There seems a generalised belief in most works that nanotech will automatically be self replicating, von Neumann machines. And none of them compare with living cells and realise the most important function is the energy generating and storage system (batteries are really not on
), taking up far more of the total mass than information storage.
But actually, whatever the first uses of nanomachines might be, unless you are going to evolve them (extremely dangerous) many generations are going to come out of factories, work while they last then wear out and replaced by new ones. No self repair, no complete instruction kits for constructing "offspring", so lower initial cost (but continuous running costs) no "grey goo" syndrome, no risk of mutation.
Which could be a good thing for those cancer zapping nanites (assuming they can be trusted to recognise the right cells; and we're not going to start injecting people with them until we're dead sure of that, are we?), since you can energise them from outside (think of an induction coil run at a frequency that a tuned circuit inside the machine could receive and pull power from: a very high frequency, given the size of them). Now, except when you're within this coil, they're passive, and floating around the body (and being filtered out of it by kidneys), totally harmless. And each class of nanite would have a specific function, and a specific powering frequency, so you could inject cancer zappers and cholesterol disposers in one patient, and run tests as to which one is useful
now.
All these "geo" words are based on Earth (yes, even "geometry", measuring the Earth); if your crew are going to mars shouldn't they be "Areologists" or areophysicists or something?
