Go PTS! :) And I feel for you chfjfuchdjd, I started writing a response and noticed that I'm writing basically same thing you did, but anyway: The most import

Go PTS! :) And I feel for you chfjfuchdjd, I started writing a response and noticed that I'm writing basically same thing you did, but anyway: The most important thing is industrialization, using resources already in space (moon and NEOs), independent as possible from things lifted from earth. It sounds a far off goal to - for example - plaster moon with solar panels, but there's no physical reason why it couldn't be done, mostly automatically, with very small input of materials from earth. "Want to go live to moon? There's nothing there but vacuum, wasteland and radiation, and you need to take everything with you, at costs of 1.2M$/kg. Nice view of stars.", doesn't make much sense, does it. Try this: "Want to go live to the moon? You'll have 40B$/day of free electricity [1] and absolutely massive, if still somewhat one-sided manufacturing capability to use. Build anything you like (including a garden, just bring the soil and seeds with you). Yeah, it costs 1.2M$/kg to get there, but you'll have a big underground shelter and free oxygen for kickers. Nice view of night sky, btw.". There, now it's suddenly making a lot more sense! The problem is not that there would be some single insurmountable technical challenge stopping us, there just there isn't an agency or program whose purpose would be to attack the huge number of those technical challenges needed to overcome. Like, how do you build magnets for electrical motors without plastic or carbon for wire insulation? Maybe you could find formulation of glass to coat wire with. Maybe you could deposit alternating layers of conductor and glass directly to core to make magnets. Maybe you could 3D-print the wires directly inside the insulator. Or something completely different. It could take better part of decade to find out and scale, and perfect the process down to a T. And the end result would be something onerous and expensive to manufacture in earth (where's my vacuum for that vacuum deposition step?), and probably somewhat less efficient than we can muster with earthy materials. It doesn't matter, it could still work brilliantly on moon. But it's quite unlikely that such a technology would be an offshoot of some other R&D project here on earth, or an offshoot of some space science project. None care about the poor plastic-free magnet wire. There are some really interesting projects in NASA, for developing ISRU, and say nuclear power (could be handy for bootstrapping moon base), but these never seem to be the ones that literally fly. At least for an outside punter (correct me if I'm wrong) they seem to be just a side dish, not part of concerted effort/vision to develop technological base for manufacturing. What flies seem to be science probes (good!) that can also take nice pictures (a bonus), and ISS. Which brings me to humans in space. It seems a huge waste to haul people to LEO, and then support them there for decades for great expense, and hope that will somehow help the march of civilization to space. You need a reason first for those people to be there. You need to extract energy and material from the environment, and create useful and interesting things with them, because that's what life does, both pooping humans, and clanking machines (together what I'll call civilization). What proportion of space civilization should be humans, and what proportion machines, should be strictly rational affair, you lift as many people as is optimal for economic growth at that moment. I would not worry, as long as machines need humans at all, it's probably useful at some point to have some people nearby to troubleshoot processes gone awry, do R&D, or just hang around at spa, even if life support is expensive. If humans actually become redundant, we will probably become redundant some point on earth too, anyway. tl;dr: Self replicating solar power industry on moon - build it, and they will come. [1] Roughly the amount of electricity that 200km wide strip of solar panels around moon equator would produce at 0.04$/kWh and paltry 4.3% efficiency, possible with even crudest of processes.