perfidy

This article about John Carmack (developer of Doom and Castle Wolfenstein) and his efforts to get into space hits at one of the key problems we've had in space development over the last forty years:

Testing is key for Carmack, who doesn't want to work for months only to find out a rocket doesn't work. He believes the more testing done, the faster the crew can work out any kinks.

"Some people have commented that I am trying very hard to make aerospace like software, and that's the truth," he says. "If we looked at what we do in software, if we could only compile and test our program once a year, we'd never get anything done. But that's the mode of aerospace."

Only one space program since the end of Apollo has used a rapid development process, and that was the DCX. Typical NASA programs involve millions of dollars and years of testing before there is even an attempt to cut tin and actually construct a prototype. Aerospace engineering is not so cut and dried that we can make a perfect design on the computer, build it, and expect that it will fly.

Cost overruns, failed expectations and cancelled programs are the result of this design centric philosophy. The key to success is to build early, test early. Lessons are learned quicker, and applied easier through a regime of rapid prototyping and testing. Just like in software development. In a matter of months, the DCX team went from a standing start to a 1/3 scale flying prototype. And spent a fraction of the money that was ultimately spent on the X-33 which replaced it, and which never once flew.

The growing provate space industry is largely funded, if not actually run by successful software magnates. They seem to be applying the lessons they learned in developing other technologies to the problems of space. They are expending effort where it does the most good - gaining experience in building spacecraft. Even if the first, second, third attempts fail, at the end they will have a wealth of experience that NASA has lost in the days since Apollo. NASA has not designed a new working vehicle in almost thirty years. They have forgotten how it was done in the golden age, for what was the sequence of Mercury, Gemini and Apollo but a series of prototypes and testbeds to gain the practical engineering skills to reach the moon? Test early, test often.

What would have happened if NASA had spent the period between the launch of Yuri Gagarin and Apollo 8 designing, redesigning - on paper - the perfect launch vehicle? A giant explosion, most likely. And that is why I am certain that of the twenty teams now competing for the X-Prize, at least several will have successful flights by the end of next year.