←back to thread

388 points pseudolus | 3 comments | | HN request time: 0.013s | source
Show context
Bukhmanizer ◴[] No.43485838[source]
I’m surprised not many people talk about this, but a big reason corporations are able to do layoffs is just that they’re doing less. At my work we used to have thousands of ideas of small improvements to make things better for our users. Now we have one: AI. It’s not that we’re using AI to make all these small improvements, or even planning on it. We’re just… not doing them. And I don’t think my experience is very unique.
replies(21): >>43486104 #>>43486264 #>>43486456 #>>43487649 #>>43487671 #>>43488414 #>>43488436 #>>43488988 #>>43489201 #>>43489228 #>>43489488 #>>43489997 #>>43490451 #>>43490843 #>>43491273 #>>43491336 #>>43491568 #>>43491660 #>>43492193 #>>43492499 #>>43493656 #
baazaa ◴[] No.43488436[source]
I think people need to get used to the idea that the West is just going backwards in capability. Go watch CGI in a movie theatre and it's worse than 20 years ago, go home to play video games and the new releases are all remasters of 20 year old games because no-one knows how to do anything any more. And these are industries which should be seeing the most progress, things are even worse in hard-tech at Boeing or whatever.

Whenever people see old systems still in production (say things that are over 30 years old) the assumption is that management refused to fund the replacement. But if you look at replacement projects so many of them are such dismal failures that's management's reluctance to engage in fixing stuff is understandable.

From the outside, decline always looks like a choice, because the exact form the decline takes was chosen. The issue is that all the choices are bad.

replies(33): >>43488541 #>>43488644 #>>43488809 #>>43488874 #>>43488894 #>>43488954 #>>43489176 #>>43489496 #>>43489529 #>>43489552 #>>43489570 #>>43489702 #>>43490076 #>>43490205 #>>43490296 #>>43491212 #>>43491465 #>>43491538 #>>43491547 #>>43491626 #>>43491950 #>>43492095 #>>43492352 #>>43492362 #>>43492581 #>>43492773 #>>43492829 #>>43492886 #>>43493251 #>>43493711 #>>43495038 #>>43495649 #>>43495778 #
Art9681 ◴[] No.43489570[source]
You're just getting older and looking at the past with rose colored glasses. No one is going backwards in capability. It is about how accessible and cheap the thing is. In the 90's, a license to install Maya or 3D Studio Max, or Lightwave was extremely expensive, those products were not promoted nor available to the general public. They would cost tens of thousands of dollars, for the software alone, not to mention the hardware.

Today it is a commodity. So we are flooded with low effort productions.

With that being said, we have more capability than ever, at the cheapest cost ever. Whether businesses use that wisely is a different story.

There will always be outliers. I see many comments with people who derived value from whatever they perceived as something uncommon and unique they could do. Now AI has made those skills a commodity. So they lose their motivation since it becomes harder to attain some sort of adoration.

In any case, going forward, no matter what, there will be those who adopt the new tools and use them passionately to create things that are above and beyond the average. And folks will be on HN reminiscing about those people, 30 years from now.

replies(4): >>43489580 #>>43490428 #>>43490760 #>>43492946 #
yubblegum ◴[] No.43489580[source]
Boeing calls to say hello...
replies(3): >>43489725 #>>43489779 #>>43489869 #
SR2Z ◴[] No.43489869[source]
This is an excellent example. The 737 used to crash a LOT MORE than it does now, and even the version with MCAS is a safer airplane than your average 1990s jetliner (which was also probably a 737, which is kind of my point)[1].

So it's a slightly less safe (in the grand scheme of things) airliner that's vastly more fuel efficient and cheaper to run than any in the past. Obviously this is of no comfort to the families of the people who died in the crash!

But to suggest that Boeing has somehow regressed decades in technical capabilities is just plain wrong.

1: https://en.wikipedia.org/wiki/List_of_accidents_and_incident...

replies(6): >>43490090 #>>43490271 #>>43491284 #>>43491896 #>>43491913 #>>43495236 #
hylaride ◴[] No.43495236[source]
It wasn't that the 737 MAX isn't (or can't be) a safe aircraft, but that Boeing prioritized making it fly like its predecessor over all else - including safety.

The fact that two brand new MAX's crashed killing all aboard within 2 years of its commercial introduction (out of only ~600 models flying at the time) is a brutal safety record for the jet age, especially as the cause of the crash was the plane itself. That list you post includes any and all reported incidents that merely involve 737s (and involve incidents that were caused by factors that aren't necessarily related to the safety of the place itself).

replies(1): >>43499858 #
1. SR2Z ◴[] No.43499858[source]
> It wasn't that the 737 MAX isn't (or can't be) a safe aircraft, but that Boeing prioritized making it fly like its predecessor over all else - including safety.

It's misleading to say they prioritized making it fly like its predecessor over safety.

In theory there was absolutely nothing wrong with a system LIKE MCAS. In fact the 737 MAX is still approved to fly with it.

The flaws were in the specific implementation and documentation around it, not with the idea of the system itself.

> The fact that two brand new MAX's crashed killing all aboard within 2 years of its commercial introduction (out of only ~600 models flying at the time) is a brutal safety record for the jet age, especially as the cause of the crash was the plane itself.

If you want to be pedantic about it, the reason for the crashes is that the pilots failed to recognize trim runaway during takeoff. The trim runaway was caused by MCAS, but this is not a new failure mode for ANY aircraft and pilots get extensive training on how to manage it [1].

MCAS failing was not an unrecoverable error [2]. It failed several times in the US, as well, but American pilot training standards are very high compared to the places where there WERE disasters and the pilots recognized and recovered quickly.

I say this not to deflect blame from MCAS. Its original implementation was unsafe and should never have been approved.

A large part of why modern jetliners are so safe is exactly because of flight control augmentations like this - both Boeing and Airbus have been implementing these for decades and they have made flying much safer. Your suggestion that any system like MCAS is always unsafe (or that Boeing was somehow doing something wrong by adding it) is totally wrong.

1: https://www.aopa.org/news-and-media/all-news/2017/july/pilot... 2: https://www.nytimes.com/2019/09/18/magazine/boeing-737-max-c...

replies(2): >>43505337 #>>43505378 #
2. hylaride ◴[] No.43505337[source]
> It's misleading to say they prioritized making it fly like its predecessor over safety.

The internal Boeing emails literally say otherwise.

> In theory there was absolutely nothing wrong with a system LIKE MCAS. In fact the 737 MAX is still approved to fly with it.

I never said that MCAS had any issues in theory. And the 737 MAX was mostly "approved" by Boeing's self-regulators, where emails trails (again literally) had anybody raising questions or concerns sidelined.

> The flaws were in the specific implementation and documentation around it, not with the idea of the system itself.

Yes, because Boeing's top priority was making it so that no expensive extra training was required to fly the MAX, despite the fact that MCAS was designed to deal with some situations that could cause the plane to fly differently.

> If you want to be pedantic about it...

Yes, I am being pedantic about it. The trim issues in the crashes were (intermittently) caused by MCAS, but there was no specific documentation or specific training as to how to deal with it in the case of faulty MCAS sensors. There were indeed several MCAS incidents in western flights, but they were different as the failures were different. The two crashed pilots did indeed attempt disabling MCAS but the intermittent failures masked the problem and there were insufficient checklists by boeing, because had they existed it could have allueded to the fact that such situations may need new simulator time.

The MCAS issue was totally and completely recoverable if it were properly documented, but doing that would have almost certainly guaranteed the simulator time that was Boeing's top priority to avoid.

Almost all the reports about pilot capability differences had more to do with experience than it did training. These "developing" countries have younger airlines and pilots who don't have the same pipeline of pilots with decades of experience, including the military like in the US. MCAS "acted up" on several other Lion Air flights that the pilots corrected for as well, but again those were different failure modes.

The fact that the Ethiopian Airlines had perfectly acceptable safety record on other planes negates that these are "poorly trained" pilots. They've had one major accident in 2010 that was attributed to pilot error, but most of the rest were due to bad luck (eg bird strikes) or hijackings.

> A large part of why modern jetliners are so safe is exactly because of flight control augmentations like this...

I never even mentioned MCAS by name. Yes, modern jetliners are safe because of these kinds of systems. Airbus planes will not allow pilots to do many things no matter what, even. But these systems are documented, pilots trained on them, and go through rigorous testing because in most cases they're designed to make a plane safer, not try to deal with aerodynamic changes.

Boeing wanted no new simulator training despite the MAX being a very different aircraft due to changed engine placement. That was the cause. If Boeing wasn't trying to avoid new simulator training the 737 MAX is a perfectly fine aircraft as far as we know.

The 787 had similar issues as the overriding goal of the program was to get as much capital expenditure off of Boeing's books, but all of the outsourcing led to a nightmare when trying to assemble the plane and there was no unified quality control program, or even a straight line of responsibility.

3. ApolloFortyNine ◴[] No.43505378[source]
>A large part of why modern jetliners are so safe is exactly because of flight control augmentations like this - both Boeing and Airbus have been implementing these for decades and they have made flying much safer. Your suggestion that any system like MCAS is always unsafe (or that Boeing was somehow doing something wrong by adding it) is totally wrong.

The common person often doesn't realize this at all. Every modern plane is flying itself essentially, with hints from the pilot on what to actually do.

>MCAS failing was not an unrecoverable error [2]

Also this is frustrating, especially in the case of the second crash where every max pilot knew the procedure (including the one that crashed), they even performed the procedure but then disabled it a minute later. Both the NTSB and the BEA (French equivalent) agreed pilot error/CRM played a role in the second.