> The sudden rush to judgment on this case is unjustified by the evidence.
It might not be justified by a theoretical formal proof. It irks me that a lot of the HN population seems to always want to reason using formal proofs like if it was a silver bullet.
In this situation, bayesian inference is appropriate, and tells us that grounding planes is highly justified. This succession of two relatively similar events in a new plane is enough to raise the probability of "this aircraft model has a problem" a lot, and definitely enough to justify grounding them.
Also, the risk / benefit ratio is not symmetric, what is better:
- Ground them for nothing and lose money
- Not ground them to save money, and see a third crash happen, killing 150 again
If you take into account those two effects, it's clear that grounding the planes is a sane option.
Finally ask yourself: would you board this plane today if you had a ticket? What would you think when there is turbulence after takeoff and the plane suddenly pitches down for a second? What color would your pants be?
And there is a perfectly plausible scenario in Lions air and preliminary data on the new crash point in the same direction (erratic altitude before the crash). It’s not like there is nothing, there are technical details that do match.
- Not ground them to save money, no crash happens before the cause is found and a fix is deployed
- Not ground them to save money, and see a third crash happen, killing 150 again
And that also doesn't consider secondary effects, like the loss of confidence from the public on that aircraft type, or the potential confusion from the public between 737-MAX8 and 737-800.
I'd say it's the opposite: the first scenario (ground the aircraft type) would suffer less from those secondary effects. Once all the airlines on your region are no longer flying that aircraft type, people no longer worry about whether they're flying on a MAX8 or a non-MAX8; they know they're flying on the "safe ones", so they have less reason to memorize the name, to check the aircraft type on their tickets, to tell their loved ones "avoid that airplane", etc.
The third scenario is the worst case, if there's a third accident with the same cause (or at least perceived to have the same cause), the reputation of the 737-MAX8 will get tainted, and that taint could even spill to other 737 variants (less knowledgeable people don't know or care about the many variants of the 737, to them "737" is the airplane type).
It might not be justified by a theoretical formal proof. It irks me that a lot of the HN population seems to always want to reason using formal proofs like if it was a silver bullet.
In this situation, bayesian inference is appropriate, and tells us that grounding planes is highly justified. This succession of two relatively similar events in a new plane is enough to raise the probability of "this aircraft model has a problem" a lot, and definitely enough to justify grounding them.
Also, the risk / benefit ratio is not symmetric, what is better:
- Ground them for nothing and lose money
- Not ground them to save money, and see a third crash happen, killing 150 again
If you take into account those two effects, it's clear that grounding the planes is a sane option.
Finally ask yourself: would you board this plane today if you had a ticket? What would you think when there is turbulence after takeoff and the plane suddenly pitches down for a second? What color would your pants be?