SSI Activity (and its rebirth?)

> I have to say, I'm with Mike on this one. Many nanotech researchers don't think nano-scale
> assemblers will ever happen; some that are, are looking at biological processes for
> inspriation. The problem with the latter is that it's entirely possibly the replication
> processes will require things that the Moon isn't known for, to be able to replicate, such as
> an organic soup at ready disposal.

See my other note to Mike for part of this.

On the issue you raise of biotech, there are many webs of different
strains of bacteria that live deep beneath the Earth, chew through rock,
etc. and organisms like molluscs that make protective and translucent
shells, so in terms of just biotech, I would think this is feasible
(even if I wouldn't go at it that way -- through see some of Hamilton's
sci-fi writings related to his "Edenist" habitats which grow from an
organic seed and raw materials delivered to it while getting power from
magnetic fields of gas giants).

> So, if you want to create a self-replicating habitat, you probably don't want to start on the
> Moon. This isn't a concept killer, but it complicates things.

The general assumption is that asteroids are much better sources of raw
materials, with many containing carbon and water in abundance, and some
ones of those also containing metals and silicas. I would not target the
moon for this activity (unless water and carbon was found there in
abundance -- more likely if there is an asteroid collision with it,
recent (and perhaps intentional)). Someone from NASA once told me the
astronauts who landed on the Moon were specifically told not to pick up
rocks that might have been from asteroids or comets (they wanted pure
"Moon" rocks for the scientists) so we many have a very poor idea of
what is actually available there. But I also think the advantage of
asteroids is that solar panels can work almost 100% of the time there,
whereas on the moon powering things during a two week night is more
problematical (still, a lunar replicator might be designed to work only
every other two weeks if it could survive the cold night -- although it
coudl not easily make a habitat at Earth normal gravity there, which
means the mass driver etc. which adds complexity to the entire system).

> What attracts me to O'Neill's vision is that it's based on present-day technology. (A caveat:
> while my copy of The High Frontier is on order, I haven't read it, but I'm working here with
> the ideas I've read on the Web). No major advancements have to be made; in a sense, it's
> just looking at building construction when you subtract gravity from the force equations.

I think part of the issue is what is meant by "present day technology".
Is something present day technology (like a space habitats) if there are
no complete 3D plans for them? Or related very detailed simulations for
testing out if they work? Or there are no building permits issued yet
for related things (rectenna farms). Just look at the issues Biosphere
II unearthed as they actually tried an experiment for a small closed
ecology. Part of my frustration with SSI in the past is that space
habitats are classified as present day technology and so not worth of
SSI research funding to actually design them. This reminds me a tiny bit
of an AI-related essay where the author (forget who) wrote that the AI
field was crying out for a few "well documented failures" in order to
make progress, because what really happened in practice was graduate
students got PhDs for projects that didn't really work as described in
any general sense and they would also never share the code for someone
to build on or examine (in part for this reason), and so that every
interesting area of research was now considered "done" (even though it
wasn't) and so no person could study anything useful any more, leaving
the AI research field drowning to death in its own supposedly successful
theses. :-)

> I think what could be done 'cheaply', however, is setting up an automated mining system,
> which could do much of the initial activity O'Neill had in mind without involving people.
> Sort of an ant colony that packages together bundles of minerals to be sent to an L5 point
> via a mass driver, etc.

That's all true, and I think some part of what he envisioned. Still, the
quesiton is is everythgin shipped up? And a big part of Gerry's vision
recolved aroudn the moon and the mass driver, which as you point out
isn't a great place to start for lack of much water etc. (even if there
may be pockets if you could find them and then work in specific locations).

> I like the idea of such robots being able to reproduce somehow - possibly by sending just
> the manufacturing plant and a few starter robots to tap into lunar resources to start the
> process of generating more of them - but the problem here is again the available
> resources on the Moon. One of the most important resources for manufacturing is water,
> especially when doing anything high-tech, like making computer chips. Literally tonnes of
> water go into making a single computer, for example. While this could no doubt be done
> with less water - the main reason it's not is economic - it's still a requirement.

Modern manufacturing plants can have basically no discharges if they
clean their water (or air, etc.) and reuse it. They only use a lot of
water etc. because it is cheaper in the short term to dump it and let
someone else suffer the consequences than filter it (this is a general
problem with capitalism -- you can make profits and be more competitive
by avoiding paying for such external costs).

? Assuming
> that there's no water on the Moon (the jury's still out on the presence of water at the
> poles, so I'd say it's important to plan for the case where there's none), then you'd have to
> ship lots of water to the Moon with your assembly plant, and said plant would have to be
> able to manufacture robots, purify water and recycle it while losing only a negligable
> amount of the stuff.

Again, asteroids are where it is at. Some Earth crossing asteroids are
easier to get to than the Moon in terms of fuel. All asteroids are
certainly easier to get off of, especially if you make a big object
intended to be moved to where people can easily travel to it in a matter
of days.

> Alternatively, it could manufacture some components, have the rest periodically shipped
> to it, and make more robots when the resources become available.
>
> I don't think the idea of just growing a colony out of a rock is realistic, at least in the near
> term. Getting robotic mining working does seem doable to me - it's more along the lines
> of Rodney Brooks' research (http://www.ai.mit.edu/people/brooks/index.shtml) than
> anything absolutely bleeding-edge. Getting it working is doable in the short term, if
> challenging.

Actually, the reason we don't have robotic mining right now everywhere
is mining unions. We have the technology for 100% automated mines with
no human casualties. It's a sad thing to read of mining disasters still
occuring.

> Getting it reproducing is still science fiction. That doesn't mean it's
> impossible, but there's a lot more work that has to be done first if it is.

James P. Hogan has some descriptive writings on what this might look
like in his _Code of the Lifemaker_ novel. Also, his _Two Faces of
Tomorrow_ has a space habitats with automated industrial plants (one
novel read in college that inspired me very much down this road).

I agree it will be a lot of work. The issue here is, in part, is it less
work than CATS? And, even if it is more work, is a better objective than
CATS? With CATS you still have no where to go. With DOGS :-)
(Development of Generative Space-Habitats) you have a place to go to in
space. Which is better, no place to go and a cheap way to get there, or
a place to go and an expensive ticket to ride? I'd rather there be a
place to go, since that would eventually drive the ticket price down by
itself.

--Paul Fernhout