if the funding is there. Given the ever increasing radio interference on and above the Earth, it may become necessary to do it if radio astronomy is to survive.

Overlooking the all too common stupid mistake in the article's title about "the dark side of the moon", there are several approaches to developing a farside observatory.

1. You need electrical power. This might come from solar panels mounted in one of those places where the sun always shines on the moon and using cables to carry the power to the observatory. It might come from some form of solar power satellite. It could even come from a small nuclear powerplant if the Greens didn't kill it. Regardless, you need electrical power around the clock. Simply putting solar panels at the observator site isn't a very good solution unless you're willing to shut down for about 2 out of every 4 weeks for the long, cold lunar night. As we've discussed before, electronics equipment doesn't really like getting very cold...

2. You need communications between the observatory and the controllers on Earth. Once again, this might be accomplished by putting an antenna somewhere on lunar nearside (I suggest at the same place you'd put the solar arrays) and running a cable to the observatory. You might even be able to superimpose the communications signals on the same wire as the electrical power. An alternative is to put some simple communications satellites in lunar orbit.

3. You need your antennas at the observatory. If you're going after low frequency signals, then you antenna is going to be big. If the site is to be developed robotically, it might be possible to use a spin-off from the rumored "black world" Royalite ELINT satellites. According to "Deep Black", these satellites sport some really big antennas to pick up weak terrestrial signals from satellites out at geosynch.

4. Your observatory electronics would need to be rugged, reliable, and redundant. It's a long way to go make a service call. It needs to be able to withstand harsh heat and cold, radiation, etc. and capable of operating for years without human intervention. Derivatives of existing satellite electronics should do the trick without a lot of extra R&D.

In concept, the idea is pretty simple. Actual implementation is a bit more difficult but that's engineering for you. As always, the first and hardest battle will be getting the necessary funding. This stuff isn't cheap.