It seems obvious that the lunar settlement must pay its own way eventually, & should pay as much as possible as soon as possible, to the extent that necessary activities are not interfered with. In other words, we need to begin bringing in money (as in commercial income) as soon as possible -- for choice, before First Landing, but certainly as soon thereafter as possible. We also will need to have diverse revenue streams which do not depend on some particular precondition. For example, it would be unwise to stake our commercial prospects on shipping oxygen to low-Earth orbit, since we cannot be sure there will be any users there & it will take us some time to complete the launching apparatus. We should be prepared to exploit this opportunity to the extent it exists, but making ourselves dependent on it is a recipe for disaster. Likewise with raw materials for solar power satellites, Luna-based solar power transmission, satellite repair, &c.
This is the thread for making the case for profit-making ventures.
I will start off with one which might not be obvious to people : astronomy. Amateur astronomy has become big business in the past two or three decades, at least partly because it has become much easier to buy & use telescopes with the advent of mass-market production, computer drives, &c. With the advent of the Internet, there are now robotic, "remote" or "unattended", observatories in dark-sky locations which allow a user in a built-up urban area to do his observing in the comfort of his own home, rather than driving out into the countryside & huddling in the cold. These break down into "private" & "shared" instruments. The shared observatories may be in the hands of a club, or may be owned by a commercial enterprise, or owned by some other entity (such as an established observatory) & rented out. For example, "slooh.com" has its own telescopes & also rents a telescope from a Chilean university, & sells subscriptions which allow users to specify what objects or parts of the sky they want photographed & (to some extent) to schedule the photography. A private remote observatory is typically owned by an individual, who has some contractor set it up in a promising site (typically several are grouped together under common management) for a fee, & may never see it or interact with it except by the software interface on his computer.
Now, if there is a sect of amateur astronnomers willing to pay $50 000 for a remotely-controlled mountain observatory, there must be a smaller sect willing to pay $1 million for a space telescope. Telescope foundations could be made from cast basalt & put in an area relatively unaffected by lunar industry, with its dust & vibrations, & the most massive part of the telescope itself would presumably be the mirror, which could be made of lunar glass coated with lunar aluminum, or (in the absence of air oxidation) simply machined from an aluminum blank. The imager (essentially a commercial digital camera) would probably have to be imported, & at least part of the control electronics (although the circuit boards might be fabricated locally from glass), but that is really a handful of components. It would probably not be difficult to cut mirrors up to 1 meter diameter, which would be considerably larger than anything an amateur would ordinarily use -- in fact, such a telescope would have more than a little appeal to professional astronomers, especially if it could be fitted with multispectral optics. In the year after First Landing, we might put 100 or more telescopes in place. At a million-dollar price point, that would be a nice step toward offsetting our costs. It is not entirely unreasonable to suggest that we might sell a thousand at $1 M, and (some time later) another thousand at $250 000 or so, which would recover $1.25 billion (1/5 of our launch costs on the Skylifter basis).
