Re: @John
"Seriously: a radio-isotope power source is as "hot" as it will ever be at launch. If the launch fails, some fairly short-lived really horribly radioactive crap gets showered into the atmosphere or the ocean."
Erm, Plutonium is embarrassingly long lived. Well, embarrassing to your point, that is.
Pu-238 has a half life of 87.74 years.
The current thermoelectric generators operate off of the decay heat of that Plutonium.
Plutonium also has the distinction of being one of the most toxic of artificial radioisotopes.
Now, a reactor uses far more uranium (plutonium can be used, but it's not really practical due to expense. Thorium can be used, but there is currently no microgravity rated design to separate out various "poisoning" isotopes that would halt a reactor) than an RTG though. Many, many kilograms more.
Plus really nasty, toxic metals.
Now, think of the last few Russian fireworks displays that were supposed to be orbital shots. Think of a handful of tons of really nasty metals and hundreds at a minimum of kilograms of uranium raining from the sky.
Compared to a few pounds of that PU-238. In a hardened, small and survivable unit (note the design in the video, including the aeroshell).
I'll go with the RTG. Less chance of a loss of containment in a launch accident than a reactor.