Hacker News Reader

Comment by readams | Hacker News Reader

The underlying assumption here seems to be that if only the FAA had examined in detail every aspect of the development of the plane, of course they would have found all the problems. Underlying this assumption is the idea that Boeing is either trying intentionally to create unsafe planes (as though this is good for Boeing) or that the FAA could hire engineers that are way better than Boeing's engineers.

I don't buy it.

View on HN

The latter half of the article makes specific claims to how lack of FAA expertise and manpower, and increasing deference and complacency to the industry's self-reporting, led to missed signals. For example:

> When company engineers analyzed the change, they figured that the system had not become any riskier, according to two people familiar with Boeing’s discussions on the matter...So the company never submitted an updated safety assessment of those changes to the agency. In several briefings in 2016, an F.A.A. test pilot learned the details of the system from Boeing. But the two F.A.A. engineers didn’t understand that MCAS could move the tail as much as 2.5 degrees, according to two people familiar with their thinking.

> Under the impression the system was insignificant, officials didn’t require Boeing to tell pilots about MCAS. When the company asked to remove mention of MCAS from the pilot’s manual, the agency agreed.

The "change" mentioned is Boeing overhauling MCAS to have a four-fold increase in magnitude of control over the stabilizer (2.5 degrees vs 0.6 degrees). Your belief is that the FAA, even with a better-staffed review team, and less of a culture of rubber-stamping industry engineer assessments, would have reached the same conclusion? And/or do you think the MCAS change is actually not a problem it's scapegoated to be, and that Boeing was right to tell FAA to remove it from the pilot's manual?

I'm saying I don't buy that the FAA was ever likely to catch the problems with this system that the engineers at Boeing didn't catch. In hindsight it's easy to say the system was a problem and clearly it was, but the FAA was never going to catch anything.

The meeting with the FAA pilot you mention only goes to reinforce this; nobody in the FAA is ever going to have all the right context to put this together. Rather than trying to find some meeting with the FAA where someone could have caught it, there were probably many many meetings inside Boeing where it should have been caught.

No one intentionally makes shitty products. Or at least I’ve never heard of anyone who does.

But I can sympathise with the people that made the decisions to put the plane in the air.

I never intended to make products that customers wouldn’t like. But I’ve shipped things that didn’t meet my own (low?) standards due to financial and social pressures for me and the companies I worked for.

Most employees have done the same. Most business owners have as well.

Except that if you launch in production a bad web site to sell bicycles people don't die because of it. You need to make the difference between bad products and dangerous products - if you put in production a bad plane, car, medicine or gun people can die and there is no sympathy for that.

And you “don’t buy it” because? Blind faith in Boeing and the FAA not being corrupt?

Fact is, the FAA has outsourced large parts of certification to the very manufacturer whose plane it was to certify. Either physically (by using Boeing manpower) or conceptually (by not testing everything because Boeing told them nothing has changed even though that was not true).

Parent poster's claim seems to be that Boeing is incentivised to make safe planes, and is likely to have better aeronautical engineers than the regulator. In that scenario, it wouldn't matter if the regulators were corrupt or Boeing certified itself.

Unfortunately there's a massive hole in that argument, which is that (i) Boeing might be incentivised to do things which aren't in the public interest, such as take risks to make a great deal more money, and (ii) it might suffer a decline in competence due to contingent or management factors.

Perhaps in a perfect world incentives could be so perfectly aligned that no industry would need regulating. But that's not the world we live in.

If there aren't any ramifications for Boeing building unsafe planes, why would they be incentivized to do so? Shareholders don't care about the lives of the passengers, only ROI.

Boeing looks around at all of the slaps on the wrists other companies get, including bail outs. Why shouldn't they hop on that gravy train also?

This was not regarded as a new aircraft development but a variant of an existing design. This meant less scrutiny. Variants deliberately push the boundaries of what it means to be the same design to avoid the scrutiny given to a new design

Reply on news.ycombinator.com