Rainbow Goo

[See GreyGoo]

  1. Reproduction means "the survival of the fittest" starts to apply.
  2. "The fitness of the whole is greater than the sum of the fitness of the parts" means increasingly more complex (fitter) ecosystems, with their component individuals, will evolve.
  3. Thus individuals that can team up and create synergy will more often survive.
  4. Specialized individuals create more synergy than homogenous individuals.
  5. Therefore, I'd guess RainbowGoo :)

That'll work fine if your goo can mutate. Otherwise, it just stays grey. I suppose, from a human perspective, it makes little difference - all the works of man get obliterated either way.

But at least the Rainbow stuff could eventually evolve into self-awareness again. Perhaps those rainbow-minds would look at the stuff that makes them up and wonder at how random chance could have resulted in such clever engineering.

I think the point of the original author was that that is exactly what has happened -- we humans are RainbowGoo. -- StephanHouben

Regrettably, no. To quote RichardFeynman, "There's plenty of room at the bottom". Human cells are many thousands of times larger than the putative EricDrexler assemblers. They're microscopic, not nanoscopic. Compared with the assemblers they're also hideously slow and inefficient. Assemblers designed by intent seem likely to be vastly more powerful than the stuff that we're made of. --PeterMerel.

But how do you keep the nanoscopic assemblers tamper-resistant the way human cells are? I think that in order to make assemblers tamper-resistant, you'd end up adding so many auxiliary mechanisms that they too would end up microscopic instead of nanoscopic. --TaralDragon

Well, for a start, you can encase your assembler in a shell of diamond and still keep it 1,000 times smaller than a human cell. Now you might still manage to tamper with a beast like that ... but you'll have a much harder time than you'd have tampering with a jelly-like human cell.

To get an idea about the scale and style differences, think of a couple of thousand ants eating your picnic. Human cells are comparatively picnics of molecules performing most of their tricks through osmosis and statistical assembly. Nanotech Assemblers are pick-and-place robots like ants, but a million times smaller and faster. The reason for the difference in scale and style is the assemblers are engineered for efficiency, where the cells just have to harmonize with other natural processes.

How can we do engineering like this? We're not smarter than nature - we are natural creatures ourselves, and thereby instruments of nature - but we've often found ourselves able to engineer devices that are faster and more rapacious than the ones we find in the natural world. Whether this really can be done isn't an opinion poll - dig out Drexler & Merkle's NanoSystems and see the designs for yourself.

But how do you keep the nanoscopic assemblers tamper-resistant the way human cells are?

Human cells aren't tamper-resistant. What makes you think they are? And what makes you think that nanomachines need to be?
One thing I realized about the runaway nanotech scenarios - if they were likely, wouldn't we see a cellular equivalent in nature already? Yes, the nanomachines are postulated as smaller, but we've seen bacteria in seriously hostile environments, so why haven't we seen something the size of a virus that can disassemble a cell? My hypothesis is that there is something that is being ignored by the GreyGoo theory, something akin to the recent report (I saw it on Slashdot) that showed a more difficult time getting a nano-manipulator to release a molecule than previously expected.

wouldn't we see a cellular equivalent in nature already? No, because if there were such a thing, we wouldn't be around to see it. That doesn't mean it can't exist, just that it and us can't co-exist. In general the nastier something is, the more readily it is constrained by epidemiological considerations, that is, the nastier you are, the more poorly you spread around.

What do you think Tuberculosis, Ebola and Necrotising disease are? They're examples of bacteria disassembling the human body. The only reason some freak disease hasn't wiped out humanity completely is because the human body can fight back and diseases are highly environment and species specific. Needless to say, nanomachines built on a non-biological basis will not be species specific. If someone designed them that way, they could munch on every species on the planet, animal, vegetable, bacterial and mineral, photosynthesizing all the while.

And unless someone specifically designed them for it, nanomachines will never be digestible by macrophages or neutralizable by T-cells, quite unlike the weakling Ebola virus. I mean hell, some people actually survive Ebola. Further, the small size of nanomachines compared to bacteria is sufficient to confer on them awesome power. You can fit the functionality of a bacteria in the size of a virus. Now imagine for a moment self-reproducing Ebola viruses. That survive and are transmitted by dry air. That infect the entire freaking biosphere. Enough said.

It's worse than that. The best viruses and bacteria can do is be transmitted by dry air. It appears that nanomachines can be designed to infect dry air -- to use air to build more copies of themselves. --AnonymousDonor

I'm not doubting the theoretical power of a nano-disassembler - I'm questioning the practical power - the Slashdot-referenced article discussed that the research was finding that it was much more difficult to let go of and atom/molecule than most nanotech theories were postulating - IOW, it was not possible to do the precise positioning unless there was another force (such as substrate covalent attraction) to draw the atom/molecule to it, and even then, the nanobot might not be able to let go. Look at viruses - they've had, what, trillions of generations, and not one of them has approached the efficiency that nanotech mavens tout. Nature is hardly the best designer, but She does get aimed in the right direction a vast majority of the time.

Can't agree with this last. Humans have exceeded the capabilities of StochasticEvolution? in everything they've ever built. We see no supersonic birds. No steel buffalo. No silicon DNA. Not in trillions of generations. But EvolutionByDesign? produced all these things in just a handful of human generations. When EvolutionByDesign? occurs on the nano-scale, we should expect nothing less than a whole new ball game. It likely won't look much like what we envision today - but it sure won't look like what's been happening without us.

Now where I agree with you is that GreyGooIsEasilyContained.


View edit of January 22, 2004 or FindPage with title or text search