I just saw Superman Returns last week, and the opening sequence (featuring Sir Richard Branson in a fictional Virgin Galactic space shuttle replacement) got me on a bit of a space kick.

To NASA’s credit, they’re a dedicated and persistent organization. Yesterday, however, cracks were found in foam insulation on Discovery’s fuel tank. Of the five Shuttle Orbiters built, two have been lost to accidents – hardly an acceptable ratio. More meaningfully, there have been a total of 114 Shuttle flights, with 2 accidents that killed 14 crewmembers. The Russian Soyuz program has had 95 manned and over 20 unmanned launches, with three launch or landing disasters that killed a total of four crew. As far as I can tell, the last Soyuz fatality took place in 1971.

The Shuttle’s fatality rate is awful – and the events much more recent – when compared to the only other long term launch system in operation. Why are we taking so long to replace it? Projects that were nearing completion, like the Lockheed VentureStar, have been cancelled, leaving NASA working on a cobbled-together program called Ares that will eventually succeed the Shuttle. It will re-use existing boosters and engines in a new, single-shot configuration.

The work being done, both within and outside of NASA, is a good step, but we should have been somewhere else already. The Shuttles were designed for 100 launches or 10 operational years – yet they are now 20+ years into their operational lives and have only achieved around 20 or 30 launches each. Operational costs were vastly underestimated, as were turnaround times.

Failures like NASA’s are why new ideas and competitions like the X-Prize are so important. Burt Rutan’s SpaceShipOne, and his recent collaboration with Branson’s Virgin Galactic, is the right step. We need brilliant visionaries, not engineers in offices. The only clever launch systems recently are Burt Rutan’s and the Boeing-Energia SeaLaunch collaboration, which effectively created a floating equatorial launch platform mid-ocean. Is spaceflight something in which private enterprise should have an increasing role? It seems that might be the only way we’ll see practical advancement, but then there’s the disadvantage of access to space being commercially-owned.