A paragraph or so from Czysz/Bruno’s “Future Spacecraft Propulsion Systems,” where they discuss a little problem in aerospace: over the last generation or two, as our ability to collect and analyze data has increased, out actual physical progress has stalled or even slipped backwards.
I’ve been involved with the whole spectrum of aerospace… from two guys testing rocket components out of the back of a truck in a mostly-empty parking lot, to major aerospace contractors working on NASA manned vehicles. And I gotta tell ya… there was much more progress – certainly much more forward momentum – at the small end. Why? Because we *didn’t* have all the analysis hardware and software. If we wanted to know how hot something would get, we had to actually put it in the fire. In doing so, you often learned things that you would *never* learn by simulation.
At the end of a simulation run, all you’ve done is shift some electrons around, perhaps consumed a whole lot of processed tree slices, and increased the entropy of the universe. At the end of a test run, you’ve actually *built* something. If it works, you are *far* ahead of the simulator. If it didn’t… you are very likely *still* far ahead of the simulator, if you have the wit and grit to understand what went wrong and to correct it and try again.