Think Like a Rocket Scientist

When Dan Goldin became NASA administrator in 1992, NASA was facing a crisis: cut costs or go out of existence. It was under this extreme pressure that Goldin hatched his slogan, �Faster, Better, Cheaper,� and sold it to Congress. Goldin�s idea was born out of self-preservation; fi erce political winds were blowing and only the strong would survive. Goldin came up with his threepronged optimal solution: NASA would be �the best� by all three measures. But as Cliff Stoll said in his enlightening book High-Tech Heretic, you can�t have all three. (Stoll was the astronomer who broke up an international spy ring by noticing an excess charge of seventy-fi ve cents on his Harvard main frame computer; he tells his spy-thriller, scientist�detective story in his gripping memoir, Cuckoo�s Egg.) Let�s say you�re hungry and you want to eat dinner. As Stoll points out, you have three choices: you can have it better and faster by going to an expensive restaurant, you can have it better and cheaper by taking the time to cook it at home, or you can have it faster and cheaper by settling for a burger at McDonald�s. You have to give up one of the three: economy, speed, or quality. But Dan Goldin was no dummy�he understood that he could pressure NASA to really look at this optimization trade. There is a branch of optimization theory that considers multiple goals that are contradictory. We face such problems all the time in real life. In the Galileo project, I had to design a sequence of eleven orbits about the planet Jupiter that would provide the principal investigators with observational opportunities to �maximize science return.� Unfortunately, there were three major objectives: study the moons of Jupiter, study the atmosphere of Jupiter, and study the magnetic fi eld of Jupiter. And guess what? These goals were at odds with each other. As a mission designer, I found myself in a role similar to a AAA travel agent trying to satisfy a family of three�each of whom wants to visit a different sight. What makes their problem diffi cult is that they�re on a tight schedule and budget and have only one car to drive. The travel agent can only draw one Trip-tik�only one path that goes from A to B to C. Similarly, maximizing Jovian atmospheric observations would reduce opportunities to photograph the moons or to study the magnetosphere of Jupiter. To maximize each scientifi c investigation, we really needed three separate spacecraft, but we only had one spacecraft. And planetary scientists (the principal investigators) were a lot like backseat drivers on a family vacation�they weren�t above throwing temper tantrums to get their way. In the end, mission designers (in their cool, logical, Spock-like way) had to compromise and so did the scientists (though there was some screaming and shouting and shedding of tears). A trajectory (an orbital tour) was designed that would get �adequate� coverage for all major science objectives. The idea was to plan a feasible mission that would not give inadequate return to any major area. So when Dan Goldin cooked up his shibboleth of �Faster, Better, Cheaper,� he knew what he was doing. His statement is illogical, taken literally, but behind those three words there is a compromise and that was the only way NASA could survive the government budget cuts at that time. In real life, we deal with these seeming paradoxes frequently. It is helpful to get specifi c about what your goals are (such as faster, better, cheaper), but you also need the wisdom to decide what combination of these goals