I highly doubt it's forgotten just because it's not at the top of the news cycle, I will call that tragedy an avoidable accident likely caused by a sudden cargo shift perhaps due to improperly secured cargo, likely due to safety measures bypassed by a contracted freight operator under pressure from Amazon to make its schedule.

Since the MCAS issue was already known to everyone, and a software fix was still pending, what is beyond me is why the system was allowed to be left active at all - there should have been an order to either disable the system, or voluntarily ground MAX aircraft if the operator did not want to do that.

There may have (and should have) been a relative risk assessment of the options done before deciding whether to do any of the three (disable, ground, leave as is) or probably any of a number of other options while waiting for corrective action. It may be that leaving the MCAS active was seen as the lowest risk option - Boeing may have statistics on how often it engages and in what conditions to support the decision.

And with only 14 in the fleet.

Yes. I'm not an expert on the details of the certification process, but one of the things that's been mentioned on this thread is that Boeing convinced the FAA that the 737 MAX was a modified 737, meaning that a pilot who is rated for the 737 MAX will also be rated for other 737 models and can move between them.

One of the changes I can foresee being implemented, is that pilot [re]certification will be required at a much lower plane-change threshold than what is currently thought to be sufficient.

Those who fly the MAX9 for UA spend most of their time flying -700, -800, and -900 737s.

Do note that there have been some recent cracks (or what seems like cracks), as this video claims. Yes, it's another airline, but it's relevant as we're talking US commercial airline culture.


Of course, we're never sure if this is more political than anything else. But safety is something that we should never take for granted. Slacking and cost-cutting in safety is a continuous risk, and although we like to think the opposite, US aviation is not immune from that.

See, I would argue the PCU servo issue was even more serious than the MAX problems... there wasn't a well-established procedure in place to recover from such a scenario, and the Eastwind flight only safely landed because the servo failure spontaneously resolved. The Lion Air crash, though not in final yet, most proximately involves a flight crew failing to recognize a scenario for which they had been trained (runaway stabilizer) and continuing to fight a losing battle. It's a pretty textbook case of being 'behind the airplane'. If the ADS-B data profile shows great similarity to Lion Air 610, it's probably another tragic case of the same response. Did a poorly-designed system contribute? More than likely. But it's always the result of a panoply of factors.

As you note, the commercial interests won the day in the 90s, and the 737 was not grounded. You'll also recall that Boeing went to great lengths to characterize the incident as an improper response to wake turbulence.

All this is exactly why I'm not in favor of letting social pressure and political influence dictate safety decisions made by the FAA and other regulators. Would it have been more appropriate not to ground the MAX if this series of accidents took place five years from now, with upwards of a thousand in service?

Because, absent an instrumentation failure (faulty AOA sensor), the MCAS actually protects against a stall when flying under certain conditions. That's why the Boeing fix won't disable, inactivate or substantially modify the system; instead, it will probably include functions to automatically inhibit it and alert pilots when there is evidence of an instrument malfunction, like an AOA disagree. The simplest fix appears to be a software update to the MCAS logic to accept two AOA input channels, allowing a comparison of data to signal some sort of instrumentation failure.

What's most baffling to me is why this kind of fail-safe logic wasn't a part of the system in the first place... it seems pretty intuitive and is consistent with redundancy of other automated flight control augmentation systems, like yaw damper. Instead, the MCAS went into production models with a single-channel input for AOA data, which came from only one AOA sensor, creating a single point of failure. This starts the cascade of failures and worst-case-scenarios (there were several) that most likely led to the Lion Air crash and is now implicated in the ET302 disaster.

I know what the MCAS does and it's purpose - I maintain all of these "techie" aircraft gizmos designed to enhance flying safety put too much decision making control in the hands of computers, or at the very least, may deny the pilot access to accurate information about the aircraft's current condition due to misinterpretations or faulty data collection. When I am flying under instruments I always know my life depends on trusting my instruments, but if there is a computer system standing between the mechanical collection of flight conditions and my instruments, now I need to think about whether my instruments are accurate, and is the computer accurate. Honestly, that makes me feel uneasy.

I've been flying airplanes for a very long time, and maybe I'm too much of an old-timer because I don't fly airplanes full of computers except the single one in my head, but since the advent of computerized flight control and management systems, I am not sure the statistics lead to a long term conclusion of improved flight safety when compared to the era of aircraft with fewer computerized systems, and many flight and aeronautical science professionals feel a similar way about this issue.

Agree, but is it 100% confirmed that there was actually a single input channel? It's very hard to imagine several levels of Boeing engineers would've signed that off.

The pilot can always overrule the computer or the MCAS for that matter.

I fully expect a culture clash between older generations who are fully aware of how computers can fail and younger generations who've never seen a Blue Screen of Death. But the fact is that safety has continued to improve drastically since the advent of the computer on the flight deck, all while "pilot error" has become a more dominant cause of accidents. And the defensive way the pilot unions are reacting right now seems to indicate that pilot error is again being considered for the recent 2 accidents.

As you'll probably agree, the ensuing discussion will not be about the amount of automatization on the flight deck. No big aircraft producer has indicated a desire or need to reduce that amount, even while they've been hearing the calls from pilot unions about the decrease of hand-flying skills for years. Boeing has indicated that - because of a lack of engineering background - it struggles to convey necessary technical information to flight crews in a way that they can comprehend. The discussion will probably shift more towards how we should define the role of the pilot and the computer on a flight deck that is expected to be increasingly more complex and safe in the years forward, and as you may guess, it may not bode too well for the pilots.

Is a yaw damper a "techie aircraft gizmo"? What about Mach trim? Conceptually, no different from MCAS, and an uncorrected failure can absolutely result in an upset from which recovery is virtually impossible. Like I said, my issue is the lack of redundancy.

As far as the last point about whether higher levels of automation have led to improved flight safety as compared to the era of less automation, the statistics are pretty clear that they absolutely do. Automation dependency is an issue of unintended consequences stemming from increased automation, and over the last twenty years we've seen professional pilot training begin to re-emphasize manual flying skills. But are you really suggesting that flying is demonstrably less safe in the current era than it was 20, 30, 40 years ago, and that there are credible professionals who actually believe it?

I could be wrong, but everything I've read indicates that it was. I'm not sure this particular failure mode (faulty AOA input, improper MCAS activation, failure of crew to react appropriately to stab runaway, loss of pitch control) was found to be a reasonable possibility... but those kinds of engineering errors are not unheard of.