Overall, it's a very good article. I do support going back to the moon and beyond. In the 6 Apollo missions, we barely explored a few square kilometers of the lunar surface, while the moon itself has a surface area bigger than Africa. The whole, "Been there, done that" argument against lunar exploration is quite weak, IMO. There's much more to see and learn there. In addition, lunar farside would make a dandy place for radio observatories. I also like the idea of developing the basic technologies for manned exploration on the moon and then taking them to other places as soon as possible.

However, I have some quibbles with a few of his paragraphs.

Although of fairly ordinary composition, the moon contains the resources of material and energy that we need to survive and operate in space. With its resources and proximity to Earth, the moon is a natural logistics and supply base, an offshore island of useful commodities for use there, in space and ultimately back on Earth.

Water is an extremely valuable commodity in space -- in its liquid form, it supports human life, and it can be broken down into its two components, hydrogen and oxygen. These elements make the highest-energy chemical rocket propellant known. Water exists in the dark and cold regions near the poles of the moon. Scientists estimate that each pole contains more than 10 billion tons of water, enough to launch a fully fueled space shuttle once a day, every day, for over 39 years. The ability to make fuel on the moon will allow routine access to Earth-moon space, the zone in which all of our space assets reside.

Lunar resources, if they're harvestable, would be invaluable for further lunar exploration. If we accept the current launch cost of $10,000 per kilogram to Earth orbit (and that's a number to be taken with a grain of salt - boosters besides the Shuttle can do it considerably cheaper), then the cost of landing that kilogram on the lunar surface is several times higher. You must first accelerate that kilogram to Earth escape velocity (requiring either many kilograms of propellant or months of time and ion thrusters) and then land it on the lunar surface. In short, you'd probably need at least 10 KG (at $10,000 each) in LEO for every KG on the lunar surface. Every kilogram you can avoid having to send to the moon represents a big cost savings.

However, as many have pointed out here, if he's talking about using the moon as a refueling point for further exploration like to Mars, it isn't such a good fit. I believe for those missions, a better alternative would be to return the lunar water to Earth orbit (perhaps using the steam rocket idea) where it'd be processed and used to fuel outbound missions and perhaps space tugs.

The moon's slow rotation, unclouded skies and abundant local materials make it possible to build installations specifically designed to harvest solar energy there. Solar power, collected on the moon and beamed to Earth and throughout the space between the two, can provide a clean and reliable energy source not only for space-based applications but ultimately for users on Earth as well. Lunar solar power solves the apparent "showstopper" of other space-based solar power systems -- the high cost of getting the solar arrays into space. Instead of launching arrays from the deep gravity well of Earth, we would use the local soil and make hundreds of tons of solar panels on the moon.

I'm less enthusiastic about this idea but at least he didn't mention Helium-3. Between the radio propagation losses due to distance and the Earth's rotation, putting energy transmitters on the lunar surface for terrestrial usage is not as good an idea as putting solar power satellites in geosynch. Those solar arrays could be made from lunar materials and returned like the fuel, greatly lowering the launch mass. Also, geosynch satellites are in sunlight over 99% of the time (with eclipse season centered on each equinox and lasting for a maximum of 72 minutes per day), so unless you're talking about building the lunar surface solar array farms in one of the few places on the moon where there's continuous sunlight, you're talking about roughly 2 week on/2 week off power availability. GEO power satellites offer lower propagation losses, continuous visibility to a given ground station, and potentially greater power availability than transmitters on the lunar surface.

Lunar power generation might have one interesting benefit - you might be able to transmit electricity to ion or plasma powered vehicles on their way from Earth orbit to other planets. I like the idea of a reusable tug that could accelerate from LEO to the proper escape velocity, release the interplanetary probe, then decelerate and be aerocaptured back into LEO for subsequent missions. This idea could also be used to launch GEO spacecraft on much smaller boosters. You wouldn't need an upper stage or apogee kick motor to put a satellite into GEO. In the case of the Proton booster, the old numbers that I remember were that it could put 20 metric tons into LEO or 2 metric tons into GEO. The 18 metric ton difference was the upper stage. Imagine the cost savings if you didn't need to launch and throw away that upper stage every time.