aren't better than teleoperating ROV's for efficiency. The beauty of ROV's is that if you lose one all hell doesn't have to break loose and the whole space program shut down for months or years. The vast majority of time spent on ISS has been the building and replacement of parts to keep humans alive, not science. The same thing will happen on the moon.

If you don't have the humans (at first), you don't have all that expense of taking equipment and supplies along to keep the humans going. The time lag for teleoperation to the moon means you can have quick responses to virtually any situation you are teleoperating. Rather than single super expensive rovers like we now have on Mars with a much longer lag time, you can have lots of cheaper and more heavily virtualized machines that could have single real time operators.

Building CAPS or anthing else on the moon, would go vastly faster if you don't have to worry about keeping humans alive at the same time. Raul's point that there can other (political prestige) reasons for putting manned presence out there is true, but we've already put humans on the moon. A heavier point would be to get down and do some real cost effective exploration right now. It is a bootstrapping process and the longer we dilly dally and don't let real engineers design the most cost effective straight line way to do the bootstrapping the longer the whole process is going to take.

Put up a bunch of high bandwidth communication satelites and then lets start droping rovers and other teleloperated machines on the moon and get to the task of doing some real exploring and building.

I'm a believer in space colonization, but I believe the fastest path is to use the tools we have developed in the most cost effective manner leading the way. That means teleoperating to build the infrastructures that man will eventually make use of to live off planet.