The system should have been evolving already by the time of the Challenger failure; by that time it was already clear that the system could not meet its goal of

The system should have been evolving already by the time of the Challenger failure; by that time it was already clear that the system could not meet its goal of quick and/or cheap re-usability. But by that time NASA already had a full "fleet" and a space station to assemble, and had no taste (or money) for developing a new system--or even variations on the existing STS system. There were some design improvements over the years, but the system remained basically as first flown in 1981 until it was shut down, with no successor in mind. The STS was supposed to be a first generation system, to fly for 10-15 years before being replaced. By now we should be on a third-gen STS, having learned the lessons of the first few. Had that happened, they might have a pretty good system by now. But NASA got locked into a flawed 1st gen system by over-buying the first run out of over-confidence, by politics, inertia, etc. (Part of the problem was that the orbiters were designed for 100 flights, so NASA wanted to get something like that out of them. But the flight rate ended up so far below expectations that the darn things just wouldn't wear out.) I guess it's just as well; NASA doesn't really have the structure for cost-effectiveness. STS v3 might have been good enough to stifle the emergence of something like SpaceX, which has a much better shot, long-term, of really making spaceflight work.