Ok, I'm not that familiar with the plans with starship. I think the premise of the article is wrong anyway. I think no matter what the system is like, it stil

Ok, I'm not that familiar with the plans with starship. I think the premise of the article is wrong anyway. I think no matter what the system is like, it still makes sense to optimize for mass in a reasonable way. That means, not sending dead weight to high delta vee. Leave big tanks or engines behind. It's a bit like you don't transport your car with the airplane. It would be madness. Exponentials are unintuitive. The total mass of all fuel in all stages is directly proportional to the mass that needs to be sent to the highest delta vee. So for example (to take a few shortcuts) if your moon ascent stage is 1% heavier, everything has to be 1% heavier, including the first stage of the launcher. So 1 kg in the ascent stage might not look like much but it grows the first stage at something like 1000 kg. That's why I'm really skeptical of the architecture of landing atmospheric stages to the surface. They are just unnecessarily heavy. You have maybe quadrupled your total launch amount from the surface (haven't done an analysis). Early Apollo planning had direct landing because docking in orbit had not been demonstrated. Gemini validated that. Orbital docking works fine and Apollo could continue with muchh more efficient architectures. Actually, docking has gotten a lot better, we did have Apollo, yes, and we have robotic tugs routinely flying to ISS. Engines haven't made significant progress since the sixties, not enough to warrant really big architectural changes towards less mass efficiency. If anything, the improved docking and automation that has already been demonstrated would swing the architectural plans for moon or Mars to have even more docking, berthing, refueling and staging! Direct landing might only make sense with something like nuclear propulsion...