> "To address the unsafe condition, the FAA proposes to require four design changes: (1) installing updated flight control software (with new control laws) for the FCC operational program software (OPS), (2) installing updated MDS display processing computer (DPC) software to generate an AOA disagree alert, (3) revising certain AFM flightcrew operating procedures, and (4) changing the routing of horizontal stabilizer trim wires."

> "The first design change is intended to prevent erroneous MCAS activation. The second design change alerts the pilots that the airplane’s two AOA sensors are disagreeing by a certain amount indicating a potential AOA sensor failure. The third design change is intended to ensure that the flightcrew has the means to recognize and respond to erroneous stabilizer movement and the effects of a potential AOA sensor failure. The fourth design change is intended to restore compliance with the FAA’s latest wire separation safety standards."

Notably the FAA does not required a 3rd AoA sensor, but simply for software to monitor both sensors, compare the values, and if they disagree beyond a given threshold, to light an "AoA Disagree" lamp and disable MCAS for the remainder of the flight.

MCAS will also only be allowed to activate one time per "High AoA" event. The AoA sensors must return to a normal range before MCAS is allowed to activate again.

Finally, they must limit the maximum MCAS command authority within a set range so that manual control can always maintain altitude, whereas previously MCAS would command horizontal stabilizer adjustments without any regard to the current position.

It seems that in an AOA DISAGREE situation, the flight is still permitted to take off.

In my non-expert summary, they seem to be doing the absolute minimum amount of work possible to "address" the problem, and dodge completely the fundamental contradiction of why one would implement an unreliable-by-design MCAS system, hobble the control authority of that system, and further, permit flight when that system is known to be disabled.

In light of these admissions, I cannot comprehend why the MCAS system exists at all, and how the added complexity (variance in airframe operation) is worth any possible benefit.

In Elon-speak, the best part is no part. The best system is no system. So if you admit you can fly without it, why leave something in which is unpredictable, unreliable, and already proven to be deadly?

EDIT: The total cost of compliance for applying these changes to 73 existing airframes is estimated to be ~$1 million, 70% of which is the wiring harness change. <s>The software change itself seems to amount to about 20 lines of code, so let's call it $20 million.</s> Sorry, this seems like an absolute joke, and right now I'm pretty angry that this is what they came up with after "60,000 hours of review".