I've enjoyed reading your site over the past year... I particularly liked that "Space Travel" thought experiment you published a few weeks ago. It's a depressing but true feature of most systems that most of the energy goes into waste heat, and stuck in a vacuum that is likely to be a very large limitation.... However, I had a couple thoughts about some possible ways around this. I'll talk about two areas - first, the power distribution system, and then possible ways of generating power without generating destructive amounts of heat.
Power distribution: I seem to remember a discussion about high temperature superconductors. Maybe we'll have one for electrons someday, but we have one for photons right now: fiber optics. So assume for a moment that we have a high power laser (yes, I know _that_ would generate a lot of heat, but I'll get to that in the next section :). Right now, we easily achieve .2 dB/km of loss in fiber optics cables. That can certainly be improved. I'm not sure when the nonlinear breakdown starts to occur, but I think people typically use 10 W or more of continuous power in a single mode fiber without any trouble. (Indeed many people go to great lengths to observe nonlinearities at all). Of course, if you are using these fibers for optical power transport and not data transport, you care about the nonlinearities that bother people less; your main concern is not scattering the energy out of the fiber and not destroying the fiber. Atomically even fibers that are available today are fantastically resilient if you don't have a sustained throughput of power. An interesting factoid is that a nonlinear pulsed laser, which basically operates by compressing a high power laser into very short bursts, has an instantaneous power throughput of more power than the entire US power grid output. So I suspect that fibers could in principle be constructed to conduct kilowatts of power or more, with sufficient engineering. There might be bundles of thousands of these, which could collectively transport very large amounts of energy with very little loss or heat generation. Furthermore, there is work on new fibers that have an air (and conceivably eventually) a vacuum core, which would of course increase capacity even further.
So to come back to the original point, if you assume a large laser beam as a power supply, you probably do have the ability with very plausible technology to get that energy anywhere you want on your ship. No cooling or special treatment would be required for these fibers, and moreover, they are quite light compared to metal. It's pretty easy to very efficiently convert this light back into electricity at the extremities of the ship, if that is needed. though I would bet that many things you are assuming require electricity could be done optically. From the standpoint of entropy, it's very nice to work with light because you don't increase it much by transporting it from place to place, and even in space you have a very convenient way of getting rid of all of it, by simply radiating it directly away. The light can also be an entropy sink, of sorts.. if you take a laser beam and do something that chops it up, that might be in principle a way to get rid of entropy from other processes without producing heat.
If you've read this far I owe you my theory of how such a laser beam might be constructed. First of all, I was being a little dishonest before for simplicity, there's actually no reason to even have a laser beam in your power system in the first place. If you're going to generate electricity at an extremity for the purpose of a conventional activity, it really doesn't matter if you shoot in incoherent, broadband light. It will all get absorbed by the photodiode and be converted to heat and current. But let's say you do need a laser, as you would for optical logic. The first thing to realize is that there is really a continuous spectrum between laser light and normal "random" light; it is just a matter of what the "coherence time" is. In other words, if I measure the E field at time t0, how long must I wait until a new E field at t0+dt is totally uncorrelated from the original time? If I am willing to filter out part of the incoherent spectrum, I can obtain an arbitrarily spectrally pure signal which will have an arbitrarily long coherence time, at the cost of lowering the fraction of the power I get into my new "laser" beam. I'm pretty sure that such a beam would end up being indistinguishable from a true laser beam, but I could be wrong on that one. I haven't done the math. The entropy of the light beam is not lost, of course, it just resides with the rest of the unfiltered light. Such filters can be constructed in high Q resonators, and with sufficient engineering will not dissipate much energy as heat, allowing the entropy to stay in light form.
But how to get the incoherent light? Well, an obvious source of the entropy and heat for the ship is the power source. Something either has to react or burn to create the energy, presumably. An obvious candidate for other reasons is fusion, although I think things work out such that any ship relying on this would have to be very large, because fusion is more efficient for larger volumes. But if that's just a function of the surface to volume ratio (ie. too much heat escapes relative to what is genera
