As a counterpoint: NASA has an _extremely_ careful process and their success rate is high. It's expensive, though, and you have to develop an engineering proces

As a counterpoint: NASA has an _extremely_ careful process and their success rate is high. It's expensive, though, and you have to develop an engineering process that supports it. When you're sending a probe to Mars, the trial runs are very expensive. They spend a lot of time making sure that the mission will be successful. I'm skeptical that anywhere near the same level of care is going into the private missions, but I the reason is cost, not that they can't bootstrap themselves up with Earth-environment experimentation. Yes, I'm aware at least one of the Mars probe missions totally failed. I don't think that detracts from the overall point. They weren't screwing up a bunch of missions before they got one right.