Boeing's 737 Max Software Outsourced to $9-an-Hour Engineers

[Simples]

to the full and complete quote - this one:

"There is one and only one social responsibility of business – to use its resources and engage in activities designed to increase its profits so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud."

Clearly where Boeing erred is in the coverup/obfuscation/what have you - not the $9/hr programmers (if they did their job right and accuracy was verified). Seems they didn't and it it wasn't.

Optimizing the planet and ecosystem to increase currency? Why not just print more money, more currency to all. Oh, just more currency to a select few. QE fixed that already!

 

[Simples]

With over 30 years experience as a software engineer and engineering architect in a field requiring mission critical software ( financial systems ), I respectfully disagree on this point.

With lower wage programmers in the US, you will get inexperienced or lower quality individuals. With outsourcing, the turnover rates are huge and this multiplies the potential for quality problems with new personnel coming into the middle of projects. With quality, well-paid, self-motivated software developers, the bug rate is much, much lower, and people with pride in their work will not allow untested, minimally tested, or questionable code to even reach QA without effectively testing it thoroughly themselves. QA is the backstop (double-check, insurance policy, etc) not the primary method for avoiding quality problems.

QA is important, and you should pay those people well for sure, and a great QA resource is worth high dollars. But with great well paid developers, you won't be betting the company on QA resources, they will just add to the quality. You will also minimize developer turnover and allow people to build specialized domain knowledge which helps eliminate a lot of potential problems and improves efficiency for future projects.

There is a dangerous myth surrounding QA these days in that you can catch all bugs with automated testing, CICD pipelines, TDD, etc. It's simply not true, especially in the mission critical systems which have real world scenarios that are not able to be replicated in a build test. Problems involving multi-threading, race conditions, error handling ( especially compound errors ), surviving component failure, database deadlocks and such are very difficult to detect using systematic QA testing. Sometimes the error frequency is very low for certain scenarios, but people could die if it happens. It's also human nature to quit caring about certain types of errors if the frequency of it is low enough that it won't happen within a number of years, particularly when the coding cost to prevent it is high.

Many of the current crop of developers coming out of college have no idea about hardware architecture, do not understand system level machine behaviors, and are basically just coding up stuff by grabbing a bunch of open source software and slapping it together like tinker toys. There are some good ones out there, but if you have a specialized situation involving mission critical systems such as medical, flight systems, banking systems, nuclear systems, power systems, self-driving cars, etc. I would not recommend applying commodity programming resources there. Someone can and will die, and you don't want to be the victim of stupid corporate idiots trying to maximize their executive bonuses all in the name of being cheap and hiring the cheapest developers they can find.

In the name of lead time to market, QA has increasingly become the sacrificial sheep and bottleneck for the past 20 years already. There certainly have been advances in security, robustness, scalability and componentization as well, but overall, the customer experience has not improved very much. Stuff is relatively slower, more bug-ridden, unintuitive and driven more through social reinforcement than any inherent quality of the interface itself. Problem is, technological excellence won't give you new customers, as through market forces, people choose the cheapest and dirtiest alternatives. However, nothing beats hammering out excellence in every part of design and implementation from the very start. In such cases, QA would almost be an afterthought, though in the aviation industry, as we'll see in the self-driving car industry, QA-processes and regulations over decades are a necessity since the industry don't have the means to overcome competition AND regulate themselves!

 

[bone]

And Boeing is reaping the whirlwind for it’s stupidity.

The stock price has been sodomized accordingly by capitalism - I’d guess orders of magnitude more off the market cap than whatever they saved on software outsourcing. And reputations in safety-sensitive manufacturing are very difficult to mend.

 

[traderslair]

Welcome to my life for the last 20 years.

Fixing things from teams in India who always say Yes to land a contract while they have no clue what they are talking about.

Here's a link for you:

https://www.thehindubusinessline.co...are-development-jobs-study/article9652211.ece

95 per cent of engineers in India are not fit to take up software development jobs.

The study further noted that while more than 60 per cent candidates cannot even write code that compiles, only 1.4 per cent can write functionally correct and efficient code.

That is cheap India for you. Enjoy your flight.

 

[GRULSTMRNN]

I wonder then why many software end products that come out of American firms where there are reportedly a lot of Indians toiling are relatively decent.

So what is going on, can't be both? And if someone is such crappy developer how can they hold a job between h1b sponsorship and attaining a greencard?

Welcome to my life for the last 20 years.

Fixing things from teams in India who always say Yes to land a contract while they have no clue what they are talking about.

Here's a link for you:

https://www.thehindubusinessline.co...are-development-jobs-study/article9652211.ece

95 per cent of engineers in India are not fit to take up software development jobs.

The study further noted that while more than 60 per cent candidates cannot even write code that compiles, only 1.4 per cent can write functionally correct and efficient code.

That is cheap India for you. Enjoy your flight.

 

[Simples]

I wonder then why many software end products that come out of American firms where there are reportedly a lot of Indians toiling are relatively decent.

So what is going on, can't be both? And if someone is such crappy developer how can they hold a job between h1b sponsorship and attaining a greencard?

Any generalizations, for such huge swathes of land and diverse groups of people, are indeed wrong, as indicated above.

From experience, it's both, having worked abroad at such centers, most people who stay there are decent or excellent in their jobs. They have to be, to survive one hellhole, to get promoted to the next. Often times, it's these two factors: A) The sheer number of people applying for jobs, attempting to better their lives B) All the shithole exploitative circumstances surrounding such ventures. So there's always a promotion somewhere or someone leaving for one reason or another, and there's always fresh meat to the continuous meatgrinder. The rest is the result of circumstances and fallacious management. I guess customers are a-OK with this. If the market is saturated or moving, this perhaps won't be improving the situation either.

 

[Simples]

And Boeing is reaping the whirlwind for it’s stupidity.

The stock price has been sodomized accordingly by capitalism - I’d guess orders of magnitude more off the market cap than whatever they saved on software outsourcing. And reputations in safety-sensitive manufacturing are very difficult to mend.

Capitalism doesn't live in its own bubble. In other countries where restraint and regulations are non-existing, the same wouldn't make a dent to the ownership, who often may have foreign ties.

 

[GRULSTMRNN]

Generalizations to characterize individuals are wrong. But the average of generalizations are always correct by definition.

Any generalizations, for such huge swathes of land and diverse groups of people, are wrong.

From experience, it's both, having worked abroad at such centers, most people who stay there are decent or excellent in their jobs. They have to be, to survive one hellhole, to get promoted to the next. Often times, it's these two factors: A) The sheer number of people applying for jobs, attempting to better their lives B) All the shithole exploitative circumstances surrounding such ventures. So there's always a promotion somewhere or someone leaving for one reason or another, and there's always fresh meat to the continuous meatgrinder. The rest is the result of circumstances and fallacious management. I guess customers are a-OK with this. If the market is saturated or moving, this perhaps won't be improving the situation either.

 

[Simples]

Generalizations to characterize individuals are wrong. But the average of generalizations are always correct by definition.

Interesting take, and my comment wasn't meant as direct response. The average, mean, median, is statistics though, there are many, many averages to choose from. Correct by its own standards, but may be misinterpreted to gain incorrect or out-dated beliefs.

 

[newwurldmn]

I wonder then why many software end products that come out of American firms where there are reportedly a lot of Indians toiling are relatively decent.

So what is going on, can't be both? And if someone is such crappy developer how can they hold a job between h1b sponsorship and attaining a greencard?

The people who always complain are the software engineers who aren’t able to command top dollar for their talents and thus have to compete with lower cost workers.

It’s not unlike the auto workers complaining about the Asians and the Mexicans in the 1980s and 1990s.