You know, for an 'inherently flawed and unsafe design' it did pretty well for almost 30 years, outliving it's expected life by, what, 15?
I would be inclined to think that the reason it "did pretty well" is more to do with beating the odds than good design or good management. Read what Richard Feynman had to say about his role on the Challenger investigation board (Rogers Commission) in "What Do You Care What Other People Think?". It's fascinating. The people on the ground who had the most to do with the Shuttle put the odds of a catastrophic mission failure at much shorter odds (1 in 100 ISTR) than managers (something like 1 in 100,000 - sorry for not being more precise I don't have the book to hand). These were the same managers who were much less obsessed with the safety of the shuttle and crew than they should have been and pushed for launching when they shouldn't have done. I suspect managers with similar figures for failure in their heads were the ones to ignore concerns of more junior staff when the hole was first detected.
At the point where the shuttle broke up it was obviously a non survivable event, but I'm of the opinion it didn't have to be if appropriate steps had been taken when there was first a problem detected. I also feel anything they can learn now from Columbia to help design a better vehicle that ups the odds of surviving a catastrophic failure in future is a good thing.
To go back to my original point, I do think it is extremely misguided to say that just because a thing hasn't happened before means it is safe or well designed - it may just mean that so far it's beaten the odds, and I don't think that should be overlooked by NASA when they come to finalise future designs, or plan future missions.
"In the face of entropy and nothingness, you kind of have to pretend it's not there if you want to keep writing good code." -- Karl Lehenbauer