> Cold fusion is only ten years away. What's weird is that is has been ten
> years away since 1945.
> >Also, someone recently proposed (was it Mr Clarke?) that cold fusion would
> >become a household phenomenon, and kill the power production industry in
> >next couple of decades.
I'll believe it when I can order my C.F. household power source from the net!
Hot(!) Fusion has been 10 years away since 1945. Current designs would get more power out than they put in. They still have a way to go to have a good way of extracting the power for real use. Cold fusion is still a very questionable phenomena and there is no evidence that it can be scaled into something useful.
Yes, I can buy a little rocket and launch it in my backyard, and it is possible to build big rockets to launch people into outer space, but I still don't (personally) have one.
Contrast rockets with the huge amount of evidence that computer chips can continue scaling down to the atomic levels, that fiber bandwidths will continue to increase for quite some time, and that self-replicating machines can build interesting organisms of a variety of sizes.
When you make a prediction and you are extrapolating from known trends or things for which there is evidence all around you, you are probably making a relatively safe statement. When you make a prediction that requires absolutely new science and engineering you are sticking your neck way out.
I'll give a counter-example of something that would totally deflate the cold-fusion balloon (even if it could be made to work).
Fusion produces neutrons, neutrons have this nasty property that they go through things and require a lot of shielding. They also make materials radioactive. So it is likely that any cold fusion reactor you produce (if you intend to put it in your home) is going to be heavy and take up a lot of space, at least if you intend to get a useful amount of power out of it.
On the other hand, as Eric I believe has pointed out, you can get nanoassemblers to assemble solar cells that are extremely thin by current standards, much less than 1 Kg/meter^2.
Fusion & nanotech are not comfortably compatible technologies (the nanobots have to run around, constantly repairing the damage being done by the neutrons while being damaged themselves).
So, it would appear that on the basis of the quantity of materials required and on a reliability basis that solar cells will be less expensive than fusion reactors.