Hacker News Reader

Comment by mehrdadn | Hacker News Reader

When is the media going to get the message that this wasn't a software issue?

It will eventually fly with only software fixes, which will reinforce that picture.

The final verdict isn't in, but it appears to have been a systems engineering problem with a software component.

The software performed exactly as designed. The problem is a faulty spec for how it should behave. The software is pretty much incidental.

You are correct, however... I can't imagine a single engineer with any kind of proper education (formal or not) implementing any piece of software or hardware without thinking:"One sensor? Really? What's the failure rate on this thing". That might have been enough to stop the chan of events.

You get $9 for hypothetical blood splattered on your hand or you get worse than nothing for nothing

Or may be they pointed it out but got reply "just work as said, you're just $9 engineer"? Or they were never told about the exact purpose?

The engineers may not have realized they were relying on one sensor, or they might be the $9/hour variety with no avionics experience who don't know any better.

The main problem was higher up. The overall design was bad, and Boeing didn't seem to have a handle on it.

I suspect they were told that runaway trim wouldn’t be a serious problem, and they believed it since they wouldn’t have the expertise to analyze the whole thing from first principles.

Or they have a "obey all orders even when it is blatantly criminal" clause.

Software is an integral part of modern flight. Somewhere up the chain, there’s a highly-paid software engineer/manager who helped design and signed off on the spec.

Would a highly paid software engineer be expected to understand the potential consequences of runaway trim? I’d think that would be up to the aerodynamics folks, who most likely said that it was not a problem.

If the issue was in the specs, then it is software. Producing code is only really one piece of software development.

Software was definitely part of the issue.

But it wasn't a software issue! /S https://philip.greenspun.com/blog/2019/04/08/boeing-737-max-...

Any competent engineer would raise moral and ethical questions when working on software that is "mission ciritical". Causing a passenger plane to continually go into a nose dive because of data coming from a single sensor is absolutely ludicrous. Just because the software was horribly designed, doesn't mean it isn't a software issue.

It is a hardware design issue. There's a concept of N+1 redundancy in safety-critical systems, a hardware design decision. This basic philosophy was missing from the MCAS system[1]. You can argue that the software could just ignore bad sensor readings, but that's just software making up for hardware deficiencies.

1. https://www.seattletimes.com/business/boeing-aerospace/a-lac...

But but but... there's a computer! It has to be a software issue!

Reply on news.ycombinator.com