[danmaz74]

Airlines are forced to certificate their pilots for the plane they're going to fly. How many lives have been saved by this regulation, in comparison with those which were lost because Boeing tried to cheat at said regulation? Ignoring that is the same as pointing to people killed by airbags, without considering those saved by them.

[AnthonyMouse]

That's the false dichotomy again.

One example of a third way would be that when a plane is 97% the same as one you're certified on, have something between no and full recertification which is only 3% as time consuming because it only covers the differences between the planes instead of going back to square one.

[breatheoften]

I think you are also guilty of over constraining the solution space ... airlines want a new more fuel efficient plane with a certain size and passenger carrying capacity. Building a new plane for this need either involves training pilots on this new plane type or inherently unsafe engineering to deliver the new product without requiring the pilot retraining expense.

You suggest that we should sacrifice the rigor of pilot training to allow delivering planes without safety compromises ... I think there’s absolutely no justification for safety compromise in this scenario - the new plane will pay for retraining costs in a slightly longer amount of time than it would’ve had retraining not been required — oh well. That’s the cost of doing business and it’s absolutely without doubt that in this case - designing a plane that wasn’t inherently flawed would’ve been a better strategy for Boeing and for all of their 737 Max customers ...

The real question to learn from this tragedy is not how can we make it slightly less expensive to produce a new air frame — it’s how can we ensure that safety is not allowed to be compromised during the design phase of new passenger aircraft in service of _truly_ minor cost savings ... if we had a real regulator that was not captured - they should’ve had the flexibility to say no to the 737 Max very very early in the proposal process — and if necessary they should’ve been allowed to offer a tax incentive to offset some of the pilot retraining cost required by them not budging on safety requirements.

Building and deploying safe airplanes into real world use is more expensive than building unsafe airplanes — but that doesn’t mean it’s not worth the cost to society...

[AnthonyMouse]

> You suggest that we should sacrifice the rigor of pilot training to allow delivering planes without safety compromises

Describing this as "sacrifice the rigor of pilot training" is illegitimate. If someone is already certified on a similar plane, the training they need is really only on the difference between the planes.

> I think there’s absolutely no justification for safety compromise in this scenario - the new plane will pay for retraining costs in a slightly longer amount of time than it would’ve had retraining not been required — oh well.

The problem is that monetary costs are still safety costs. Now the new thing costs more and people choose alternatives that may be less safe. If new planes are more expensive then older planes that are more likely to suffer wear-based mechanical failures stay in service longer. If air travel costs more then more people drive.

Boeing obviously screwed this up, but we gave them the incentive to screw it up, and that's on us too.

> The real question to learn from this tragedy is not how can we make it slightly less expensive to produce a new air frame — it’s how can we ensure that safety is not allowed to be compromised during the design phase of new passenger aircraft in service of _truly_ minor cost savings

But that is how you do it. You figure out how to achieve the cost savings without compromising safety. There is no law of nature that says they're mutually exclusive, and without forcing them to be at odds there is no longer an incentive to choose the wrong one.

[Eopia]

>Boeing obviously screwed this up, but we gave them the incentive to screw it up, and that's on us too.

No WE didn't. THEY wanted to cut costs and it appears to me THEY cut corners to achieve it. This is like saying the professor made me cheat by making the exam artificially complicated and learning would have cost to much time.

[AnthonyMouse]

Yes WE did.

This isn't a game or a class. If you have rules that result in preventable deaths in actual reality then you have bad rules. Even when there is plenty of blame to go around.

[tempguy9999]

> ...but we gave them the incentive to screw it up,...

> ...Yes WE did...

Who is WE here? The general public who use the planes? The airlines who bought the planes?

The public and the airlines relied on the FAA. The FAA allowed boeing to self-cert. The FAA's job was safety. The FAA failed in that. It was their job and they didn't do it properly. It's the FAA at fault not some amorphous WE.

[NickNameNick]

Training to fly two very similar planes can be more dangerous than flying to completely different ones.

For example, some marks of Spitfire had engines that spun in the opposite direction to the rest of them. The aircraft were otherwise similar.

If a pilot gunned the engine during the take-off roll, the torque reaction will cause it to swerve. The pilot can catch it with the rudder if they are expecting it. If the engine is spinning the wrong way, and the pilot inputs opposite rudder, the aircraft will wreck it's undercarriage and roll over on the runway.

Whilst training to fly any variant, when a pilot has experience in other variants is obviously going to be quick and easy, expecting them to swap back and forth between variants is also clearly unsafe.

[cmurf]

The aircraft type rating is exactly difference training. You don't start from scratch just because a type certificate differs between two airplanes. The whole point is to require that difference training, and for it to be specifically laid out.

Whereas without that, what did Boeing do? They gave pilots an iPad for difference training, and kept them in the dark about the differences between 737 NG and 737 MAX. That was their choice. The lie of omission by manufacturers is the whole reason why we have type certifications in the first place.

[AnthonyMouse]

> The aircraft type rating is exactly difference training. You don't start from scratch just because a type certificate differs between two airplanes. The whole point is to require that difference training, and for it to be specifically laid out.

You're talking about training for completely different planes. Type training is commonly multiple weeks long. What I'm suggesting is something intermediate for two variants of a plane that are much more alike than that ("subtype training"), so that it's hours or days rather than weeks because there aren't as many differences to cover between two 737 variants than between a 737 and a 787.

> Whereas without that, what did Boeing do? They gave pilots an iPad for difference training, and kept them in the dark about the differences between 737 NG and 737 MAX.

Right, exactly. So they have the tiers too far apart. There is too big a jump between "here's an iPad" and tens of hours of training over multiple weeks, so there's a need for something in between for smaller variations.

[cmurf]

I'm not talking about completely different planes. If you have a Boeing 737 type rating, obtaining a 757/767 type rating is more expedient than it is for someone who has no prior Boeing 737 type rating.

Further the regulations say whether a type certificate is required, and what goes into it, is up to the administrator. It's not strictly defined. You can in in effect have subtype ratings.

FAA regulations are very much dependent on delegating authority rather than detailing every nitpicky thing. They really aren't that complicated.

[danmaz74]

I'm all for having good regulations. The dichotomy comes from those which say that all deregulation is good (which implies that all regulation is bad).

In this specific case, though, I suspect that measuring the "X% the same", and what the new training should be, is much more difficult than it looks at a first glance.