><i> The price for all of that capability is a huge and heavy spacecraft that is simply too expensive to operate for the benefits returned.</i><p>I'd tend to differ. The problem with the Shuttle isn't its design....okay, the <b>main</b> problem isn't its design, it was in the way the program was structured. Simply, the fleet is too small. There should have been seven operational orbiters built - 6 based at KSC and one based at VAFB. Six orbiters is the largest number that could have been accomodated in the current facilities (3 OPF bays, 3 high VAB bays), and is a large enough fleet to give an adequately high flight rate. At 3 flights per-year per-vehicle we would have had a launch every three weeks, more than enough to satisfy <b>global</b> launch requirements. I believe that even with the current Shuttle design, it's not unreasonable to expect 3 launches a year.<p>One modification that would have <b>significantly</b> increased the flight rate would have been a modular payload bay. The payload bay module (PBM) would have simple, standardised interfaces to the Orbiter, and could be removed and re-installed in a short period of time. The interfaces to the payload would be made to the PBM which could be re-configured and/or customized to suit each payload. This would have dramatically increased the availability of the Orbiters, as currently a significant amount of work (and time) is spent configuring and deconfiguring the payload bay for each mission. The beauty of this system lies in the fact that an arbitrary number of PBMs could be made and they would even allow payload integration to occur at the manufacturer's facilities, with the PBMs being shipped to KSC for installation as little as a few days before rollout (or even at the pad). It would also have simplified the job of flight manifesting as payloads could easily have been switched between Orbiters.<p>Together, these two changes could have made the Shuttle economical - remember, it's the large fixed cost that</p></p></p>