Matthew Hagston wrote:
The faster you could 'spin' it (get it closer to the speed of light). The
less mass you would need to start with though right?

The only advantage this gives you is in density. In
all other aspects this scheme is dramatically more
difficult than just using a pile of matter. And it
is insanely more difficult and cumbersome than just
using acceleration (e.g. rotation). Unfortunately,
even that benefit is very tiny and almost
inconsequential. The problem is that ordinary matter,
even diamond or buckytubes, is just not strong enough
to store enough rotational energy to significantly
affect its mass. Consider that the maximum amount of
energy storable via this method is similar to the
energy stored in the chemical bonds of the material,
and that energy is similar in degree to the energy
released by chemical explosives. Now compare that to
the energy density of matter and the amount of energy
released by just a tiny amount of matter (e.g. much
less than 1% of the mass in the case of fission or
fusion explosions, 100% in the case of anti-matter,
matter annihilations).

Even given all that (namely, the ability to store
on the order of 20 *megatons* (TNT equiv.) of energy
per kg) you would have improved the mass/energy
density by a mere factor of 2, if that. Most likely
though, since you would be forced to use materials
with strong covalent bonds (e.g. carbon compounds)
you would start out at a relative disadvantage in
terms of density compared to using the densest
matterials avaible (e.g. Lead, Uranium). So you
would gain no advantage at all.