And we better use robots for solving one problem a time, at least in early experimentation. For example, make robots measure absorbed radiation, to test those antiradiation shells. Or make them build a sustainable water supply and farm in the target planet, to prove it can be done.

That could be cheaper, and less risky in human lives, than trying to juggle all balls at once.

That's the point. They won't become relevant until actual misions make them relevant.

Until we know what's out there we can't know when we'll want to send people far away.

Add to this that we won't want to send people far away because we haven't (still) the right technology and you have a nasty vicious circle.

This remembers me those other news on HN: the duck tape programmer. We need a duck tape exploration program, that's it :-)

There's no vicious circle. Once we find something worth sending humans to, we have a motivation to develop the required tech.

Also, the space station get resupply from Earth. Doesn't solve GP's "autonomous and sustainable life-support".

The space station isn't a complete waste, but it just doesn't tackle those problems. (Disclaimer -- I'm very much not a rocket engineer.)