[extropy-chat] Nanotech educations

[Chris Phoenix]

Adrian Tymes <wingcat at pacbell.net> wrote:
> There are those who want to discuss the technology
> (but don't necessarily have a clue about its
> realities), and then there are those who want to
> practice and develop the technology (but suck at
> beating their own drum).  There's a lot going on that
> you don't see if you don't look for it.

Like what?

I doubt it's a matter of being bad at self-publicity--rather, it's that
no one dare admit to working on "nanobots" or even toward them.

>> The problem with being covert is 
>> that it doesn't let us discuss the implications of
>> the technology.
> 
> Some people don't see this as a problem.  [GMOs]
> ....
> Then again, I do wonder what specific benefits are
> sought from a public discussion of the massively
> unknown.  (Beyond just alerting people that technology
> X exists: in truth, while it's not yet developed, it
> can be argued not to truly exist yet.)  

It's not just the public: the more some academics say "nanobots" are 
impossible, the less anyone can talk about what'll happen when 
nanofactories arrive.

And I do think it's important to plan ahead for such powerful technology.

I dislike suggestions that an elite should make decisions for "the 
public" while keeping them in ignorance.  I recognize the truth of "A 
little learning is a dangerous thing."  And I know my dislike of the 
suggestion doesn't mean the suggestion is wrong.  But I really hate to 
see people being blatantly misinformed (as in Whitesides *still* 
spreading the fingers fallacy on BBC).

"Robert J. Bradbury" <bradbury at aeiveos.com> wrote:
> On Wed, 23 Jun 2004, Chris Phoenix, commenting on my comments wrote:
>> The AIDS problem is a lot more constrained than molecular manufacturing.
> 
> Oh??? Its bad enough that it has a robust system for stealing replication
> resources but it is self-evolving on top of that.  I would hardly consider
> that "constrained" when one is trying to figure out how to stop it...

I didn't say the virus was constrained.  I said that the problem 
(stopping the virus) was constrained.

> I would agree that we could probably have developed simple diamond
> mechanosynthesis by now if a greater amount of attention had been
> focused on it.  I am less confident with respect to the extremely
> varied assembly chemistries that might be required for the parts
> Eric, Ralph & Josh have designed where they make very creative use
> of atoms other than carbon.

I was just talking with Eric about this tonight.  I took the position 
that hydrocarbon-only (Merkle-style) was the way to go.  Eric asserted 
that today's computer resources would be sufficient to design such 
reactions/operations.  Previously, I'd asked Rob Freitas (who's looked 
into whether he thought pure hydrocarbon could implement machines; IIRC 
he said that he hadn't looked into it in detail, but he thought so.

So either way, it should work.

> We shall henceforth call this the Bradbury Ice Cream Sandwich
> analogy for Nanotech Development.  In its short form it can be
> summed up as "we are nibbling around the edges".
> 
> I'm not saying that is good -- I'm just observing that is what
> seems to be taking place.

I'd argue that most of this nibbling work is aimed at eating the 
sandwich rather than making interesting shapes out of it.  It's possible 
that after enough practice eating ice cream creatively, if you'd 
suddenly wanted to impress someone by creating an ice cream sculpture 
with your teeth, you would have found it easier to do.  But there are 
more efficient ways to learn, if you know what the goal is.

> Yes, I would agree that it would be nice to have multiple labs
> focused on real MNT using perhaps each of the 4 or more processes
> we could probably outline to get there (mechanosynthesis, a
> combination of retrosynthesis and organic chemistry, lithography,
> etc.).  But until we have have *real* proof of concept I do not
> see the pressure (either from within the scientific community
> (bottom up?) or from government (top down?)) to make it happen.

I see that as a flaw in the system.

> Could we not have previous examples of this in the history of technology
> development? The one which comes to mind is the time from the development
> of the theories that were the basis of atomic fission to the demonstration
> of the atomic reactor in Chicago to the detonation of the first atomic
> bomb in New Mexico.  But I suspect if we really go back and look at things
> like the airplane or the automobile engine or even the steam engine they
> were relatively long processes.

The A-bomb was theory-limited.  Your other examples were 
technology-limited.  MNT is not technology-limited.  And it's only 
theory-limited because no one has paid attention to it; the theory ghas 
turned out to be surprisingly easy, at least every part I've studied 
(which is a pretty wide sample).  The other limitation is politics. 
Which varies from place to place... which is why I'm so worried about an 
unpleasant surprise one of these years.

>> No... I believe you can start from a different perspective.  Protein
>> engineering is hard, because protein folding is complex.
> 
> Granted -- but three things are combining to mitigate this -- ...
> So protein engineering is on a path towards being much more tractable.

I don't disagree.  But diamondoid engineering was tractable years ago.

>>> Using a different chemistry that maps fabrication to shape more simply would
>>> substantially cut the design cost.
> 
> Granted.  And I've been relatively surprised at the rate at which
> people have been advancing self-assembly and related processes
> towards this.

As I said--the theory turns out to be surprisingly easy as soon as 
anyone looks at it.  I'm currently starting to study (on paper) a 
chemistry that I think is quite promising...

>> I've been told that students are told not even to read the literature in
>> molecular manufacturing.  IIRC this was at Northwestern University.  I
>> strongly suspect it happens elsewhere as well.
> 
> Interesting -- *if* so then you need to launch a strong campaign
> against this -- I can't imagine a class on bioethics (or philosophy
> or history of biology, etc) where one is told to *not* read reasonable
> literature if only to find out why it may be wrong (e.g. Lamarckian
> inheritance).  

Ah, but Nanosystems isn't reasonable literature, it's science fiction 
full of cartoons.

Suppose I could document this... what would the campaign look like?

> Sounds like it is more likely that the profs/admins
> don't want to deal with sticky issues as to where the literature
> is incorrect (which means they would actually have to *read* it).

Yep!  Especially since no one has found a showstopper problem in 
Nanosystems in over a decade...

> I don't think it is *that* huge -- you may be able to get a grant
> from the NSF on the process of rating how the programs are doing.
 >
> Steps I might take are:
> a) Use either the web and/or a call to the NSF to find out what Univs.
>    are being funded (and/or any international centers mentioned in the
>    news).
> b) Determine who the directors are.
> c) If its a Univ. order a course catalog from each school and review
>    it for details about any nanotech programs (or perhaps do this online).
> d) Give a call to the directors and query them about specific courses,
>    course materials, etc.
> 
> Just (a & b) on the web would give people like Devon a significant
> ....Then its a simple task to setup a matrix that people like Devon
> can review.

Yes, this doesn't sound that hard.  But I don't have time, myself. 
Would someone like to do this?  Of course I'd be appropriately cautious 
about putting CRN's reputation behind someone else's grant proposal. 
But we might work something out.

I'll also pass this on to World Care (our umbrella organization, with 
educational mission) and see if they can find someone.  So thanks for 
the idea--it may go somewhere.

Damien Broderick <thespike at satx.rr.com> wrote:
> Interested players such as Chris and Robert might care to review this set 
> of Wired projections from just on 9 years ago:
> http://www.wired.com/wired/archive/3.08/reality_check.html
> and see how they think things are crystallizing.

The responders were obviously talking past each other.  So I'll just 
say: I expect with pretty high confidence that someone will have a 
nanofactory in less than fifteen years, and that the follow-on 
applications will be like Drexler said (just a few years for even the 
tough problems).

Chris

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