[extropy-chat] Social Implications of Nanotech

[Eugen Leitl]

On Mon, Nov 10, 2003 at 03:44:25PM -0500, CurtAdams at aol.com wrote:
> 
> Right.  The distinction between nanotechnology (machines able to operate
> at very small scales) and Drextech (self-replicating nanotechnology) is

It is very important to keep track of the lingo here. Orelse, we arrive at
the "dental nanopaste" and "scratch-proof nanoparticle car finish" kind of
nanotechnology (I kid you not, last Saturday I was privy to what the Deutsche 
Museum NanoTag/VDI understands by nanotechnology -- cheap potshots at Drexler & 
Merkle included).

> critical here.  If machines have to be constructed then there's no reason to
> construct excess (costly) so they will, as you point out, have similar
> economics to macromachinery.  Only if they replicate will you get a situation
> like biology where the cost of something (in the long term) is basically
> the cost to feed it.

I suggest we talk about molecular manufacturing when we mean molecular
manufacturing, and leave the nanofabs (which, somehow, magically came into
being, and then, suddenly, somehow can't make parts of themselves, nor
otherwise assist in their own manufacturing) roam the plains along the
bands of elves and flocks of unicorns. 

I don't at all think that http://www.jetpress.org/volume13/Nanofactory.htm 
is how it is going to happen, but we shouldn't regard self-replication as
something magical. Even we can do it; why can't machines?
 
> Even with self-replication current IP laws would defer most of the effects
> for the duration of the patent.  Given the demonstrable clout of major 

IP laws has remarkably little clout when it comes to limiting the scope of
information dissemination. This not only includes directly
monkey-consumables, but news and virtual machines (software) as well. All
that while people go to jail for it. Why should blueprints be different? It's
just information, after all.

> patentholders (e.g. Microsoft) they might figure out how to extend patent
> indefinitely.

Even if we introduce capital punishment (all over the planet? fat chance) IP
violations will occur routinely.
 
> >2.  A big question is by what factor general manufacturing devices are less 
> >efficient than specialized manufacturing devices, either in terms of 
> >production time, material waste, or final product quality.

Computers are all-purpose information processing devices. They have
almost completely destroyed all special-purpose information processing devices. A
nanofab with a structural repertoire equal to or a bit beyond biology will completely
transcend biology.

We can make computronium from proteins just fine. It's just spintronics in
SWNT matrix is so much better. And each iteration of the technology assists
its own ascension.
 
> That's a great question and I know there's work on this.
> Specialization -> efficiency goes back to Adam Smith.   Unfortunately
> my vague recollections from reading some of this stuff is that the
> results are contentious. Anything quantitative on this would be great.
> The hip pop versions of complexity theory certainly encourage the idea
> that there *would* be fairly general rules, but I've never seen anything,
> so it's probably not there.  There might be some studies from chip design.
> 
> >3.  PGMDs embody almost *fully* automated manufacturing - if they need 
> >people to step in frequently to diagnose and fix assembly line problems, 
> >they become much less attractive.  While many manufacturing plants today 
> >are highly automated, it may cost quite a lot to produce designs for fully 
> >automated production processes.  So design costs may be a lot higher.
> 
> Drextech vs. nanotech again; only if you self-replicate can you erase

Let's call it molecular manufacturing, please.

> your inital design costs.  The trend in the modern world is to higher 
> specialization.
> The big exception is computer chips.  But, to be fair, to *use* the 
> generalized
> chip requires a lot of *software* design and I'd expect the same for 

I get all my software from the global network for free. I could use silicon
compilers (for free, of course) to design my own circuits, should I want it.

If I had my own desktop nanolithoprinter, you can assume I'd quit my job and
started tinkering. So would many, many others.

> nanotech.  If Drextech doesn't fly, or at least doesn't soon, I would expect
> nanobot to be made by methods similar to chip manufacturing, etching and 
> layers large numbers of duplicates on a single block.

I would assume large-scale autoassembly and bottom-up to produce the first
simple self-rep systems. 
 
> >4.  The manufacturing fraction of the cost of most consumer goods today is 
> >rather small (15%), and only part (~1/3) of those manufacturing costs now 
> >are the physical capital, rather than labor and design.  So it is not clear 
> >how just lowering those manufacturing costs will have a huge effect on the 
> >economy.
> 
> Good point; not too much effect there.  I'd think more in terms of nanobots
> in consumer hands, able  to exert physical force on scales an in locations
> not currently possible.  Dishwashers that can actually scrub; carpets that
> can roll themselves up; home under-gum plaque removers; indwelling
> periodic catheterization devices; stuff like that.

Machine which can actually think and make improved copies of themselves;
stuff like that.
 
> >5.  If the cost of designing and building an effective self-reproducing 
> >PGMD is much higher that of ordinary PGMDs, there might be plenty of 
> >ordinary ones around before any self-reproducing ones appear, minimizing 
> >the social impact of this transition.
> 
> I think manufactured PGMD are almost guaranteed to precede self-reproducing
> PGMD.  How can we build something to do something as complex as 

PROCESSIVITY. Think about it. How do you scale nanograms/h in a large hall 
to kilotons/h? Not without self-rep.

> self-replication before we're *very* good at making it?  I think the
> transition from manufactured to self-replicating would still have 
> *enormous* social implications, though, if it happens.   The values 
> for centralization and for current manufacturing stock could go "poof" in
> a big hurry, depending on the speed of the transition.

We should see the impact of cheap photovoltaics on decentral energy
production within the next 10-15 years. Ditto wireless networks linking up to
a global mesh. The centralist paradigm is fighting a battle uphill.

-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144 http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20031110/edfee085/attachment.bin>