Sending robots to establish infrastructure is one way to do it (and I think we should do as much of that as we can), but current robotics technology isn't very

Sending robots to establish infrastructure is one way to do it (and I think we should do as much of that as we can), but current robotics technology isn't very good when it comes to operating unattended in a harsh environment. If the price of shipping goods to Mars drops faster than the capabilities of robots to act independently, repair themselves, and construct new parts from raw materials improves, then sending lots of people to manage the machines is going to be the easier solution. I would love to see more work on problems like "what's the smallest device I can send to another planet that's capable of making some kind of bricks and stacking them into vaults/domes that can be covered in dirt and then sealed from the inside to form a habitat?" or "can we build a robot that can assemble a personal computer from a stack of components in their retail packaging?" or even "what does it take for a general-purpose robot to be able to tie shoelaces?"