It goes between two places with at least some sort of atmosphere by default (or one with and on without for the Lunar variant) so while indeed dragging aerosurfaces with your in the vacuum of space might seem wasteful on the first look, you can save huge amounts of delta-v once you arrive at your destination due to atmospheric braking.

A purely built for space vacuum only craft could be built to be much lighter but would also need to loose all that excess interplanetary speed with engine power alone, potentially negating the advantages quite a bit.

There are, of course, various hybrid approaches: - the space only craft still could have some limited aerodynamics to do earocapture to orbit at the destination or at least do aerobraking from a "cheap" eccentric capture orbit - you could have a space only "hotel" ship on a cycler orbit and get passengers and cargo using Starship like "taxi" ships from a planet - a big complex space-only ship could have fancy high tech low thrust & high ISP engines that would not realistically fit on craft that needs to be fairly aerodynamic

Also as for in space construction - indeed, it will enable really massive and efficient space structures, but we are not there yet space infrastructure wise. And Starships so far have shown how easily they can be made in their current form using very basic tech during their assembly, even compared to tech used for existing decades old expandable rockets! :)

So no doubt eventually big efficient high tech in space cruisers will carry the most interplanetary traffic, it seems like Starship is a surprisingly good fit for the time being & for putting in place the building stones of that future. :)

Heck, you don't need orbital assembly for that. SpaceX could put in orbit a single- or two-engine starship with no landing legs and no heat shielding for transfers. Eventually they could decommission one such truck to make an orbital station.

> Anyway, $100-200/kg (or less!) and huge volume/high cadence is so radically disruptive that it's kind of hard to predict how space will look a few decades down the line. It's going to be a wild thing to watch, with lots of opportunity.

There might be so many satellites in orbit as to permanently disrupt astronomy on Earth, necessitating a fleet of replacements for the James Webb telescope.

Sorry not familiar with these plans - does this mean that we assemble a bigger rocket (by using earth based rockets) at ISS level orbits and then relaunch to another planet?

  - Reusable launch system
  - Intercontinental transport
  - Space tourism
  - Earth–lunar transport
  - Mars colonization
  - Multiplanetary transport

Because of that, I would say starship still has to prove it is commercially viable.

Space tourism, in particular, highly depends on how safe one can make the trip. A 1:1000 risk of death likely would make many would-be passengers think twice before booking a trip.

It also would depend on making the trip nice. The first few tourists might go for bragging rights, but after that, one would need attractions in space. Building those there will increase the cost of a trip.