Soyuz as a rocket just had a failure in 2018. Russian Space program had massive issues with reliability. Soyuz as a capsule has a fine safety record but most ca

Soyuz as a rocket just had a failure in 2018. Russian Space program had massive issues with reliability. Soyuz as a capsule has a fine safety record but most capsules have a good record. Lets remember that somebody in the Soyuz manufacturing drilled a hole into it and hid the fact, this was only noticed when ISS had a pressure lost. > Of course not. But a pressure to drive the per-launch cost downwards will indeed lead to shortcuts being taken. We won't actually know how safe the vehicle is until it's launched successfully a great number of times. SpaceX Falcon 9 even reused is fully certified for humans, for the highest level of NASA Science payload and for all military missions. Those are three very difficult to gain certifications. Falcon 9 even reusable has already launched more then most rockets in history and is overtaking even the Ariane 5 shortly. SpaceX knows more about these things then anybody in the world. Re-usability and the certification for it have been worked out. > I'm not so sure that's how this works. Starship was never considered as a lunar-bound vehicle Wrong. Starship was always designed to be able to land on Moon, Mars and Earth. > and whatever lander module they're going to build will be a brand new design, designed by folks who've never built a lander before The engine and control structure are the the same no matter where you are. Its not a brand new design, its Starship slightly updated. > There's no way launching satellites into LEO or even visiting the ISS (also in LEO) builds any sort of experience for landing on the moon. No but landing on earth does. > NASA certainly is, by having a single design. Congress is cutting corners and NASA does the best it can.