Giant Leaps … off a cliff: why U.S. space policy is all wrong

However, the spaceplane, loaded down with heavy wings and draggy wheels, never carried as much payload as ballistic missiles; even less could it achieve the massive thrust needed to reach orbital velocity. For instance, the X-20 Dyna-Soar--say that zippy name aloud--was canceled in 1963 precisely because the Gemini rocket provided the same bang for a third of the total launch mass.

The shuttle was conceived as a halfway house between the spaceplane's reusability and the expendable missile's capabilities. Alas, the halfway house was built to the wrong specifications, and we got a mindblowingly expensive, dangerous quasi-rocket that looks like a plane but that can be used just once before its "reusable" boosters have to be completely rebuilt. The result? A manned-spaceflight program flying in circles from the '70s onward, and a deformed hybrid that provides a lot of jobs in important voting districts, all while robots and probes have begun performing the kind of miracles astronauts were once expected to do.

NASA, clearly, has lost itself in space. But you don't have to be a rocket scientist to understand rocket science: It's all about reducing the cost per pound of payload taken into orbit. If this cost could be brought down to an affordable minimum, the entire calculus would change and every aspect of spaceflight--including missions to Mars, lunar bases, unmanned probes to Jupiter, space stations, etc.--would suddenly become not just feasible, but possibly commonplace.

There are two keys to lowering this cost. First, spacecraft, fuel tanks, and their boosters must be truly reusable, and reused scores of times. While the initial expenditure may be larger than that required to buy disposable versions, the former will be amortized over time, thereby lowering the cost per flight. Which partly explains why it's just $149 for you to fly cross-country on a reusable Boeing 777--a complex piece of machinery with 3 million parts--where the cost basis is roughly a penny per pound, compared with $10,000 per pound for the shuttle. In other words, if we want to conquer alien worlds and build orbital colonies, we're going to need to build a starship with airplane economics. Even the United Federation of Planets couldn't afford to maintain a fleet of ships on the NASA model.

The other key is competition--and I don't mean the Ansari X Prize recently won by Bert Rutan's SpaceShipOne, which, its impressive engineering notwithstanding, cannot hit anywhere near orbital velocity. No, the real competition must focus on drastically cutting launch costs to Low Earth Orbit. One contender is SpaceX, a company founded by Elon Musk, co-founder of the Internet-payment service PayPal. SpaceX has developed the Falcon rocket: a two-stager that is jettisoned after the initial blast stage, parachutes into the water, is recovered by tugs, and is then reused. To keep costs down, SpaceX's "Mission Control" is an 18-wheeler, and the staff consists of just 30-odd people. The potential savings are enormous: A satellite launch on the small Falcon I is estimated to cost $4,000 per pound compared with $14,000 on the comparable Pegasus, an air-launched booster made by Orbital Sciences. Longterm, as Musk's rockets get bigger, they could be used for manned flights to the moon and Mars. SpaceX, which is set for at least four launches this year, may eventually fail--commercial space activity is still in the doldrums--but its importance lies in demonstrating that non-NASA/ Boeing/Lockheed space exploitation and exploration need not be expensive or difficult.