Eliezer S. Yudkowsky wrote:
> Doug Jones wrote:
> > I've been thinking about this question off and on for years, and the
> > timeline keeps shrinking- the only hard limit I can see is the doubling
> > time of the assemblers and the power available for making fuel. With a
> > little bit of design ahead, and a lot of really clever (but non-sentient
> > software), it should only take hours.
> How clever does the software have to be? I mean, what exactly are you
> envisioning it doing? (I'm assuming that we're using design-ahead to
> build nanocomputers that run Linux so that the software can be tested,
> in miniature, on existing computers.)
It would be nice to have something close to a "design team in a box", with complete materials properties, stress analysis, CFD flow modeling, detail design, and manufacturability analysis all in one integrated package. "I want a throttling valve for 500 kg/s liquid oxygen at 20 MPa, 10% to 100% flow and cutoff, to fit in this volume and require minimum mass, reliability of six nines, fail closed (or open), and don't forget materials compatibility so that it won't self-ignite." The engineer then chooses from an array of options presented by the software, and fifteen minutes later, after the CDR, the complete part spec is downloaded to the manufactory. Part is in hand an hour later, assembled into the vehicle an hour after that.
> What kind of "design-ahead"? And how much do you need to know about the
> exact nature of the assembler breakthrough beforehand? Suppose the
> drextech is soft instead of hard? How does the complexity required to
> build a hard-nano spaceship in a vat compare to building soft-nano grey
> goo that reproduces in the wild?
In that case, you roll back the materials to bulk compositions available today, such as zero-CTE dispersion hardened copper instead of diamond (sigh). Then you have soft nano robot machinists operating conventional machine tools for most of the fabrication. If effective micro-scale hard manipulators can be built cheap, these can be sacrificed by the million for tasks such as propellant tank welding (in the extreme, a 100 micron crawler is mostly a thermite charge for doing a spot weld. It crawls into place and goes phhht on command. Kinda like dinner in the Restaurant at the End of the Universe.)
> How much power do you need for making fuel? Can the fuel be
> manufactured in advance? If not, is there some way you can use
> nanotechnology to get around the problem by concentrating existing
> resources, at least on a once-off basis? How about fusion drives?
I was limiting myself to known physics and chemistry, not considering things like Bussard's electrostatic confinement fusion. Certainly if the economic resources are available to the conspirators, er, planners, then thousands of tons of propane and LOX could be stored well in advance. If soft nano can build decent solar arrays, and the doubling time is short, then a few square miles of desert land (or a bloom at sea) could provide gigawatts of power to synthesize kilotons/day of propellants. Fuel could be made most quickly by fermenting biomass, and LOX is a nickel a kilo even these days, mined from the air. One gigawatt could produce 100 kg/s of propane/LOX from CO2 and H2O via electrolysis and the water shift reaction; using air and biomass would get that up to 1000 kg/s easily. That would be 40,000 tons of propellant in under 12 hours, enough to put about 500 tons into orbit or 100 tons onto the moon with low performance rocket designs.
> > Bottom line: Give me the right tools and I could be out of here tomorrow.
> How fast does that scale up? Could you evacuate cities, or at least
> provide the evacuation vehicles to do so? Could you fire cities, or at
> least buildings, directly into orbit?
Here you mostly run into the rocket equation and power demands, which then have economic and political impacts. Sure you can put 500 tons into orbit in one lump, but the beast is 40,000 tons at launch, burns 270 tons/second of propellants, and the jet power is over a terawatt. The noise at 100 km range is 120 dB- folks will *object* to this. Not to mention the 1000 ton empty first stage that falls out of the sky at the speed of sound 200 km downrange. Hard to evacuate LA that way.
With cheap labor, cheap engineering design, and cheap raw materials (mostly scrapyard steel and seabottom ooze), big rockets could be built quickly and cheaply. Details would depend on the exact capabilities of the soft nano, but with soft nano doubling like bacteria, a day would be enough time for the manufacturing. I'd want to launch a few and analyze the results before riding one myself, call it a week for the sluggish humans to catch up with the machinery.
-- Doug Jones Rocket Plumber, XCOR Aerospace http://www.xcor-aerospace.com