> because we live in the software world where if you see something that is done then cloning it is generally just a matter of time and effort. But there's certainly some 'secret sauce' at SpaceX that does not seem easy to replicate

In software, thanks mostly to the open-source movement, we all have access to the same materials and manufacturing processes.

In manufacturing, there are huge differences in materials and processes that we need to account for. On the flip side, this can allow one to leapfrog the other rather than replicating them - it’s likely the next reusable Falcon 9-class vehicle will use a methalox engine that’ll be easier to maintain compared to Flacon 9’s kerolox ones. China has demonstrated a novel approach to vehicle recovery as well (in addition to landing legs). Future competitors will have access to materials current incumbents didn’t have and can’t easily adopt with their current supply chain.

People also forget SpaceX subverted a lot of the expectations for reusable rockets - they didn’t use parachutes, but opted to carry extra propellant for landing; they didn’t eject the engines, opting instead to reuse the whole stage; and they built landing barges because their rockets would launch towards the ocean, saving fuel for a boost back to the launch site. They also tested capsule landing using the launch abort rockets, but opted for parachutes and splashdowns instead. Still they keep all the RCS motors in the capsule, making their expendable service module very inexpensive in comparison to Boeing’s.

All these decisions point to proven mission architectures so that followers don’t need to explore architectures that were explored and dropped.