Today, I saw a piece in The Space Review about what makes spacecraft launches complex and difficult. It occurred to me that this was a rather odd essay, coming as it does on the heels of the successful, high-profile flight test of a rocket that promises to seriously shake up the launch game. The essay is a full-throated defense of the Old Launch paradigm; the idea that the people who have been approaching space the same way for decades are the best at it by virtue of their heritage. If this essay had come out a few years ago, when SpaceX was experiencing strings of launch failures, it might be relevant; but now it is a perfect illustration of what’s wrong with space industry thinking.
Building and launching spacecraft is hard, no doubt about it. Satellites and rockets are complex systems. A lot of things have to happen very quickly, and some things have to happen in regimes where we don’t fully understand all the physics. The success rate for space missions is not 100%. (These days, though, it’s pretty darned close.) However, the inherent difficulty and complexity of space exploration and exploitation is a poor reason to shy away from innovation.
The Space Review essay opens with the following paragraph:
One of the most challenging aspects of launching payloads into space is that you not only get only one attempt for a particular set of hardware, but usually that one attempt is the first time that particular set of hardware experiences the actual flight environment. It may even be the only time that overall hardware configuration ever flies. Every flight is a test flight, like it or not. For that reason it is very, very important that the hardware gets built every single time in exactly in the same manner of other examples that were found to work properly. This is not easy; in fact, it may be hardest single requirement in the space launch business.
I’ve added some emphasis to a statement with which I cannot disagree more. The author says that the most important requirement for space hardware to meet is that it should be exactly the same as other space hardware that has already flown. I think that what he should say instead is that it’s important to be sure that your hardware will work. Whether you prove that by simulation, analysis, experiment, back-of-the-envelope calculation, derivation, or by comparison to flight heritage is immaterial to me!
I think that this notion of valuing flight heritage above all other considerations is detrimental to the space industry, for a couple of reasons. First, it stifles innovation. If, over the past sixty years, we really hadn’t sent anything into space that hadn’t already been in space, we wouldn’t have any satellites at all. Or, if I’m going to give humanity the benefit of the doubt, we might have a couple satellites but they would all look like this. Space is a challenging but rewarding environment. Purely in economic terms, it’s worth it to stick our necks out a little and accept a couple failed launches in return for all the infrastructure that we have been able to deploy in space, from weather satellites to Earth imagery to military support. The more capabilities we want from our spacecraft, though, the more we need to innovate. Sometimes – heck, often – that means we have to build a vehicle that looks different from the things that have gone before.
Second, I don’t like the idea of flight heritage because it involves an implicit logical fallacy. Spacecraft engineers sometimes confuse a solution that worked in the past with the best solution to a problem. Sometimes, spacecraft launch with really state-of-the-art devices and programming. But, other times, they launch with only good hardware and software. Every now and then, they even launch with something on board that’s actually sub-par – and sometimes, that causes a problem. An engineer might think that if a design has heritage, it’s certain to work. But no such guarantees for success actually exist. Spacecraft are not like mass-market consumer goods: we can’t test thousands of samples and get a good statistical sense of whether we have the best design or not. We have to deal with small-number statistics for successful missions.
It’s important to look at spaceflight heritage with a critical eye: What worked? What didn’t? And why? Do we have the best solutions? Can we make them better? If so, what would it take? These are questions that drive innovation. They are more likely to come up at a New Space company – which has to innovate in order to survive – than an established Old Space company. I have great respect for the engineers that have been able to launch whole series of operational spacecraft. But I am wary of an approach that views prior success as a standard of perfection.