A CNN story on a WSJ story:

Boeing says it has been working on a software fix since the Indonesian crash. I'm not sure if they have actually been working on it or have just been talking about working on it until the Ethiopian crash. It's reportedly due to come out "no later than April". Does someone with more knowledge have an opinion on what a reasonable software update timeline would be?

Depends on the complexity and also when the requirements and scope were locked down, which seems to have been an issue with Boeing and the FAA not agreeing on that (and then further delayed by govt shutdown if they didn’t get to agreement with the FAA before Christmas). They may also have been waiting on more data from the first crash and close call before locking it down.

Then I’m sure they do much more extensive QA than most software and have to consider all of the edge cases, not just the 95% of cases.

If they rush out a fix it’s possible it could make it more dangerous bc the fix affects something else in a negative way, so they don’t do 2 week sprints and release as they go.

The de Havilland Comet 1, the first jet airliner, was notoriously grounded for ever in 1954 after it killed 110 passengers and crew in a series of accidents over a year (mostly due to metal fatigue, poor window design and insufficiently thick skin, the effects of regular pressurised jet flight being little understood at the time). The Boeing 737 Max has just killed 346 passengers and crew in two accidents over about five months. The aircraft has a thrust rating and centre of gravity well outside the envelope that the airframe was designed for and the 'fix' is an automatic computerised pitch control which the pilots cannot affect or override regardless of the manual or autopilot mode selected. If the computer wants to pitch down (and pile the aircraft into the ground) for whatever reason, it'll pitch down, and there's nothing the pilot can do. And you know what computers are like.

This appears to be a bit of a problem.

Where did you get that from?

I’ve seen that they can override it but need to know how to

Martinggale fallacy?

I’ve only seen that in a few media reports and less frequently than the reports that pilots can override it. Any sort of official report or release hasn’t mentioned the pilots being unable to turn the system off.

This is an interesting article in a bunch of the issues with the MAX
https://www.theatlantic.com/notes/20...37-max/584791/

Flights aren’t independent. An accident and the things learned/recommendations/changes that come after that accident change the probability of future flights crashing.

There is the huge legal liability basically admitting there was a defect in the design and programming.

NYT update on pilot communications right after take off and also graphics showing similar odd takeoff patterns of both flights.

I never know how to embed links

https://www.google.com/amp/s/www.nyt...lines.amp.html

Shocked to hear American Airlines has 328 Max 8 planes. Boing has back orders for 4000 plus planes

Dumb layman question but jet travel is indisputably safe and has been for a long time... So why the need for this MCAS system at all? Seems to add another level of complexity to solve a non existent problem.

.

What have we learned from these accidents? Isn't this an ongoing investigation?

Eventually we will learn something, I just don't think anything has changed yet, hence the grounding of the aircraft seems justified.

They released a directive to make pilots aware of a potentially fatal situation and the appropriate actions to take. Airlines became aware of the fact that a particular sensor needed to be working correctly for a plane to be airworthy. Of course things changed. Which is why your martingale comment was silly.

This idea that they don’t learn anything until the final report is also silly. It’s not like investigators keep all this information to themselves until the report is released. In lots of cases actions are taken to make planes safer long before a final cause is officially reported.

You don't actually know for sure what is causing this, if similar advice or reassurance was given after the Comet crashes, the actual probability would have been exponentially higher for additional flying hours with that aircraft.

A similarly cumalitive fault is possible but unlikely in this case, it still has to be considered in our risk calculations.

You keep moving the goal posts.

Letting corporations regulate themselves is bad.

Similar story. https://www.bloomberg.com/news/artic...workers-warned

Now...
Prosecutors, Transportation Department Scrutinize Development of Boeing’s 737 MAX



https://www.wsj.com/articles/faas-73...ed-11552868400

I would not underestimate the seriousness of this problem for both Boeing and the FAA. As mentioned previously, the 737 Max has now killed an awful lot more people than the Comet 1, apparently for technical reasons inherent to the aircraft (and not, for instance, weather or pilot error). In the case of the Comet crashes, neither de Havilland nor the Ministry of Aviation nor anyone else on Earth knew what was wrong before the Comet fleet was grounded. A Boeing engineer reputedly said, 'If if hadn't happened to de Havilland, it would have happened to us,' such was the novelty of pressurised jet flight. In the case of the second 737 Max disaster, it appears that the problem was in fact known, and yet the manufacturer and the authorities continued to allow the public to fly on a dangerous aircraft.

The liability is so great it's practically out of sight. So they'll do everything to try and get round it.

Of course, I forgot to mention: the liability may not only be civil. And you know what the other sort is.

The pilot's statement at the end rings a bit hollow as pilots did not make the decision to ground it.

Lion Air Flight 610 (the first 737 MAX 8 crash) had the same nose-dive error in the flight just prior to the fatal crash. Luckily there was an off-duty pilot who assisted the flight crew (who were struggling) in disabling the feature that causes the auto-correct. They landed safely and the plane was subsequently checked out and cleared to fly. Crashed 13 minutes later.

This is starting to look like it was a much bigger problem than pilots were aware of. Multiple occurrences of faulty sensor readings. It also looks like the pilot community wasn't thoroughly trained in deactivating the malfunctioning system.

Boeing is going to take a massive hit in reputation and probably money.

Grunching a bit to react to this theoretical question,

1% is WAY too high to be acceptable. Like unbelievably too high, not to mention the difficulty in assigning a reasonably accurate % above 0 to "We have no clue"

From the article I linked to:

Does anyone know what role simulators are playing right now, and how many are available with Boeing and outside of Boeing?

I am just assuming that right now Boeing has secured as many of the airlines 737 MAX configured simulators as possible and are running them 24/7.....?