[extropy-chat] Re: Nano-assembler feasibility

[Chris Phoenix]

This post will be about politics.

Brett Paatsch wrote:

> The vision articulated in Nanosystems and by Eric Drexler and
> others in the Foresight Institute has already played a very useful 
> role in getting funding into nanotechnology. 

The nanotech that's being funded is very cool, but the NNI is busy 
distracting us from the fact that they're not working on Drexler's ideas.

> The question is is it possible? And the answer is we don't know
> until we see a systems specification that shows a full set of 
> components necessary to produce an assembler (at any scale
> would be a very good start)

No, we don't know for sure.  But if we're talking about a massive 
national security issue, is "We don't know" an acceptable state of 
affairs?  Or should we be working to learn more?

The more we learn, the more likely it looks.  No one has come up with 
anything that would keep it from working.  The designs are getting more 
detailed, and more scary, all the time.  And everything we look at turns 
out to be easier than we thought.  If you want to say "We don't know" 
until it's demonstrated by someone else, go right ahead.  That's as bad 
as NASA saying "We don't know that there's a problem with the foam."

> "In politics, as in engineering, opportunity costs matter as there
> are only a finite amount of resources to go around."

Well, we have some scientists and engineers saying "This looks possible, 
we think it could be developed within a decade, and if so, it would 
change the world."  We have others, for a variety of plainly 
identifiable political reasons, saying "We don't think it's likely and 
we don't want you to study it."

If you were in charge of national security, would you adopt a strategy 
of ignoring it and hoping that the politician-scientists were right?  Or 
would you spend a little bit of effort extending the studies, looking 
for weak points in the theory and strong points in the engineering?

If you were in charge of a technology business, would you bet your 
business that the politician-scientists were right and there was no 
possibility of a nano-enabled manufacturing revolution on the horizon?

> On Sunday, March 28   Chris Phoenix wrote:
>>If we're talking politics, then we should  mention the geopolitical
>>implications of some ambitious nation making a breakthrough
>>while the current superpower is saying, "Let's not try anything till
>>we see it done." 
> 
> Talking about politics (of funding and risk management):
> 
> Without a specification of the full list of component parts how
> would you *quantify* the risk of that?  It looks unquantifiable
> to me.

I'll start by noting that "quantify" isn't a word normally used in politics.

So we're asking, "Is the risk that this is possible, and that someone 
else is working on it, and will succeed before we start, significant 
enough to worry about?"

The possibility of it working at some point, vs. being impossible, is 
not affected by the presence or absence of blueprints today.  No one has 
tried to make blueprints.  So to estimate the possibility of it working, 
I'd look at the solidity of the theories, their application, and the 
engineering built on them.  I'd try to poke holes in it.  Since this 
isn't a technical post, I'll merely note that no one has managed to do 
this in over a decade, and even very smart scientists dedicated to 
debunking it can only handwave about strawmen--while the proposal gets 
more and more detailed, with fewer and fewer uncertainties.

BTW, I wrote Whitesides a week or so ago, asking him what he thought 
about superlubricity vs. what he's said about nanomachine friction.  He 
didn't answer.  I wrote to Mark Ratner asking about whether there was 
any evidence that mechanochemistry couldn't build diamond.  His answer 
invoked the scaling problem--which is not a problem if the rest of it 
works.  When I pointed this out and asked again about chemistry, he's 
been too busy to answer for a month and a half.  He says he'll be less 
busy in April... I'm looking forward to seeing whether he has anything 
to say.

The chance that any particular group (nation, company) is already 
working on it is very hard to quantify.  And we're likely to 
underestimate this for two reasons: First, because we forget that not 
everyone listens to NNI propaganda.  Second, because it's very rapidly 
getting cheaper, and the number of groups that could develop it is 
exploding.

The chance that they'll get it before we do depends on whether it's 
possible--and at this point, the burden of proof is on the doubters. 
And on whether we'll start looking at it before someone else does a 
start-to-finish program.  If you are representative, then we probably 
won't start looking.  We'll keep hoping that it's impossible.  So the 
chance of someone else doing it first also looks very high.

Should we worry?  Yes.  Either there's a gaping but unnoticed flaw in 
the theory that's been around for well over a decade.  Or the people 
studying it most closely have underestimated its difficulty by multiple 
orders of magnitude.  Or we'll have it by 2015, perhaps quite a bit 
earlier.  Are you willing to bet your life that the engineering work is 
flawed?  That would be pretty stupid.  And by arguing the way you do, 
you're betting our lives as well.

Chris

-- 
Chris Phoenix                                  cphoenix at CRNano.org
Director of Research
Center for Responsible Nanotechnology          http://CRNano.org