As I write this, contestants are preparing to launch (quite literally) their efforts to win the Ansari X-Prize. According to a report in the Houston Chronicle, Burt Rutan's SpaceShipOne (which I think could fairly be called the favorite in this competition) is scheduled to make its first launch of the two required the day this column appears, with the second launch (required by the contest to demonstrate fast turnaround capabilities) planned for next week.

That might slip, rocket launches being prone to such things. (The Canadian DaVinci team has just announced a delay, for example.) Or everything may go off perfectly.

But I think it's worth stressing just how far we've come, and how much the X-Prize has accomplished already.

It's probably no coincidence that we made rapid progress in space back when the attitude was "win at all costs," and that our progress slowed down drastically once the attitude became "no mistakes allowed."

You don't want to risk people's lives, or expensive equipment, foolishly, of course. But we often learn the most from trying projects that stretch our knowledge and capabilities, not from playing it safe. And, at any rate, the urge to play it safe usually doesn't come from concerns about protecting people or equipment, but from concerns about protecting bureaucrats from criticism.

That's certainly one of the things that has been wrong with NASA's manned space program. The goal was to avoid things that could be criticized as failures, yet it has produced, well, a lot of failure, and not a lot of things that can be praised as successes. (The early days of the space program produced a lot of failure, too, but it was failure that led to later successes.)

The prize structure helps offset that. First, the people involved aren't bureaucrats. Political criticism isn't such an issue for them. Second, the very nature of a prize competition presupposes that some people -- all but one, in fact -- will "fail." Instead of asking "why didn't you deliver what you promised," we tend to respond "better luck next time." And why shouldn't we? It's not our money at risk, but theirs.

I expect that SpaceShipOne will be successful, whether or not it's on schedule. And I wouldn't be surprised if several of the X-Prize competitors manage to fly successfully, even though only one can win the prize. I also expect that even those who don't win will demonstrate technologies and approaches that someone else is likely to find useful.

It's hard to structure government programs so that they produce this kind of an effect, and even harder to maintain them in the face of a political and media environment when learning from failure is seen as indistinguishable from failure alone. But in all sorts of areas -- from space and jet aviation in the 1950s and 1960s, to computers in the 1970s and 1980s, to the X-Prize today -- it seems that we make faster progress when we have lots of parallel efforts, with freedom to experiment, and to fail. Sometimes we got that sort of thing within a government program; other times it happened outside. But it seems to be an important formula for success. That's something we might want to keep in mind.