An article for @global-risk-sig.org
There are three kinds of lunar resources - those we can use someday, those we can create someday, and those we fantasize based on faulty reasoning. Most claims for lunar resources are wholly concerned with the last.
"Use someday" will involve designing machines and processes entirely different than those we've spent giga-man-years developing here on Earth with the kinds of stuff we find on Earth. Silicon, vacuum, and space UV will be excellent raw materials for very different and very useful machines, and we WILL learn how to make them with somewhat less than a giga-man-year of development effort (but much more than a mega-man-year).
Not germane to near-term goals.
"Make someday" will be the long-term gardening of the inner solar system. REALLY long-term, millions of years, way beyond the imagination of exponential-millenialists. Most of the rock flying around up there is stone gravel. I can get that delivered for dollars a tonne. I have 100 megatonnes of undifferentiated rock under my back yard.
But some of it is planetary core, differentiated by heat and gravity. Nickel iron is immediately useful.
The asteroids are Lyapunov-unstable; perturbations grow exponentially. Very small nudges grow to very large trajectory changes with megayear time constants. Job one is characterizing and tracking EVERYTHING tonne-size and larger to micrometer precision, then forecasting future trajectories.
Job two is using microthrust to steer the "useless" rocks into Mars and the "useful" rocks into the Moon. It will take millions of years to clean out the inner solar system of threats to Earth, but there is plenty of material with near-term collision potential. Shifting trajectory by a centimeter per second can turn an Earth hit into a Moon hit in a couple of millennia.
We can "do trajectory" to micrometers per second today, through a thick blanket of turbulent atmosphere. Global precision geodesy is defined by the LAGEOS satellites, with which we observe continental drift, and calibrate GPS. Imagine what we can do with MANY observatories in space.
Someday, we will deliver specific precision-assayed asteroids to specific prepared locations on the Moon; I imagine deep curve-walled "melt pits" lined with refractory rock. The arrival kinetic energy at the Moon is about right to melt but not vaporize the asteroid. Ladle out the melt, zone refine it, and mold an enduring space civilization from it.
ENDURING; not the infection-model machine plague flaunted by Silicon Valley software geeks. As is, we've got another half billion years on this Earth of ours, should we choose that path. Wisdom is about what we SHOULD do, not what we CAN do.
A tiny tiny fraction of the asteroids are "in the groove" for Earth impact (or lunar impact processing) in the near term, but there are A LOT of asteroids out there. We can arrange a steady supply of useful materials; just the Near Earth Asteroids contain about 300,000 years of nickel-iron at current global usage rates. There is vastly more material between Mars and Jupiter, the source of the NEAs we have now, waiting for just the right nudge.
The process reminds me of log management at the McMillan-Bloedel lumber mill at Port Alberni, British Columbia, which I visited for a technology tour a couple of decades ago. The storm-battered coast of Vancouver Island produces some of the gnarliest trees (and logs) in the world. As logs were cut and delivered, they were 3D x-rayed for knots and imperfections, bar coded, and stored in a vast log pond, with their data stored in the small hard disks of the puny Sun workstations of the time. When an order for custom-cut lumber came in, they sorted through their log data base, mapped the 3D boards into the 3D trees, then pulled the logs and cut them to avoid the knots and imperfections (which became particle board or paper pulp). That mill produced some of the finest lumber in the world: long clear-span blemish-free beams for the repair of temples in Japan. Such unique "perfect" wood earns a 20x to 100x markup over ordinary construction grade lumber. Crap to gold.
If we are WISE and SMART, we can do the same thing for the entire solar system, all 380 trillion terawatts of it, and over a VERY LONG TIME extend the process to the entire galaxy. No, you and I will not live to see it; but then, the way most of us eat and exercise, we won't live long enough to see much of anything.
Choose life, not pathology. Create aids for human accomplishment, not autonomous robot plagues. If you can't tell the difference, leave the software cubicle farm and go work on a real farm for a season or two. You might invent something that helps farmers, so they aren't as likely to vote for people you don't like, out of justified fear of what you are inventing now