I never considered NASA responsible for Challenger. IIRC, the White House pressured NASA to make the launch for publicity reasons, otherwise they would have delayed for temperature reasons. And that brings up the "real" testing issue in this case..... how well did Morton-Thiokol test those O-rings for low temperature safety, and was it documented in a chart showing the failure-vs-temperature probability. From what I recall, they just relied on some engineer issuing a politically pressured guess. (Of course Morton-Thiokol might just have been a scapegoat, being small enough to blame without too much adverse corporate disrepute).

Everyone gets alot of "patriotic confusion" over the space race, and the blame game gets very distorted, since they feel failures are a reflection of God's love/hate for our country. As a matter of fact, I had a woman blame me personally for the Challenger disaster, because the night before at dinner, I was discussing whether it was even ethical to blast rockets into space. She claimed I psychically caused it by bad thoughts. Unbelievable!

And then there is the Columbia's reentry burnup.... that one that burned up over Texas. Are you telling me that they didn't have a contingency plan? They just crossed their fingers and prayed it would make it.... well it didn't. They devalued those astronaut's lifes. After that, they made a contingency plan to assuage public uproar.

Now to inject my own psycho-babble blame for those 2 disasters......they both occurred under Republican administrations....Reagan and Bush Jr., who made it a point to curtail non-military spending.

Anyways, I don't want to bash NASA in this thread about software testing, because those were hardware and political failures.

The biggest most unbelievable software blunder in space, related to poor testing, was that infamous "metric-vs-english" value for the gravitational constant which destroyed that Mar's probe a few years ago. Now that was negligence! Are you telling me, that their simulations didn't detect that gross error? A complete failure of proper software testing.