional safety systems that lower cost carriers in the undeveloped/developing world didn’t buy, which would have prevented most of the incidents.

don't blame it on the airlines. Boing shouldn't have been able to certify a plane without redundant sensors.

Yes. But it wasn’t just about training the pilots, those systems/sensors should have been present even in the cheaper planes (or the cheaper planes simply shouldn’t have been sold).

It's worse because we literally did all this before. When the stick-shaker came out to help pilots know of impending stall conditions, the second one on the copilot's side was an optional extra! This despite the fact that the copilot is the Pilot Flying quite often in standard practice.

Do you have a reference for this? To my knowledge there was an optional AoA disagree indication (which was on a multi-function display so it's hard to argue it was any kind of cost saving). It's unlikely this would have been a factor in preventing either accident. A small warning indication with no actionable steps would have been low on the priority list for a pilot wrestling for control of the aircraft.

The 737 MAX is certified on the same type certificate as the 737 NG, 737 Classic, and 737 Jurassic. Structurally and systematically, they are nearly identical. The only major changes are new flight deck instrumentation, new engines and engine installation, new wingtips, and a re-lofted tailcone.

And those engines played a significant role in the problems.

The original 737s by Boeing were all-American masterpieces that had pretty great safety records considering the sheer number of miles flown and delivered airframes.

The 737-MAX, courtesy of Post McDonnel Douglas-Boeing merger [0], with code written by offshored 9$/h Indian coders was a complete disaster.

[0] https://www.theatlantic.com/ideas/archive/2019/11/how-boeing...

Oh dear.

https://en.m.wikipedia.org/wiki/Boeing_737_rudder_issues

And speaking of safety records, A32* is better than B737* by any metric.

I'm not sure it's fair to compare the record of a plane family that's 20 years younger than the 737. The first 737 entered service in 1968 vs 1988 for the A320, and during those 20 years, airplane safety improved significantly. Airbus benefited a lot from Boeing's learnings over that time.

As long as Boeing keeps flying the same vintage airframe, they are comparable. Boeing could have also built a more modern, safer plane, but they chose not to, which is a defensible position IMO, as the 737 is pretty safe.

$9 converted to Indian currency is Rs720 with a conversion rate of 80Rs per dollar which is around 3 lakhs a month, which is a lot in India, equates to a mid six figure salary here based on what you can afford. I'm not sure how you are trying to make a point here. Also it's unfortunate that you retort to blaming the coders and mild racism, instead of accepting Boeing messed up. FAA not once in it's reports nor any aviation experts blamed the code, Boeing went cheap and removed a redundant device the MCAS system needs during a malfunction. Even during the incident HN was quick to blame pilots for being unskilled, now it's the coders mistake? The American company went greedy and went cheap, ignored its engineers, bent FAA over it's back to make rules that suit them, ignored all the warnings. Tried to brush away opinions. And was fully aware of the issue. Instead it tried to hide it. I'm not sure where your so called $9 coder sits here.

This is a great post, but I'm confused about your math in the first sentence. Maybe I misunderstand: 3 lakhs means 300,000 INR. If 720 INR per hour, that means 415 hours per month. If we estimate 4.5 weeks per month, that is 90+ hours per week! Where did I go wrong here? And there is no way Boeing is paying less than 20 USD per hour in India for people writing flight control software!

For other readers, if the monthly salary is 300,000 INR, that would be 3.6M INR per year, or about 45K USD @ 80 INR/USD.

> equates to a mid six figure salary here based on what you can afford.

So they can convert it to six figures USD?

> Boeing messed up. FAA not once in it's reports nor any aviation experts blamed the code, Boeing went cheap and removed a redundant device the MCAS system needs during a malfunction.

McDonnell Douglas-Boeing messed up indeed. The 9$/h coders are just one of the many symptoms of a culture that lead to this kind of reckless behavior.

You are just not able to comprehend, the difference in USD and INR. I will give you a simple example for you to comprehend. The cost of Unlimited 5G internet in India with 2GB allowance per day billed at 1 YEAR is 43 USD. It's 80 USD at minimum in US per MONTH. The cost of groceries for a 4 person family per month is 243 USD. The mortgage payment on a decent 3BHK house I pay is 182 USD per month. The cost of living and expenses in India is way cheaper. I'm not sure what you are trying to achieve by emphasising 9$/hr coders. It's less in US, it's a lot in India, you are not even in the middle class at that point. They hired some competent coders if they paid that rate in India. Comparing competency of a coder paid in USD vs INR is plain foolish.

The successful Indian Space Research Organization is probably paying $15/hr to it's engineers if you are to question their competency next, I'm not sure if you even understand anything.

I mean that's the reason they outsource, it's cheaper because they have to pay way less in USD, but can hire the same level of talent, if needed. Since in INR it's a lot.

And why should they convert to USD, that's completely foolish, they live in India LOL.

Why do you think a lot of people come to US, even if it means working low income jobs. They send their money to India where it's a lot. And hoard a lot of wealth there.

> The cost of Unlimited 5G internet in India with 2GB allowance per day billed at 1 YEAR is 43 USD. It's 80 USD at minimum in US per MONTH.

That's for mobile internet I can only use in India, getting an Indian IP address. It costs 43 dollars but it useless once I land in America.

> The mortgage payment on a decent 3BHK house I pay is 182 USD per month.

For a house in India, in an Indian school district, under Indian jurisdiction. It is not the same asset because there's more to it than the number of bedrooms. See how many Mainland Chinese purchase real estate in America specifically because the assets are now under US jurisdiction and can't be seized by the ruling party, only by a US court of law.

> but can hire the same level of talent

Can you? Last I checked, Silicon Valley talent was pretty expensive worldwide. Partly because it's easy to relocate to the valley.

>all-American masterpieces

What's more all american than "Post McDonnel Douglas-Boeing merger with code written by offshored 9$/h Indian coders"?

Reply on news.ycombinator.com