Hacker News new | past | comments | ask | show | jobs | submit login
Boeing's Managerial Revolution Created the 737 Max Disaster (newrepublic.com)
282 points by johnny313 22 days ago | hide | past | web | favorite | 129 comments

> Everybody thinks his business is different, because everybody is the same. Nobody. Is. Different. [from a wall street analyst about Boeing]

This in a nutshell represents the core failure of modern managerial thinking. What started as "scientific management" many decades ago, an attempt to systematically understand how corporations should operate, quickly evolved into a group think belief that management could function as fungible & completely replaceable units of work-- that a "manager" could be trained up in various basic principles and plugged into any role

There are of course commonalities across businesses, and commonalities across managers for necessary skill, but each business is, in some core, fundamental way, different from others. Good management of a company doesn't following a paint-by-numbers rule book, it requires deep understanding of the particulars of that business, the market & industry it operates in.

I think this is changing, slowly, but the legacy of this flawed thinking is a long one.

Yea, this whole “management is a profession in itself” thing is incredibly toxic on so many levels... If you’re American you should count yourself lucky that this thinking has not swallowed the whole economy. Where I am (in Sweden) it pretty much has. :(

It's the same, or maybe even worse in Germany.

On a previous job I was working on a component for the project that was a well known Audi model. In 18 months, Audi replaced the manager for that project 3 times.

Managerial careers in German corps are a revolving door where incompetents go in, do nothing, slap the prestigious company name in their resume, then, when the shit hits the fan, jump ship to next prestigious company. Rinse and repeat.

The American economy has been swallowed up by "management as a profession" since Frederick Winslow Taylor in the late 1800s.

You're all late to the party.

Isn't that the point the article is making?

That in late-stage capitalism, all actual work is invariably outsourced, with only management and profits retained?

I'm not sure I personally see the link to outsourcing. But I do agree with the quote in the OP that (in my own words) once innovation is no longer the main differentiator you typically see the “power and money people” take over. So I guess you could say that Americans should count themselves lucky that they still have large sections of the economy where innovation is the differentiator.

> once innovation is no longer the main differentiator you typically see the “power and money people” take over.

Someone contrasting engineering culture at Tesla and current "old auto", described part of the latter as being fine with someone else making the actual vehicle, and then just slapping a brand on it. Suggesting causality can go the other way too - innovation is still a differentiator, but is deemphasized by a management culture focused elsewhere.

Re: the link to outsourcing - is it not clear? Without scapegoating the cheaper external workers - as the business gets commodified, management sees less and less of the value in the internal expertise and institutional memory.

You start to see a 50-year-old engineer not as a battle-hardened mentor who can be a leader for your teams and instead as a highly-paid shambling pile with less "productivity" than a fresh grad.

>evolved into a group think belief that management could function as fungible & completely replaceable units of work

That is just an element of consistency. Since the 1980s management 'wisdom' has been centered around the idea that all workers should be fungible, replaceable units of work. Ideas like 'if you are irreplaceable, you are expendable' and the notion that your degree of importance to the functioning of a company is also a measure of how much of a liability you are, etc. They're ideas rooted in the notion that a business should prefer to close its doors rather than do anything which could raise the cost of labor in the market. If an individual becomes important in the functioning of a business, they should be eliminated as swiftly as possible because otherwise they will remain and eventually ask for more money. Since you can never, ever, consider granting that request, you will have to fire them at that point and the damage would be even harder to repair than if you had disposed of them when you first realized their importance. So better to eject them when you first realize it and clean up whatever damage that causes when it is more containable. Perhaps that viewpoint might have had some relevance in industries centered around commodity repetitive physical labor, but I would argue that it is overall destructive in knowledge-based industries. It has been phenomenally successful at suppressing the cost of labor (between 1950 and 1980, the median wage of workers rose 75%, between 1980 and 2010 it rose by only 1%) but it's caused a widespread host of much larger problems that threaten the very foundation of the economy.

> This in a nutshell represents the core failure of modern managerial thinking. What started as "scientific management" many decades ago, an attempt to systematically understand how corporations should operate, quickly evolved into a group think belief that management could function as fungible & completely replaceable units of work-- that a "manager" could be trained up in various basic principles and plugged into any role

This is what I like about Chinese industry, for now...

They have not yet got themselves into "scientific management" coolaid.

"Sales call in the morning, prototypes in the evening" — says all about it. Find a single American electronics maker who can do that.

The crisis in American corporate culture is that people can't find out what their company exceeds in. So much focus goes to MBA things, than whatever is company's main competence gets less and less clear.

But in China, things crystal clear, for now:

Factories compete on cheap widgets

Engineering companies compete on TTM

Lawyers compete on winning cases

Car companies compete on horsepowers

Advertising companies compete on being obnoxious

Banks compete on cheaper loans

No "strategic analysis" bullshit

I expect the downvotes are because of the unmentioned "communist, single-party states are the essence of groupthink coolaid" elephant.

But I doubt you're wrong on the engineering analysis.

The physical world has the unique property of being unbeholden to vision. Either the widget works, or it doesn't.

And in a marketplace with actual competition, you're going to have a lot of people working very hard to build a better widget.

stealing your R&D and producing things using near-slave labor probably doesn't hurt competitive advantage too much either, but lets not pay attention to any of that..

Having worked for some startups and a couple of large companies, I note that while the large companies tend to fall into this generic-management, vision-free thinking, successful startups are not run that way. Successful startups are all about adapting to the particular niche they're trying to serve.

(I did once work for a startup that was run somewhat like a large company. It failed. Most startups fail anyway, of course, but I don't think this one had to fail.)

I almost burst out laughing reading this. I’ve also worked across start-ups and larger established companies, across industries and with different ranges of success outcomes.

Never saw one iota of difference in how management is handled. If anything, the reduced bureaucracy of the start-ups was a problem because it left executive behavior wildly unconstrained, where rampant religious bigotry, sexual harassment, micromanagement and overwork were elevated into practically glorified business ideals and self-serving cultish culture mandates

Executives at bigger companies were just as reprehensible most of the time, but at least had many more layers of other corporate oversight constraining them to at least give appearances of professionalism and sincerely consider employee’s needs when designing policy.

So... managers/executives are just reprehensible in general...?

Someone having absolute power over you leads to reprehensible behavior, and the most common expression of that power in the modern age is the manager-employee relationship.

Oh, be serious!

No manager has "absolute power" over you. At a maximum they have the power to fire you. As a software developer in this economy there are far more important things to worry about.

And yeah, I'm a manager!

Yeah, firing you so you can't pay rent and don't have health care isn't much power at all, is it?

And withholding a reference for their next job hunt certainly doesn't do anything either.

The OP said "absolute power" not "any power" so let's at least get that straight.

Nice goalpost moving, though.

And TBH, I don't know anyone who uses their current job as reference for the next job: probably wouldn't go over well. I still have people I last worked with 10 years ago using me as a reference.

Yeah in modern capitalism simple failure to "fit in" often leads to death for a free country you have to toe the line so tight it breaks your feet but you better not complain lol

Yeah, there is even a whole academic-grade sociology book about this, called Moral Mazes.

Interesting. Haven’t come across this book before.

A quick glance on the Wikipedia page[1] doesn’t give me the impression that it paints all managers/executives as reprehensible though... But I guess you’ve read it and come to this conclusion?

I do agree though that there is, in many of the organizations I’ve worked in, large and smallish, a perverse incentive structure in that "instead of ability, talent, and dedicated service to an organization, politics, adroit talk, luck, connections, and self-promotion are the real sorters of people into sheep and goats" [1]. But you think the solution to this is more bureaucracy? I’m leaning more towards better selection, cultural norms and perhaps some “separation of powers”. But I’m speculating.

1. https://en.m.wikipedia.org/wiki/Moral_Mazes

Most people are selfish. But this only matters when people have power. That's why people in power look so bad.

An interesting and succinct explanation.

It goes well with my own thinking too, that (in management/power) you want people who are focused on the outside world, with a “selfish” desire to affect/change it. When they act “selfishly” they will at least do something resembling executing on a strategy.

The people you really want to avoid are those that are not very talented or able and just want power for its own sake. When they turn “selfish” they will focus within the organization, promoting cronies, expanding their power base, etc, etc. when others see this the culture goes to hell...

That’s my current theory at least. :P

And were those startups successful?

One of them was, yes.

I understand what you're saying about unconstrained executive behavior at startups, but that's not what I'm talking about. I'm referring specifically to whether they ran the business as a generic commodity business, where cost-cutting took precedence over innovation. For all their flaws and misbehaviors, I'll wager that the successful ones didn't.

Yes, absolutely, I think that's a part of the success of startups, possibly any new company. The participants fully know, understand and believe in the company's products and vision,

The US military is the ultimate representation of this "managers-are-replacable-units-of-work". Officers (managers) are yanked around the world from job to job every couple of years. There's a degree of specialization early (infantry, pilot, logistician, etc) but that is really just an initialization step to make sure the organization has sufficient coverage for all the junior jobs. Future jobs progress toward generality. This leads to generals and admirals with supreme knowledge of fundamental policy issues, and deep connections with lots of peers, seniors, and subordinates. But there is a frozen middle of non-attaining careerists who are either entirely competent but were passed over for the next rank and are now looking for the door, and incompetents trying to hold on as long as possible. The paradox is that, in this frozen middle, it's the incompetents who seek power (in the hopes of getting that promotion) and the competents who are unloading responsibilities as they prepare to transition.

This has created a new problem, where the existing leadership is looking for traditional signals for promotion and seeing this bizzaro world of incompetents doing the key thing it takes to promote: exercising power at a fairly senior level (a typical officer in the O-6 grade is responsible for about 1000-5000 people), and remember, the flag officers have those deep connections, they know who's competent and they know those people aren't seeking additional responsibilities. So they got Congress to authorize bringing in outsiders at the O-6 level (clearly competent, and have exercised authority at that key 1000-5000 head-count level, or at least 500-1000), which pisses off the career-competents to no end and puts the fear of god into the career-incompetents, exacerbating the situation. Yet it did nothing because who in god's name, making 500k on the outside, is going to take a $150k job at the Pentagon and live in a shack 20 miles outside the Beltway and endure that commute on the 66?

So now the flag officers have finally also gotten authority (finally) to start forcibly retiring the dead wood from the senior ranks: O-4s, O-5s, and O-6s who have failed to select for the next rank several years in a row, frequently sitting in jobs that could be filled by O-2s or O-3s. Keep in mind, the head-count at each rank is capped by Congress, so the competents can't even promote around the incompetents. It's a war of attrition where the dead wood wins because the opportunities outside outweigh the opportunities inside. At which point it really does start looking like a jobs program. Thus these new "continuation selection boards".

This is a real problem. I had one of these incompetent O-5s sitting in an O-3 billet working for me and getting that person transferred out of my group was the single best thing I ever did for group morale. And before you accuse of me of just pushing the problem onto another group, let me just say that steps were taken to ensure this individual will also be seen further up in the CSB list.

But even that new culling process seems insufficient. The standards for record review are too high, and the observed number of forced retirements is too low. The system is driven by fear, uncertainty, and doubt. All the way down.

How about the requirement for Navy Officers and Chief Perry officers to have a picture in their service record. This always struck me. Why? So they would not advance ugly people? I never understood that. Maybe you had a mole in the wrong place? When people were going before the selection board they would pay for a professional picture to make them look the best. That seemed like the least thing to judge a person by. Their 'looks'. The whole selection process was bullshit and not based on merit. I can't imagine the politics involved at the flag level.

When you give a bunch of boys a $1T use-it-or-lose-it budget to run around playing with guns and buy shiny toys (no offence), with no war to fight and no metrics for success, well, you're not going to end up with a real efficient operation staffed with smart go-getters firmly based in reality. :)

How does pay work at a given rank in the US military? Looking at a table of pay [1] it lists a range for most ranks. For example, it lists for O-3, Captain, a range of $48560 to $79002. What determines where you are in that range if you are a Captain?

Ranges for consecutive ranks overlap quite a bit. O-4, Major, is $55231 to $92218. If you are promoted from Captain to Major, and were at the high end of Captain pay, I assume you don't drop down to the bottom of Major pay. Do you go to about the same percentile in Major pay that you had in Captain pay?

[1] https://www.federalpay.org/military/army/ranks

if you click the ranks on that site, it lets you calculate based on years of experience.

> There are of course commonalities across businesses, and commonalities across managers for necessary skill, but each business is, in some core, fundamental way, different from others. Good management of a company doesn't following a paint-by-numbers rule book, it requires deep understanding of the particulars of that business, the market & industry it operates in.

This is actually part of the problem with Boeing. They keep all the functions across the aircraft separate, or at least at arms length. So for instance if you want to talk about fuel systems, there is a 737 SME, a 747 SME, a 787 SME, a 767 SME, and a legacy SME.

Cross the pond and want to talk fuel systems at Airbus? There is one SME group that manages all of the fuel systems across the commercial aircraft with leads for each program.

What Boeing created was a system that requires internal competition for resources and promotions and spread knowledge too thin without the sharing of the commonalities across the fleet.

Early in my career when I encountered outsourcing initiatives I perceived them as management choosing cost over quality. A couple of decades later my view has changed on that mostly thanks to having worked more directly within the management structure of numerous companies (ranging from startup to ConglomCo Megasized).

In most cases those making the decision really believe they can achieve comparable results by outsourcing while also taking advantage of global wage differences and also converting the continual overhead of personnel into just-in-time overhead that only has to be taken on as needed. This is not unlike the devops trend towards using cloud-based infrastructure.

The problem is that the thinking that you achieve comparable results by outsourcing is woefully naive. In-house teams have a vested interest in understanding the business problems as well as filling in the communications gap between what is asked for vs what is actually desired as an outcome. Outsourced contracted labor does not share those same interests. Additionally, when you outsource technical labor you MUST have the ability internally to create very good designs, specifications, and oversight/feedback mechanisms to ensure that you're getting what you wanted rather than what you asked for via sloppy communication. For those companies that have previously had internal teams and decide to transition to external resources they almost never have someone with that ability and in fact are unaware of the impedance mismatch between their asks and the results delivered by internal teams.

Outsourcing means trading the overhead of personnel costs for the overhead of a much more complex and difficult communications burden. It is extraordinarily rare that the company that opts to transition from internal to outsourced talent has the capability or even awareness of that trade, resulting in predictably poor results.

This seems to happen when software companies start to be run by people who have never written code. (I'm sure the observation generalizes to other forms of engineering, but software is what I'm familiar with.) They tend to perceive the work through a very shallow lens. They see that time is spent and code is produced, and they think that that's the essence of the process. What they don't see is all the knowledge that goes into building systems that work well; they have none of that knowledge themselves, and so it's invisible to them. They don't know the difference between a well-written program and a badly-written one, so it's easy for them to think they're getting the same results for less money by outsourcing.

Additionally, when you outsource technical labor you MUST have the ability internally to create very good designs, specifications, and oversight/feedback mechanisms to ensure that you're getting what you wanted rather than what you asked for via sloppy communication

In fact there’s a word for a person who can translate vague business requirements into precise, unambiguous language.


Of course to a manager a programmer just looks like an expensive typist, one who uses too many semicolons.

In fairness to the hypothetical manager, there are plenty of people with the title of programmer who are unwilling/incapable of such translation. If it were merely sufficient to speak with Area Programmer(tm) then outsourcing would be perfectly fine.

Even if designs are done in house for outsourced production, it's stealing a great resource away from the designers: feedback.

They never see the process it takes to be created, the communication pipeline between the designer and builder is throttled, their designs are only tweaked when large or obvious flaws are found.

This is a wonderful comment. I think someone on this site said that coding is the process of understanding the solution. This is what that means.

Indeed. They are also additionally constrained by their knowledge and understanding of the universe of possible solutions rather than being able to draw from the larger combined knowledge pool of themselves and the builders.

There's another wrinkle : sometimes people outsource problems. "This process/system/unit is a constant source of trouble, other people do this well, we can't.. let's outsource it". The pattern here is that the constant complaints go away (or at least the outsourced company is now the "sin eater" and you aren't on the hook). But of course, there is no incentive to actually fix what's gone wrong, in fact the opposite is true... and so these problems stay on the balance sheet for years and decades more than they should, soaking cash that could be used for investment and dragging business performance down. But there is no way out of this tarpit - all hope of actually dealing with the problem goes with the signature on the outsource contract.

Couldn't resist posting this excerpt:


And while Boeing’s engineers toiled to get McDonnell’s lemon planes into the sky, their own hopes of designing a new plane to compete with Airbus, Boeing’s only global market rival, were shriveling. Under the sway of all the naysayers who had called out the folly of the McDonnell deal, the board had adopted a hard-line “never again” posture toward ambitious new planes. Boeing’s leaders began crying “crocodile tears,” Sorscher claimed, about the development costs of 1995’s 777, even though some industry insiders estimate that it became the most profitable plane of all time. The premise behind this complaining was silly, Sorscher contended in PowerPoint presentations and a Harvard Business School-style case study on the topic. A return to the “problem-solving” culture and managerial structure of yore, he explained over and over again to anyone who would listen, was the only sensible way to generate shareholder value. But when he brought that message on the road, he rarely elicited much more than an eye roll. “I’m not buying it,” was a common response. Occasionally, though, someone in the audience was outright mean, like the Wall Street analyst who cut him off mid-sentence:

“Look, I get it. What you’re telling me is that your business is different. That you’re special. Well, listen: Everybody thinks his business is different, because everybody is the same. Nobody. Is. Different.”

And indeed, that would appear to be the real moral of this story: Airplane manufacturing is no different from mortgage lending or insulin distribution or make-believe blood analyzing software—another cash cow for the one percent, bound inexorably for the slaughterhouse.

The 777 program is a case study in program management and engineering excellence, and that aircraft is still an amazing piece of quality design and engineering.

It is so sad that the Boeing of those days is dead and buried. Everything since the MD takeover ("merger") has been a race to the bottom.

Might there be a literal case study of the 777? I read a book about the development of the 747 and it was absolutely fascinating--particularly the (now extinct?) Boeing obsession with engineering safety.

[0] Jumbo: The Making of the Boeing 747


As you liked that, consider "Twenty-First-Century Jet: The Making and Marketing of the Boeing 777" by Karl Sabbagh. I believe there is also a video series from PBS. The 777 development program might have been Boeing's technical apotheosis.

The PBS documentary is linked in the above article:


with this quote:

"There is a scene in the first episode of a five-part 1996 PBS series on the making of the 777, Boeing’s first plane to employ “fly-by-wire” technology, in which an engineer discusses the company’s philosophy of computer-assisted aviation:

One of the things that we do in the basic design is the pilot always has the ultimate authority of control. There’s no computer on the airplane that he cannot override, or turn off if the ultimate comes, but, in terms of any of our features, even those that are built to prevent the airplane from stalling, which is the lowest speed you can fly and beyond which you would lose the control. We don’t inhibit that totally; we make it difficult, but if something in the box should inappropriately think that it’s stalling when it isn’t, the pilot can say, this is wrong and he can override it. That’s a fundamental difference in philosophy that we have versus some of the competition.

For anyone else: "the highest point in the development of something; a culmination or climax."

I highly recommend the book (Jumbo: The Making of the Boeing 747). One of the best "how succesfull engineering is really done" books I have ever read.

I remember reading a series of articles in the Seattle Times about a Boeing plane that was nearing completion. I'm not sure it was the 777, but the timeline fits, as I'm sure I was reading these articles around 1992 or 1993. What follows is from my memory of those articles--the details may be wrong, but the high level is right.

Anyway, the plane had a very different control architecture from previous Boeing planes, including more reliance on software. The FAA was concerned about the software, and the possibility of a software failure taking out all three instances of some of the redundant systems. The FAA normally required a lot more testing of such software than would fit with the schedule Boeing wanted. Boeing and the FAA agreed that in these systems, the three instances would run separate implementations of the software, writing by separate development teams working from a common spec, but not sharing anything else. The FAA would then allow the software to be deployed on the accelerated schedule Boeing wanted.

This turned out to take longer than Boeing expected, and started threatening to delay the schedule, and hurt Boeing's chances against the newest Airbus plane that was aiming for a similar market. Politicians got word of this, and the FAA was successfully pressured into relaxing the redundancy requirements, letting Boeing go ahead with running the same brand new software on all instances of the redundant systems.

The software turned out fine, so this worked out, but it still was a dangerous shortcut that the FAA allowed (or was forced to allow) that put competing with Airbus over safety. So I'm not sure things were actually all that different in the '90s. They just were luckier.

I recall reading that giving different groups the specs like that and writing different code doesnt actually work well. They tend to write the same bugs. In this case it may he that they got to choose the best of the 3 versions, so it may not have been a waste.

Formal unit testing is another idea that turns out to be relatively low value but is widely regarded as a best practice.

IIRC NASA studied both of these, but I dont have a reference.

> And indeed, that would appear to be the real moral of this story: Airplane manufacturing is no different from mortgage lending or insulin distribution or make-believe blood analyzing software—another cash cow for the one percent, bound inexorably for the slaughterhouse.

But building airplanes has a far more direct consequence to people's lives then mortgage lending and other "pie in the sky" wallstreet mumbo jumbo.

Not building a proper airplane results in preventable disasters, with a great loss of life and trust as a result.

Sadly with the current US health care system not having enough money for hospital visit may easily mean the loss of life.

The only difference with airplanes is that it's much more visible, which means that it's harder to cheat.

You often hear about required ethics courses for engineers, but it strikes me that it is management that most requires this. The whole MBA thing & the implied notion of managerialism has been a straight up disaster to everything it has touched (other than, of course, the enriching of the 0.1%)

I would argue that the mba program is basically irrelevant. The problem is that the incentive structures for management in these companies is overwhelmingly to produce short term gains in order to secure promotion and bonuses, since the metric of success is to not stay in your role very long.

The ideal situation for a manager is to be promoted through/across the company every few years. Long enough to deliver some large project (that was certainly already in the pipeline) but short enough not to be responsible for the long term impact of that project.

You don't fix that by giving the management lectures on morality, you fix it by reducing their ability to hide the issues that happens in their organisation so that when lines are crossed the manager can't be confident they can sweep it under the rug.

>I would argue that the mba program is basically irrelevant.

It is definitely relevant. The programs will pump students full of notions of "corporate social responsibility", right before they send them off to consulting gigs for the summer where short-termism is how you get the offer. It's a self reinforcing system. It would maybe be different if MBA programs weren't a cash cow for universities (especially universities that cater to international students), and instead were essentially apprenticeships at the companies themselves. I agree with the general sentiment of your comment, but I'm not ready to forgive the MBA programs' part in this.

> You don't fix that by giving the management lectures on morality, you fix it by reducing their ability to hide the issues that happens in their organisation so that when lines are crossed the manager can't be confident they can sweep it under the rug.

You also don't fix this by overlooking the fact that all publicly traded organizations are performing for the street, not the customer. How much can we sell them is the mantra now, more so, than ever. Instead of being the best we've decided that lobbying and paying the board/shareholders is the best for everyone. Those people just happen to be the 0.1% as well.

You measure what you can, and in management that's often short term things. Intangibles can't get measured, thus intangibles don't get measured -- not even estimated. The result is a distorted view of everything.

Most MBA programs have required ethics courses, so it's clear that simply having a course requirement is insufficient.

Ethics courses teach frameworks that can make it easier to rationalize commonsense unethical decisions as being ethical, so I'm not convinced they're actually the right answer

The incentives need to addressed and they're missing, there arent meaningful economic and reputational consequences for management

And software engineers do plenty of unethical things when the consequences are removed and the payoff is good. How many of us operate under the immunity of “lol they’ll blame russia”

Professional Engineers in Canada (and in some states in the US AFAIK) are required to have a valid licence, that can be revoked on ethical breach.

Although it's rar to see it as a requirement in software (most software developers aren't engineers anyway), it is more common to see it required when public safety is on the line.

When breaching ethics will cost you not only your job, but your employability altogether, that's a strong incentive to act ethically.

I was a member of the Institute of Directors and they try to espouse the importance of ethics: https://www.iod.org.nz/Portals/0/Publications/Founding%20Doc...

And, in my country's law at least, some ethics are codified into legal requirements for directors: http://www.legislation.govt.nz/act/public/1993/0105/latest/D...

s131 to s138a

Of course, in both the ethics are basically "Work in the best interests of the company". Which may not ethically align with the greater good of society. But that's where criminal law comes into it, hopefully.

As for professional engineers, sadly, there's no requirement in my country that an engineer be a member of IPENZ, as notoriously demonstrated by the case of Alan Reay, who was responsible for a substandard building that killed 115 people in the Christchurch 2011 earthquake. https://en.wikipedia.org/wiki/CTV_Building#Royal_Commission_...

He simply resigned from IPENZ to avoid their ability to censure him, and sadly, our corporate manslaughter laws were found to be woefully lacking around this (and a coal mine explosion): https://www.rnz.co.nz/news/national/345102/corporate-manslau...

Question remains too, how would you define ethics for a programmer? Some programmers consider it unethical to work in adtech or porn, yet plenty of others don't. What about facial recognition techology?

It's like the old joke - no ethical programmer would write a function called bombBaghdad(), instead they'd write it as bomb() and pass the city as a parameter.

Couldn't agree more. MBA grads in place of power have done so much damage to social fabric of US.

Economic strip mining.

your ire is poorly targeted. ethics is taught in mba programs. whether that translates into more ethical companies is certainly debatable, but don't simply assume mba holders are bad people (and engineers are good people).

the reasons for the paucity of corporate ethical behavior is complex and varied, but certainly one cause is we've collectively capitulated to the supremacy of the profit motive, and there's just not enough perceived (direct) benefit in doing so.

The ire is pointed in the right direction, but the issue isn't not rooted in ethics, it's rooted in measurability. Management decisions were made as if things that could not be measured could be disregarded.

Surely someone must have had a vague sense that relying on $9 per hour software development might not be a wise choice, but in a "managerialistic" environment, the concrete cost savings are taken into account, and the vague sense is not.

is this not the result of there being very little ideological differences between MBA graduates? I would even attribute this to a lack of ideological differences in (US) society as a whole.

Economic and managerial schools only seem to teach one ideological viewpoint, which in term results in there being very little recourse on what would be the right way to tackle a situation, philosophically and ideologically speaking.

Misaligned incentives cause good people to make bad decisions.

Also worth noting is the House Aviation Subcommittee's take on the problem:

The panel’s majority members were best embodied by subcommittee Chairman Rick Larsen, an ex-lobbyist New Democrat who delivered an opening statement of head-scratching insipidness: “As I have said before, if the public does not feel safe about flying then they won’t fly; if they don’t fly, airlines don’t need to buy airplanes; if they don’t need to buy airplanes, then airplanes don’t need to be built; and if there is no need to build airplanes, then there will be no jobs in aviation.”

Followed by Congressman Paul Mitchell's lambasting of Captain Dan Carey, the outgoing president of the Allied Pilots Association, for having the temerity to go public with his concerns: "What’s the value [of releasing a secret recording] to the system or the families?... Explain to me what warranted that, sir!" Clearly, the problem here is that the travelling public, and victims' families, have come to learn about MCAS and its role in recent crashes.

When you are trying to spin a story, however, it can be difficult to keep it straight. Here's Sam Graves, ranking Republican on the House Transportation and Infrastructure Committee:

“It’s not the plane I’m concerned about; I think the plane is very safe. We need to concentrate on the pilot ... being trained [for] the aircraft, and being able to fly a plane and not just fly the computer.”

His blame-the-pilots rhetoric does not look so good for Boeing now that we are learning the lengths it went to in order to avoid there being any additional training for pilots.

Sam Graves is a commercially rated pilot.

>He graduated from the University of Missouri College of Agriculture in Columbia, Missouri with a degree in Agronomy.

Quick the nose is pitching down and you didn't command it. What do you do? Flip the trim cutouts. True before MCAS true after.

Firstly, on previous variants, opposing the trim runaway with the yoke was sufficient to cut out the electric trim, but this was necessarily changed for MCAS-commanded trim changes. It is conceivable that a trim runaway might not be stopped by that action, but it is apparently highly unlikely -- such situations are apparently not part of current 737 simulator training [1].

This is significant, because stopping a runaway by opposing it with the yoke leaves the airplane close to being in trim, and toggling the cutout switches becomes a clean-up operation to prevent a recurrence. This, apparently, is as far as pilot simulator training for trim runaway goes. In the case of MCAS, however, toggling the swiches is the only way to stop it, and it leaves the airplane significantly out of trim.

This is what the Ethiopian Airlines flight crew did, but then they found it impossible to put the airplane back in trim because the load on the stabilizer jackscrew was resisting their attempts at manual trimming. This problem is mentioned in the documentation (though not, significantly, in the checkist that the FAA's response to the Lion Air crash directed pilots' attention to), but in that documentation, it rather vaguely says this might be a problem, and if so, it might be necessary to pitch the airplane further in the direction of the mis-trim to reduce the load on the jackscrew so that it can be cranked in manually. It suggests that several cycles may be necessary to complete the action.

If you are, as the Ethiopian Airlines crew were, at low alttude in an airplane mis-trimmed to dive, this is not a viable option, which is why they re-engaged the electric trim in the hope of correcting the problem. The article mentions a video, that was briefly available, of an exercise to test this scenario, in which the professional pilots struggled to maintain control.

It is clear from this that MCAS runaway is sufficiently different from the sort of trim runaways that pilots have previously experienced, either for real or in their simulator training, and this alone should arguably have been sufficient reason for additional training to prepare for MCAS failure. Even after the Lion Air crash, the response of Boeing and the FAA failed to make this distinction clear -- the message was 'this is merely a form of trim runaway, just like what you have trained for', which it is not.

And then there is the separate issue of not using both AofA inputs, apparently just to keep the lie that there was nothing about MCAS that pilots needed to know about.

None of the above is my own opinion - they are issues that professional pilots, engineers and safety experts have forcefully raised. Captain Dan Carey himself initially went along with Boeing's take on the issue, until more information about the issue came out. If you have contrary experience of 737 training and operation, your point of view would be most welcome.

[1] https://www.youtube.com/watch?v=SvpAsiwdM_E

Probably the most interesting and relevant part to HN

> It smacked of the sort of screwup a 23-year-old intern might have made—and indeed, much of the software on the MAX had been engineered by recent grads of Indian software-coding academies making as little as $9 an hour, part of Boeing management’s endless war on the unions that once represented more than half its employees

Link to the original article: https://www.bloomberg.com/news/articles/2019-06-28/boeing-s-...

"Boeing said the company did not rely on engineers from HCL and Cyient for the Maneuvering Characteristics Augmentation System, which has been linked to the Lion Air crash last October and the Ethiopian Airlines disaster in March. The Chicago-based planemaker also said it didn’t rely on either firm for another software issue disclosed after the crashes: a cockpit warning light that wasn’t working for most buyers."

but training HCL likely was taking tall:

"Still, for the 787, HCL gave Boeing a remarkable price – free, according to Sam Swaro, an associate vice president who pitched HCL’s services at a San Diego conference sponsored by Avionics International magazine in June. He said the company took no up-front payments on the 787 and only started collecting payments based on sales years later, an “innovative business model” he offered to extend to others in the industry.

The 787 entered service three years late and billions of dollars over budget in 2011, in part because of confusion introduced by the outsourcing strategy. Under Dennis Muilenburg, a longtime Boeing engineer who became chief executive in 2015, the company has said that it planned to bring more work back in-house for its newest planes."

I think that a lot of times management believes it's only job is to cut corners. In those cases, maybe it's better not to have managers.

Maybe if engineers were properly incentivized by profit sharing or something, they could help maximize profits without throwing out everything else like managers keep urging them to.

Lawyers and accountants become partners in their firms and are incentivized to work towards the long term health of the firm so younger lawyers will buy in and they can retire and continue to enjoy the profits of the firm.

What do Civil, mechanical, electrical, non-software Engineers have as a career path? None are in upper management. Maybe you get some stock options? You certainly don’t get to be in charge unless you start your own small, fledgling engineering company and then the chances are you only work on small projects since you aren’t big enough to sue if anything goes wrong you can’t get big contracts.

Hoping someone can shed some light on another path for career professional engineers.

I'm a big fan of the idea of codifying software "engineers" more clearly. When an architect, civil engineer, or doctor put their stamp on a job they are both civilally and in some cases criminally liable for that work. As such each profession has a professional society and legal framework for qualifying their members.

If you are building a webapp, it's probably fine that you haven't taken an ethical code, been legally qualified to write the code, etc, but those "engineers" that write code that keeps airplanes in the air or driverless cars on the road have, in my opinion, crossed that line into actual engineering work and as such should hold themselves to the same standards as other professionals in the space.

In the case of MCAS there would be a codebase with a signed commit by the responsible engineer, his or her name would be right there in bold when the first version was shipped and when hack upon hack was piled on. Of course this is no panacea but this has to be the way forward as software continues to extend it's reach into mission critical roles in our lives, right?

In aviation, DERs (Designated Engineering Representatives) fill that role [0]. They are responsible for ensuring rules/regulations are met, and need to sign off on everything before it's sent to the FAA.

[0] https://www.faa.gov/other_visit/aviation_industry/designees_...

So these software engineers would be referred to as Systems and Equipment Engineering leads in this case?

Take this post with a grain of salt - I've never worked directly with a DER. Hopefully others can correct me where needed.

> So these software engineers would be referred to as Systems and Equipment Engineering leads in this case?

I suppose you could say so, however my understanding has always been that DERs are highly specialized. "Systems and Equipment" covers a wide variety of aircraft components; I would think that's more of a category than a title. As a software engineer, on a project level, we've always simply referred to them as "The DER".

While DERs certainly have a background in engineering, they don't actively help create the product - I wouldn't classify them as a "Engineering Lead". When it comes to designing something, its safety is derived from different internal groups with different goals continuously evaluating each other's work. At each level, everyone signs off on the work they performed or reviewed. As I understand it, the DER oversees all of this, and on a technical level, is ultimately responsible for convincing the FAA everything is safe. On the software side of things, whenever there's a release, a DER has to sign off on it.

If you have an ABET certified computer engineering degree, you can become a licensed engineer. Admittedly, I'm surprised that isn't a requirement for these things sometimes.

Lots of oil and gas CEOs are engineers by trade. And as an industry they seem to value field service in their leadership.

My takeaway is embrace cartels, they're immune to market pressures and they're fundamentally conservative.

Actually profit as a motivation does not work for highly skilled work. If you put a monetary reward for solving a problem the solution usually gets worse.

"Based on studies done at MIT and other universities,[4] higher pay and bonuses resulted in better performance ONLY if the task consisted of basic, mechanical skills. It worked for problems with a defined set of steps and a single answer. If the task involved cognitive skills, decision-making, creativity, or higher-order thinking, higher pay resulted in lower performance" source: book The Drive by Dan Pink https://en.wikipedia.org/wiki/Drive:_The_Surprising_Truth_Ab...

Ted talk https://www.ted.com/talks/dan_pink_on_motivation

That's exactly the type of hogwash executives promote in order to avoid rewarding engineers so they can keep more profits for themselves.

There's one very important thing keeping me and everybody from destitution: an income.

Profit motive might not make me more creative (debatable) but it will certainly enable me to work for years on end.

Actually profit as a motivation does not work for highly skilled work.

If that is true then it logically follows that management is not highly skilled work, no?

Do the people who come up with these theories even bother to think it through at all?

Maybe they did, and the correct conclusion is that (excessively) high pay for management is entirely unjustified?

I was trying to say that high compensation levels on management level may not lead to the best results for the company. Neither may chasing quarterly profits to have a good stock level.

Thus engineering goals and management goals may not be the same.

I take it you couldn't be bothered to read the links or look at the research behind the theory before ridiculing it? The idea is that after a certain point, financial motivation doesn't lead to better work - to motivate you need to fulfill higher-level needs like ownership, purpose, autonomy, etc.

Your statement about management not being skilled work makes no sense. Do you have some evidence that paying managers more leads to better performance?

Do you have some evidence that paying managers more leads to better performance?

There is no such evidence yet manager’s reward themselves and their peers lavishly anyway. That’s the point. The research is a scam to justify underpaying skilled workers so management can keep it all for themselves.

That is literally the opposite of what the research says. It says that after a certain point, financial reward is not a good way to encourage strong performance. It says low-wage workers can be incentivized with higher pay, but high-wage workers need more intrinsic reward because pay is not an effective incentive once people are financially comfortable.

At least open the wikipedia page before making such completely incorrect statements.

financial reward is not a good way to encourage strong performance

And yet managers are highly paid and incentivised with large bonuses. How do you explain that paradox if pay is not an incentive? Managers can reward themselves and their peers with literally anything and they always choose more money. How do you explain it?

Theory and Wikipedia are fine but look at what actually happens in the real world. I read the original research in HBR when it was first published by the way. I didn’t believe it then either.

Because people prioritize their personal finances over company performance? That doesn't sound surprising to me.

This is also relatively new and it takes time for traditions to change. When it comes to the real world, you can clearly see this research in action, particularly in the world of software. There has been a shift from offering pure financial incentive (e.g. working in finance) towards more intrinsic rewards (mission-driven, autonomy, ownership, etc that most top-end software companies prioritize)

mission-driven, autonomy, ownership, etc that most top-end software companies prioritize

“Top-end software companies” like say Google? Once again you look at the real world and find that companies find a reliable means of attracting and retaining staff is high pay.

What if you got paid your bonus not in cash but in paid time off? You could use that time for more self-actualisation. But (pretty much) zero companies offer it and no employees demonstrate a preference for it. In the real world money is what people want.

Uh Google is one of the most obvious examples of a company spending a ton of money to boost the non-financial rewards of being an employee. In fact, Google is one of the best examples of this research being applied successfully in the real world...

I don't think you have a very open mind in this discussion. People who aren't struggling to pay their bills prioritize things like autonomy, happiness and purpose above cash all the time.

Engineers of yore were trained not only in the technical skills, but also in economic analysis and management of systems. They were expected to take a leadership role as their career progressed in large projects. The treky model of engineer as skilled technician has not been helpful.

Taking such a narrow view, many engineers fail to provide adequate and effective input into the program, and managers learn to ignore their concerns as not relevant to the larger picture.

I think a much better view of the issue is in the NY Times https://www.nytimes.com/2019/09/18/magazine/boeing-737-max-c...

This is a great article and now I am appalled at so many things. But the line that touched me profoundly is at the end where the father and the husband of 4 victims says this:

“I am empty,” he told the committee. “My life has no meaning.”

I can only imagine the pain he is experiencing.

Everybody should read David Packard's address to HP Managers in 1960. This was a time when HP was truly a great company.


To me, this is what i consider as true "Management" and not the BS that passes for it today.

I like to imagine a fantasy United States where the FAA (et al) would have stepped in, removed all the MBAs, and set the company on a sane course. More practically, I wonder what could be done to create structural impediments to the kind of managerial death-cult described in the article. For instance, I see no reason a company should be allowed to buy its own stock. How about requiring corporate boards to have at least 51% of non-stockholders? We should remember that corporations are purely social constructs. There is no natural law that they must not be interfered with, that only market mechanisms should control then. We created corporations for social good, and can alter them.

This call-out quote hits very close to home for me:

You are in a mature industry that is no longer innovative; it’s a commodity business. The last great innovation capable of driving major growth in aviation was the jet engine back in the 1950s, and every technological advance since has been incremental. And so the emphasis of the business is going to switch away from engineering and toward supply-chain management. Because every mature company has to isolate which parts of its business add value, and delegate the more commoditylike things to the supply chain. The more you look to the market for pricing signals, the more the role of the engineer will shrink.

I am as outraged at Boeing's ineptitude as anybody with a rational mind, but I can't read this kind of article without cringing. Sentences like "about two weeks after the system’s unthinkable stupidity drove the two-month-old plane and all 189 people on it to a horrific death" should raise alarm bells in any reader's mind. This is probably a writer who knows nothing about aviation and/or software engineering yet feels very comfortable calling a system "unthinkably stupid". This smacks of oversimplification and just plain sloppiness. There are many articles on the entire debacle that are a lot more measured, and smart about their analysis, including quite a few great reads that were mentioned here in HN.

I used to work as a controls engineer. Sensor fusion and redundancy is one of the most basic concepts in controls, and yet it was totally absent from a system that was responsible for flying the aircraft. There is more redundancy built into your office building's HVAC system than was built into the 737's MCAS. That really is inexcusable.

Actually, per the article, the lack of redundancy was central to hiding the feature from the FAA, keeping it out of the FOM, and ensuring the avoidance of any scenario that would entail extra pilot training. Whether two or three sensors, any disagreement among them would have involved a more complex system, and would risk necessitating an in-cockpit notification of that disagreement, and the ensuing training so the pilots understand the exact consequences of the disagreement and a procedure for mitigating it.

And the article explicitly dings Southwest Airlines as having provided Boeing a financial incentive to avoiding them needing additional simulator training.

> Whether two or three sensors, any disagreement among them would have involved a more complex system, and would risk necessitating an in-cockpit notification of that disagreement, and the ensuing training so the pilots understand the exact consequences of the disagreement and a procedure for mitigating it.

The in-cockpit notification light, "AOA Disagree," was a paid upgrade [1].

I don't know what the article is on about.

[1] https://www.flightglobal.com/news/articles/boeing-bosses-una...

a. The AoA disagree alert is standard, not a paid upgrade.

b. The AoA indicator is a paid upgrade.

c. In most 737 MAX's, the disagree alert did not function correctly (did not function as they supposedly intended) unless you bought the paid upgrade for the indicator.

d. Neither the alert nor indicator were considered safety equipment, they were considered advisory.

e. If the alert indicates a caution or warning, it's going to be listed in the flight manual, and decently likely a procedure for understanding and handling that condition must exist, and if so it's going to be a part of a training regimen, the very thing airlines wanted to avoid.

The two changes already planned: decoupling the disagree alert and angle indicator. Upon disagreement, MCAS is disabled. That means the disagree alert is at least cautionary now, because it means the airplane's stall behavior will be different than other 737s. That absolutely will require simulator training for pilots.

One of the most central questions is whether the assessment that the disagree alert was not "safety equipment" was wrong. That MCAS, in an angle of attack disagree condition, can so quickly induce, entirely on its own, catastrophic mistrim at low altitude, is rather damning. It suggests the risk assessment process is flawed.

Indeed, it's the motive to have done things wrong rather than in any type of correct way.

I'm an aerospace engineer who does work for the U.S. Government. When it comes to journalism, I'm a stickler for not oversimplifying the technical in attempt to make things understandable to the layperson or to support feeling-based arguments. I definitely agree the author has an unapologetic opinion/viewpoint in the writing, and I too was looking for where they oversimplified technical detail.

The sentence after "This alteration created a shift in the plane’s center of gravity pronounced enough that it raised a red flag when the MAX was still just a model plane about the size of an eagle, running tests in a wind tunnel" marks a bloc of technically sloppy oversimplification.

That criticism noted, I couldn't materially fault anything else in the writing and in fact I learned some things about my own industry. On the whole, it is a really valuable piece that I will be sharing with my colleagues.

> Sentences like "about two weeks after the system’s unthinkable stupidity drove the two-month-old plane and all 189 people on it to a horrific death" should raise alarm bells in any reader's mind.

Yes, exactly: that's the point. Yes, a lot of technical detail is left out, but the average lay person doesn't care about the technical detail. They just want the bottom line. And as someone who has spent quite a bit of time digging into the technical details, I have a hard time quarreling with the language this author chose to describe the bottom line. This is not a case of "well, it's complicated". This really is a simple case at bottom: an unthinkably stupid system, a product of a frighteningly common managerial and corporate outlook, killed everyone on board two airplanes. And the author wants that to raise alarm bells in every reader's mind--so that maybe those readers will start wondering whether letting such a system continue unrestrained is a good idea.

The article is focusing on the management part and is missing in the technical details but still presents those questionable things the hardware + software did . I mean how would you call adding hack over hack and other hack to solve a fundamental issue , sounds very familiar with the software industry where when you are rushed to put a fix ASAP and you put a ugly hack in and write a TODO that this should be redone properly. Also you have the trim wheels that you can't use them in all conditions,that is stupid IMO, there were designed probably 50 years ago and never updated when the plane size and engine were increased.

It seems like a pretty logical inference looking at the evidence.

You should see her obit of Steve Jobs which makes this look like a hagiography

The author states:

>"Southwest always had a lot to say about projected modifications to the 737, and Kelleher’s team mostly wanted as few technical modifications as possible. With the MAX, they upped the ante: According to Rick Ludtke, a former Boeing employee, Boeing agreed to rebate Southwest $1 million for every MAX it bought, if the FAA required level-D simulator training for the carrier’s pilots"

This seems like an important part of the story yet I'm not understanding if it's inclusion is meant to have significance to the tragedies of the 737. Could someone explain how of if this is germane to the crashes?

It's an incentive to make the differences between the planes seem as minimal as possible and avoid pilots being properly informed about all the details of the system, leading to pilots not knowing why the plane does what it does and how to fix it when it goes wrong?

The moral of this story seems to be "buy Airbus shares".

Maybe Lockheed Martin will buy up Boeing as distressed assets and do something with it?

Lack of openness, and tolerance for politics at corporate offices ... jfk ... bane of many people's misery. VW is another one except that it was an on purpose.

How can you guys take an article seriously that can’t even be bothered to get basic factual information correctly.

Just a symptom of late stage Capitalism.

> It is understood, now more than ever, that capitalism does half-assed things like that

Such a non-insightful aside. Boeing produced many excellent planes, and during that time functioned quite well in a capitalist system. In fact, was enabled by it. As if a non-functioning managerial class never could happen in a socialist bureaucracy.

Buying MD seems like it was mostly motivated by the profitability of defense contracts, which usually just end up distorting a company's priorities, and which are driven by decidedly non-capitalistic requirements of the military. HP avoided defense contracts for decades for this very reason.

Hopefully Boeing will return to its capitalist roots and realize that it needs to use the capital it has to satisfy the needs of its customers by building great planes.

I work with these large aerospace primes and see no consistent display of vision, long-term thinking, or integrity among them. I'm not saying these qualities don't exist within their ranks, just that they never become overriding management values. Take a look at the qualifications of Lockheed Martin's CEO or the President of Boeing Defense & Space. Yet, as the article notes, despite being helmed by non-engineers in undoubtedly one of the most technically-demanding sectors in society, these companies are richly rewarded in the public marketplace. Wall Street clearly does not give a shit that these "leaders" don't even truly understand their products.

Pretending that all these companies need to do is Adam Smith harder is crazy. As long as they have $300-$400+ share prices, there is no market incentive for them to not suck.

On the other hand, "Air Disasters" S13 E2 is about the crash of an Airbus A300 that stalled and crashed. The reason was because Airbus changed the way the "go round" mode worked, and the pilots were not aware of it nor were trained on it. The pilots kept pushing the nose down and the automation kept pulling it up, until it crashed.

Airbus also has an Airworthiness Directive out on it for pitch instability:

" This AD was prompted by analysis of the behavior of the elevator aileron computer (ELAC) L102 that revealed that excessive pitch attitude can occur in certain conditions and during specific maneuvers. This AD requires revising the airplane flight manual (AFM) to incorporate updated procedures and operational limitations, as specified in a European Union Aviation Safety Agency (EASA) AD, which is incorporated by reference. The FAA is issuing this AD to address the unsafe condition on these products."


Air France 447, Qantas 32, and Qantas 72 all highlighted significant flaws in Airbus' Guidance/Nav/Controls subsystems and their automation approach. That doesn't make the article's criticisms of Boeing the slightest bit less valid.

I didn't say the criticisms weren't valid. Why isn't Airbus undergoing the same scrutiny? Causing a crash and losing everyone on board is just as serious a problem.

Somehow I miss how is S13 E2 in any way relevant to the topic of the article here discussed, which is, as far as I understand, just about the problems brought to Boeing by Boeing's own management.

It's relevant in that Airbus management is not being raked over the coals for that disaster, which has many similarities.

Did they do their regulatory evaluation themselves, hiding what they produced, still claimed that what they did was "not important" after one crash, tried to blame the pilots, waited for a second crash, again claimed that it's the pilots responsible for the second crash, insisted for the planes to keep flying after the second crash? Was it all the effect of their decision to lie about the facts of design of the whole plane to everybody?

I still consider what Boeing did hard to match with any other event, except maybe with some other Boeing actions.

Nearly TWO DECADES before Boeing’s MCAS system crashed two of the plane-maker’s brand-new 737 MAX jets, Stan Sorscher knew his company’s increasingly toxic mode of operating would create a disaster of some kind. A long and proud “safety culture” was rapidly being replaced, he argued, with “a culture of financial bullshit, a culture of groupthink.” (Emphasis added).

But, the situation is that US air travel got safer during a lot of those two decades. Indeed, the two crashes world aren't enough to push air travel back two decades and safety standards and the rarity of crashes were enough catch the problem.

The 737 Max is a huge problem for Boeing in particular, creating a situation where the endgame is hard to see. But not as much for air travel generally, which continues to be cheap and safe. And Boeing made a lot of planes and lot of money in the last two decades.

Which is to say when things are "pushed to boards", it's hard to say exactly when they give out. The 737 Max system was about minimizing costs everywhere in the system - costs in testing, costs in training and costs in operation. The multilevel Rube Goldberg approach finally failed but it's had a long run, the 737 800 is a workhorse of air travel.

Still, I suppose the lesson is multilevel optimization can result in a catastrophic systems failure. There are probably much processes where this could happen - just think when all the cars are Internet connected.

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