Well, the guest made mention that her initial rocket science comparison was probably going to draw angry letters from rocket scientists ... but the way she delivered the comparison still leads me to believe that she thinks it’s a good one. Well, here’s the “letter” :)
While I think the guest discussed some interesting material, and I think her work is very important, she does herself a disservice by making the comparison, which reveals a profound ignorance of what rocket science is (disclosure: I don’t work as a rocket scientist anymore, but I have Aero/Astro degrees from Caltech and Stanford, and am pretty qualified to speak on the subject).
In our language, rocket science and brain surgery are held up as the epitomes of complexity. I’m sure many people would like to think that what they do is more complex than those two subjects. So, there’s a natural pissing contest aspect to this. That said, rocket science actually is tremendously complex.
The guest’s description seems to imply that she thinks rocket science is the same thing as a simplified computer model of rocket science. A discipline where everything has an exact answer, where you only have to plug in the right variables, and the computer spits out a black and white answer. She described her science as being one where you could only get average answers, that still had an inescapable variation on either side of that average, because of the complexity and variability of human psychology. In fact, that’s exactly what happens in rocket science, too.
For example, one of the difficult problems in rocket science is picking materials that hold up to the extremes of temperature, pressure, radiation, speed, vibration, etc. When you use engineering materials, you don’t know when those materials will fail. You have estimates, based on alloys of a given assumed composition (aluminum isn’t just aluminum, and steel isn’t just steel, and nothing is pure). You try to predict failures with statistics that suggest a certain lifetime, but everything down to microscopic flaws in the material will influence whether you achieve the life cycle estimate, or have early failures.
Another example is that rockets don’t get teleported into space. To get there, and get back (if they come back), they have to go through the atmosphere, and that subjects them to weather. I hope everyone understands that there’s no equation that predicts weather with perfect certainty. Remember Challenger? Even in space, you have solar weather (e.g. cosmic radiation) that fluctuates, and will affect the operation of electronics. We’re still struggling to predict sunspot activity, as followers of recent climate science news will note.
Perhaps one layman’s impression of rocket science concerns orbital mechanics, which is one and only one part of rocket science. Within an environment with no atmosphere, some of the orbital mechanics equations can be quite precise. However, when the spacecraft is relaying data back to the ground, or even to its onboard computers, it doesn’t know where it is or how fast it’s going. It only knows what sensors tell it. Sensors do not know truth. They only estimate measurements, and there’s error in any physical sensor. Part of rocket science concerns how to interpret the sensor data, and make corrections to yield the best result. Anyway, I hope this brief explanation offers a look into some of the complexities of the field. Obviously, I could go on indefinitely about nuances of rocket science completely lost on general technical minds.
I know rocket science wasn’t the topic of the podcast, but I never like to see scientists undermine their message with faulty logic, or false claims. Statements like the one made by the guest only demonstrate ignorance, and distract some members of an audience from the intended content.