Re: Leading nanotech experts put 'grey goo' in perspective

From: Jim Logajan (JamesL_at_lugoj.com)
Date: 06/24/04 

Date: 24 Jun 2004 05:26:58 GMT

Chris Phoenix <cphoenix@best.com> wrote:
> The intent was to say: there's no reason anymore for
> building anything like a dangerous self-replicator, and so anyone who
> does it should be considered uncivil / destructive.
[ ... ]
> I'll anticipate your argument: no, I don't think such things have to
> be constructed for any legitimate research purpose, even to learn how
> to fight them.

If you mean by "dangerous self-replicator" a self-replicating nanorobot
that is mobile in any way and able to forage its energy and feedstock from
the biosphere and serves no purpose except to procreate or to cause others
loss of life, liberty, or property, then your intent seems reasonable. If
your definition of what constitutes dangerous includes a larger range of
devices than that, then this seems a good point to present your definition.
I may actually find myself in agreement with some expansion of the above
set, but I'd have to see your definitions. I'm not looking to argue for the
sake of arguing - but neither do I wish to let pass without comment or
criticism any statements or pronouncements I believe are in need of
correction or rethinking.

> Whether this is
> embodied in law, in the Foresight Guidelines, or simply in
> common-sense agreements among scientists, is a policy detail: which
> will work best to minimize the number of stupid and/or malicious
> attempts?

I don't consider enforcement a policy "detail" - I think it is the very
essence of policy. In one case a person may be subject to imprisonment or
execution; in the other cases a person may be subject only to public
criticism or ridicule. The differences are not a mere "detail" - at least
not by my definition.

Education, experience, and an open and global R&D process seem (IMHO) the
best way to minimize stupid mistakes. But for various reasons that I'll
also be touching on in my coming post on the subject of a nanotech arms
race, I think it would be counter-productive and possibly even dangerous to
attempt to develop laws specifically targeting MNT (and the Foresight
Guidelines also seem to agree on this point).

> Jim wrote:
>> I disagree - construction assemblers don't need to navigate or move.
>> At least, not in any but the most trivial of ways that a nanofactory
>> must also "navigate" or locomote finished products. I happen to
>> believe that the most useful and simplest kind of self-replicating
>> assemblers are those that remain semi-attached to each other. Any
>> mobility they have is only in relation to their peers (i.e. move an
>> assembler in or out of a place in the matrix of assemblers using a
>> bucket-brigade mechanism.) Things like communication are easier,
>> among other advantages.
>
> "One to hold the bulb and six to turn the ladder."
>
> You might be right that this can be efficient enough to be worth
> looking at; but I really doubt it. Especially since a design with
> self-contained replicators would require each replicator to drag along
> a computer.

Each of your nanofactory's "production modules" has one computer
controlling several thousand fabricators - making it not only a single
point of failure for a larger system than in the self-contained
replicators, but also a feedback control choke point that will slow down
operation by a considerable factor (many of the fabricators are likely to
be idle at any given time). So the time and matter needed to build a
computer into each self-contained replicator eventually pays off in
reliability and speed. I see no reason to believe that a table-top device
built of self-contained replicators would be any less energy efficient than
a nanofactory. And since I believe the design challenges and complexities
of both classes of devices are comparable, I think self-contained
replicators are preferable (NOTE: I use "self-contained" in the above to
refer to all aspects except energy and feedstock.)

>> The genome for e-coli has ~4.6 million base pairs; the entire human
>> genome has ~3 billion base pairs [1]. Since there are a lot of
>> genomes that code nothing, the real information encoded is less than
>> those values (I believe only about 3% of human DNA encodes anything
>> useful). A complete nanorobot design shouldn't be more complex than a
>> human. For these reasons I think your gigabyte estimate is at least a
>> couple orders of magnitude too high and I still think my 200k
>> estimate is closer to the mark for a minimum.
>
> Either 1) lots of software and hardware today is more complex than a
> human or 2) engineering leads to very inefficient description
> techniques that we don't know how to compress.

I don't see that either statement holds true. What systems are you thinking
of that take the proverbial ton of information to specify?

> It might be possible
> to build a nanomachine description language that could represent a
> bonding/placement operation with less than 0.0002 byte per atom (for
> intricate nanosystems), but I sure wouldn't put money on it. But I
> admit I can't rule it out.

The key is obviously in what fraction is really "intricate" and what
fraction is repetitions of some "intricate" subsystem. Consider that the
specification for a 100 gigabyte hard disk is not 100 times larger than the
specification for a 1 gigabyte hard disk, but the former can hold 100 times
more data.

Ironically, I think your nanofactory is more "intricate" than a self-
contained replicator since you have to encode for slightly different
interfaces at each level in the hierarchy. Not a lot, but enough to add a
teeny bit more complexity and therefore a potential source of problems that
will slow development.

I'll respond to your nanotech arms race points in a new thread - though it
may take me till the weekend to find time to compose a semi-coherent
starting post.