The Great Thermite Debate...

[OnTheOtherHand]

116. perhaps it depends, again, on how one construes the problem

Of course I agree that NIST's models can't be proven to give correct results even if the initial conditions were known with certainty, which is impossible.

I don't know that this is on point, but prior to the decision to launch Challenger, Thiokol engineers recommended against the launch in part because (1) they believed that the O-rings were affected by the cold and (2) the temperatures expected that day were outside the range of experience and testing. The managers weren't relying on engineering models that falsely indicated safety; they weren't relying on the engineers.

At any rate, the O-rings are the focus -- fairly uncontroversial, as far as I know -- of what might be called the probable destruction sequence in the Challenger tragedy. But, as far as I know, no one has demonstrated that the tragedy was inevitable given the temperatures that morning. Sabotage is a logical possibility. (The FBI even conducted an investigation -- which presumably I would denominate a "whitewash" if I thought the shuttle had been sabotaged.) Still, I think the analysis of the Challenger tragedy is pretty damn good.

I think the NIST analyses are pretty damn good, too. In my view, the analogy isn't half bad, either, although the towers are more complicated than the shuttle was.