Home > The Atkinson-Phoenix Nanotech Debate -- Page 3
The Atkinson-Phoenix Nanotech Debate - Page 3 - Emails
Commentary text by Chris Phoenix, except where otherwise noted. [Bracked Italics] indicates a correction. Italics used on text indicates a comment taken from an earlier email, written by the other person.
Last update: August 04, 2003.
Page 1: Overview
Page 2: Emails through June 27, 2003
Page 3: Emails through July 14
Page 4: Emails through August 4
Page 5: The wrap-up
June 29, from Chris Phoenix to Bill Atkinson
Subject: Re: Feasibility and desirability Re: Review Response
One thing I didn't boggle at was the nanoscale frictionless bearing.
OK, good. But realize that Drexler's frictionless bearings work on a
somewhat different principle from biological bearings. Actually,
Drexler describes at least three types: the single-bond bearing, the
non-interlocking sheathed rod, and the sliding surface with force
correction. Some of these are somewhat equivalent to biological
bearings, some aren't. So five years ago, you could have said that
Drexler's bearings were unproven. But then an experimentalist, not
affiliated with Drexler, found what appeared to be frictionless-bearing
behavior in nested nanotubes, giving preliminary confirmation to
Drexler's sheath designs.
I will grant you that sheer fiddliness, by itself, is not necessarily a barrier to new technology.
Good, again. The question is whether the level of fiddliness can be
chosen, or whether the idea is brittle. There are a few places where
Drexler's proposals require brittle levels of fiddliness. You
identified one of them below: the need for removing all contaminants.
I'll get to that. But in many places, as I think you're starting to
realize, the level of fiddliness can be chosen, and designs can be
simplified until they are viable.
My problem with the MA technology you describe is that it's still completely imaginary: all calc and no grease. Workable technology, indeed any technology worth the name, has a fairly short cycle between drawing board and lathe: between idea and trial.
You have done a good job of demonstrating that molecular assemblers are
not technology at all, but speculation. Now don't blame them for being
poor examples of technology! It's far more pertinent to ask whether the
speculation has value.
Consider this: Not long ago, a group of scientists got together to
study a set of experimental data. They ran lots of calculations. They
made some interesting hypotheses about very small phenomena. But the
hypotheses couldn't be tested! So they went to the government and asked
for billions of dollars to create a research center in Texas, to see
whether their theory was right. The Superconducting Supercollider was
never finished. But no one accused the particle physicists of being
unscientific or of depending too much on theory.
You may find it very instructive to compare your criticism to this one:
A large number of drawings of the machinery are also in existence. It is
supposed that these are complete to the extent of giving an account of
every particular movement essential to the design of the engine; but,
for the most part, they are not working drawings, that is to say, they
are not drawings suited to be sent straight to the pattern or fitting
shop, to be rendered in metal. There are also drawings for the erection
of the engine, and there appears to be a complete set of descriptive
notes of it .... There remains, however, a great deal to be done in the
way of calculating quantities and proportions, and in the preparation of
working drawings, before any work could actually be set in hand, even if
the design be really complete. There is some doubt on this point as the
matter stands, and it certainly would be unsafe to rely upon the design
being really complete, until the working drawings had been got out.
Mechanical engineers are well aware that no complex design can be
trusted without this test, at least.
It was [the inventor's] rule, in designing mechanism, in the first place
to work to his object, in utter disregard of any questions of
complexity. This is a good rule in all devising of methods, whether
analytical, mechanical, or administrative. But it leaves in doubt, until
the design finally leaves the inventor's hands in a finished state,
whether it really represents what is meant to be rendered in metal, or
whether it is simply a provisional solution, to be afterwards
.... If all sorts of heavy work of this kind could be easily and
quickly, as well as certainly, done, by merely selecting or punching a
few Jacquard cards and turning a handle, not only much saving of labour
would result, but much which is now out of human possibility would be
brought within easy reach.
If intelligently directed and saved from wasteful use, such a machine
might mark an era in the history of computation, as decided as the
introduction of logarithms in the seventeenth century did in
trigonometrical and astronomical arithmetic. Care might be required to
guard against misuse ....
Have you guessed it yet? This is criticism written in 1878, explaining
why the Committee didn't think it was a good idea to try to build
Babbage's Analytical Engine at that time. The Analytical Engine was
essentially a modern computer, with an ALU, memory, logic, I/O, and an
instruction stream *including conditional branching*. It was first
described in 1837. The criticism above appears at:
An emulator has been written. In the judgement of the authors, the
Analytical Engine "was mankind's first bold venture into the domain of
Now... does Drexler deserve any harsher criticism than Babbage?
In fact the better the technology, the shorter the oscillatory period. It's a tennis game: imagine/try/iterate. MA work seems to have stalled in the Imagine half-cycle. Thus denying itself any feedback.
MA work has not stalled; the reason the foundational work has not
changed its conclusions is that it has not yet needed to. (But of
course it has been improving; the nanofactory in Nanosystems is better
than the vat of assemblers in Engines, and Merkle's designs are in
several ways even simpler and more practical.)
And it has not been denying itself feedback. In fact, it has been
begging for feedback. Whenever we find a result that's even halfway
relevant, like the sliding buckytubes, we compare our predictions with
it. I think we can be pardoned for patting ourselves on the back when
we find, time after time, that the experiments support Drexler's work
(and by extension Merkle's, Freitas', and mine).
And not all trying has to be physical. Thought experiments have to be
done carefully and treated with skepticism, but can be useful to
criticize a design or increase confidence in it.
it's evolved wilder and wilder designs - absolutely none of which have been put to the test. It was my growing annoyance in realizing this that really led to my complete loss of patience with Eric Drexler. The guy really has dispensed, it seems to me, with the necessary reality checks.
It's evolved more and more detailed projections of what might be
possible. Some of them do get pretty wild, and we know that. But some
of them look like basic engineering, and it's annoying to see them all
tarred with the same brush.
There is a crucial distinction here. It's the difference between
saying, "This design will work"--necessary for any technology--and "This
level of performance is achievable and practical"--which is all that's
necessary for policy discussion. Drexler's work aimed to demonstrate
the basis for the performance and utility claims he was making. It's
not simple to decide whether a criticism of "It won't work exactly that
way" is sufficient, or whether the much more difficult criticism is
necessary: "Nothing like that is likely to work."
You say Drexler has dispensed with reality checks. But there are
reality checks all through Nanosystems. He doesn't merely say that his
computer will make very few errors; he lists all known sources of noise,
adds them up, and figures out how often a rod will jump out of
position. That's a reality check! It doesn't guarantee that the
computer will work, of course. But it shows that if built to spec, it
will probably be sufficiently reliable.
As have you. You tell me your recent work has "solved problems." No: it's only done sums and solved equations. A true solution requires demonstration:
I disagree that a solution requires demonstration. If you don't know
how to find my house, and I give you directions over the phone, the
problem is largely solved before you ever get in the car. Of course,
this depends on the quality of the directions: "Turn left at the big
green tree" is less useful than "go 3.1 miles and turn left on Smith
St." And yes, I might give very precise but wrong directions. And a
bridge might be washed out. But directions are more brittle than
engineering: there's usually a way to make something work, as long as it
doesn't violate physical law.
and that in turn demands that fateful encounter with reality, in the person of the nanocosm. Elsewhere both you and Drexler like to cite the Manhattan Project, with its endless calcs and forecasts. But the Project met the description I set out above, for an actual technology: a high-frequency iteration between theory and practice, paper and nature, screen and reality.
We don't have anything like a Manhattan Project for MNT yet. We do not
cite the MP as a comparison for what we're doing today. Certainly not!
We cite it as an example of how R&D can be greatly accelerated if enough
money and smart people are thrown at a well-defined problem. At the
start of the MP, there was a lot of untested theory. Today, there's a
lot of untested MNT theory. At the start of the MP, they did not know
which of three ways would be used to separate the uranium--and in fact,
one was found to be unworkable and abandoned early on. But the Bomb
became a possibility as soon as the nuclear chain reaction was
discovered in, what, 1939? Pure theory, that.
We might equate Feynman's speech with the discovery of radioactivity,
and Drexler's PhD thesis or Nanosystems with the chain-reaction theory.
An immense amount of work still needs to be done to determine whether a
bomb will require pounds or tons of uranium, and whether the GNP of the
country can pay for the separation. But we think it's possible, and may
well be quite powerful.
Kept isolated, untested and therefore unproven, the greatest calcs in the world are infertile. .... Come back to me when you start your shop work. Till then I'll stay a skeptic.
Skeptics are good. We need skeptics. Please stay a skeptic even after
the shop work is started. But Nanocosm went far beyond skepticism.
You make it sound like we *want* the work to be kept isolated and
untested. Not at all! We're begging for feasibility studies! We would
love to see experimental work on any aspect of Drexler's ideas. But
where will the money come from, with a variety of people going out of
their way to debunk everything associated with the theory? You're
asking us to fight with both hands tied behind our backs.
It would take a very small fraction of the NNI's budget to, for example,
help Freitas and Merkle get their books out faster. It wouldn't take
much to start a peer-reviewed Journal of Molecular Nanotechnology in
which they could publish, rather than having to produce saleable books
for every idea they want to present. And there are a variety of
experiments we could do right now to verify some of the predictions and
assumptions in Nanosystems.
Don't do your best to cut our funding and destroy our credibility, and
simultaneously blame us for not doing experimental work.
No doubt you'll say Yes, but it's a preliminary: we're coming to the shop work. .... But something admitted to be vital, and not yet done, in a way suspends dialogue.
Depends on the purpose of the dialogue. If the purpose is to discuss
whether Utility Fog or Vasculoid are feasible, then yes, there's no way
to answer that yet. But if the purpose is to discuss policy options in
a variety of plausible scenarios... well, that can be done, and should
be done, starting now. We don't need to agree that it will definitely,
absolutely work. Just that it might, and that some big country may find
it plausible enough to try to do it. If we can reach that point, we can
start talking about which scenarios are plausible--and at that point, we
can probably agree even about a lot of the technical issues.
That being said, I have to admit - Gad, this is dragged out of me! - that some of your ideas are conceptually fascinating. I like the way the extreme complexity and intricacy of your world view's design and construction may be - I say, May Be! - offset by the extreme simplicity of its operating methodology.
Yep. You've grasped a major point here.
There seems a parallel between your view here and the main point I was trying to make in my CA discussion: ie. that at the nanoscale, manufacturing success absolutely demands a stripped-down, cut-rate, compressed, streamlined minimality in directive algorithms. Otherwise the whole concept of the MA collapses under the weight of its own complexity.
You are right. But diamondoid design can be compressed quite a bit;
instead of specifying billions of operations or millions of atoms, just
specify a few volumes to fill with diamond lattice. It's not quite that
simple, but you get the idea.
That said, by limiting the design to diamondoid, we are throwing away
large amounts of diversity and functionality. Just like digital
circuits throw away large amounts of information. Diamondoid nanobots
aren't quite "at the bottom" yet: there's still room below them for more
heterogeneous atomic structures. And maybe a paradigm that hasn't been
invented yet will allow us to build such things in a useful way: some
sort of evolution-in-place, maybe.
But there are uses for non-complex machines. Just as an automobile
engine is hundreds of times more powerful than a horse, but far dumber,
a cell-sized diamondoid nanobot will be far dumber than a cell--but far
more powerful in certain very limited ways.
Now the crucial question is, can this idiot-savant paradigm, this
engineering practice writ small, this fiddly limited artificial digital
system--can it close the loop and create duplicates of itself in a
compact system? I think we already agree that this is, in fact, the
crucial question: if the answer is yes, then diamondoid nanobots may be
quite valuable. If no, then of course they are pointless.
But I think you and I can also reach a tentative agreement about whether
such loop-closing is impossible, plausible, or likely to work. We can,
for example, discuss my claim that the control algorithms can be a lot
less stripped-down if they are fed in from outside, with minimal local
computation. Of course, to actually gain efficiency, the controls have
to be applied with massive parallelism--this requires modular design,
but we're already good at that, especially in software.
Another reality-based question. Can you really create a nanoscale clean room, without a single bit of unwanted crud in it, ever? Because a single atom in the wrong place at the wrong time would gum up the works. This may be your tallest technological order. Again, I'll believe when I see. But I grant (aghh! the tooth is pulled) it may be conceptually possible. Nice calc.
This is a brittle point in the design. It breaks down into five
sub-points; you probably know much of this, but I'll explain it for the
1) Can a small cleanroom create a bigger cleanroom? This is quite easy:
just make a collapsed flexible envelope, remove it, and unfold it.
2) Can the mechanochemical apparatus handle a variety of chemicals
without ever letting one fly away from the binding site? This is
largely unknown, but there are quite a few chemistry options, and for
some of them, we already know the strength of the binding method, and
know that it will break only extremely rarely. And if we run into
trouble, reducing the temperature of the whole factory will make a lot
more options work. So this sub-question is not too brittle.
3) Can an assembler or factory take in supplies without contamination?
This depends on sliding seals manufactured to atomic precision. The
answer is probably yes, but we need to see more work on sorting rotors.
4) (optional) Can a factory expel product without contamination? This
depends again on sliding seals that don't even let atoms through.
Again, this appears quite possible. The best theory we have says that
two perfectly flat diamond surfaces pressed together won't even let
5) Can the initial clean room be created for bootstrapping? Note that
it probably doesn't need to be very big. Also note that we can put
getters inside it: bits of very reactive material that will react with
contaminants and bind them. I suspect that filling a chamber with
molten sodium, freezing the walls, and draining the middle by vacuum to
create a cavity, would leave that cavity quite clean. If your apparatus
were rugged and unreactive and was not "wetted" by liquid sodium, you
might be able to mount it in the middle of the chamber, and have it
usable and very, very clean when the sodium retreated. This is probably
a silly idea, but it can probably be improved on.
Another possibility--clunky but workable in theory--is to develop a
technology that can make tight cleanrooms and useful bootstrapping
apparatus but does not require perfectly clean working conditions. It
wouldn't have to be very strong, and could use bigger feedstock--might
not even use covalent joining. Have the semi-cleanroom make another
with a small fraction of the volume. Then unfold it and repeat.
Eventually the cleanroom will contain on average less than one atom of
contaminant. (I think this is proposed in Nanosystems.)
But again: Why bother? The super-strong materials to which you refer are already being achieved by self-assembly, using sputtering (AIST Tsukuba) or other...
Super-strong materials are only one reason. There's also super-compact
computers and motors. If Drexler's designs are even close to right, we
can shrink most of today's machinery by a linear factor of at least a
million: a volume factor of 10^18. Cover the continental U.S., coast to
coast, with six-story buildings full of machinery. Shrink it by a
million times, and it's more or less equivalent to a thin plastic
dropcloth. That's more complexity of machinery than exists in the world
today--and Drexler estimates it could be built in an hour for a dollar.
It's as if
you've designed a nutcracker that works by firing a hammer into orbit, then falls to Earth to smash the walnut. Even if it's possible, why bother? It's unnecessarily complicated. Worse, even from the eng-calc viewpoint, it's inelegant - it doesn't tap into that organizing information that seems evenly dissolved throughout nature. Why compel when you can persuade? The molecular assembler seems nothing more than a steam-propelled gondola, or that orbital nutcracker.
Or a Horseless Carriage, or the Analytical Engine. Let me quote from
the skeptical committee again: these people knew what it was for.
"This necessity of jumping discontinuously from one figure to another is
the fundamental distinction between calculating and numbering machines
on the one hand, and millwork or clockwork on the other. A parallel
distinction is found in pure mathematics, between the theory of numbers
on the one hand, and the doctrine of continuous variation, of which the
Differential Calculus is the type, on the other. A calculating machine
may exist in either case. .... it can either calculate any single
result, or tabulate any consecutive series of results just as well. But
the absence of any speciality of adaptation is one of the leading
features of the design."
I love that last sentence. it sums up exactly what makes all the design
That's assuming, which I do not believe, that it can ever be shown to work. For example, I don't see how you can claim the nanoassembler will have universal application when it will work only on, and with, stiff materials. "Wide variety of parts and properties"? Only if they're stiff!
I don't think I claimed universal application. (The phrase "universal
assembler" was unfortunate, and I hope no one's seriously using it
anymore.) I did say "wide variety of materials and properties." Humans
consist of materials with properties as diverse as hair, muscles, and
corneas. But they're all made of cells. Spring steel is extremely
stiff. But it makes watch springs. A ten-nanometer diamond rod is
certainly stiff enough at the ten-nanometer scale to be manufactured by
mechanochemistry. But if you make it a micron-long beam, and connect
the beams in a truss, the properties of the truss will depend heavily on
how the beams are connected. It could be very stiff, or very elastic.
Yes, there's a limitation here. You can't make squishy nanoscale parts
(unless you can stiffen them during manufacture, with some kind of
nano-jig or nano-anvil or...) But from a human-product point of view,
where millions of nanoparts fit into the volume of a single cell, you
can certainly make the product have a wide range of mechanical
properties, including programmable properties.
We could throw rocks at each other forever on this one.
Actually, I think we're making good progress. I think the only rock I
threw this time was that it's unfair for you to try to destroy our
funding base while complaining that we're not doing more work.
And I think I fielded your rocks pretty well. Of course, I have not
completely answered your objections, and I won't be able to. As long as
we can agree on which issues are unsolved, and whether or not they make
nanobots unlikely or impossible, we will have accomplished something.
If you are a skeptic, you will only have a finite number of rocks. If
you are a debunker, of course, you will dig up more rocks when you run
low, and eventually be reduced to snowballs when the lab results start
coming in. But I think, despite first impressions, that you're not a
debunker. In that case, this conversation is definitely worthwhile.
Once we have classified the objections into rocks, gravel, and dust,
then we can discuss the implications. What are the chances that within
30 years, we will be able to pack the combined complexity of the world's
machines (circa 2000) into a suitcase? And what are the implications if
things actually work out this way? But it's probably a little bit too
early to get into policy discussions: we should wait till you start to
run low on rocks.
That's why I think my later response is going to surprise you. I think this whole issue could give you folks your only workable chance to get the respect that the mainstream scientific communities, and I too, have so far denied you.
I'm looking forward to it! Respectability would be very nice. I think
it was lost when Drexler committed the scientific sin of writing a
popular book about his research and its implications, and has never been
regained. (If he'd never written Engines or founded Foresight
Institute, would you have been nearly so annoyed at Nanosystems?) At
this point, any suggestions would be very welcome.
July 2, from Bill Atkinson to Chris Phoenix
Subject: Re: Dialog page...
Okay, Chris! Here's the reply I've been hinting about, the one that I think
will surprise you.
It occurs to me that at least some of our issues have to do as much with how
content is presented, as with actual content. It's not uncommon for
appearance to meld into reality, even in science - which can be far less
objective than it likes to pretend.
You referred earlier to certain areas of my expertise; I responded that
these especially include marketing communications. So: here's how I'd advise
you folks in the molecular-assembler movement to further your cause, if you
had come to my firm (Draaken Science Communications Inc.) as a client. I
offer this in a friendly and professional way; use whatever you think you
can. If you do act on any of my suggestions, by that degree my objections to
the M-A movement will be attenuated. I admit that often, "free advice is
worth the price"; but chew over what I say - sometimes the most valuable
things you'll ever hear come from your opponents!
I think we both agree that the M-A movement (which I'll henceforth
abbreviate MAM) has a credibility problem in the larger world beyond its own
adherents. Slag me if you will, but I'm not the only person whose wig keeps
flipping at the MAM's claims. For example, see the Wired Magazine article at
http://www.wired.com/news/technology/0,1282,59268,00.html - which reports
how mainstream nanoscience and nanotech are treating Drexler as "the crazy
uncle in the attic."
This situation, including the opposition from observers such as myself who
began their investigations with an open mind, puzzles and frustrates you
folks. Given the extent of your technical education, and the importance of
the subject matter, you can't understand why you "don't get no respect." As
I report in Nanocosm, the Swiss have publicly announced their intent to make
breadbox-sized factories. Moreover, rigorous current research in nanoscience
(as we have already noted) appears to corroborate some MAM predictions made
on the basis of a priori theoretical calculations - e.g. zero-friction
bearings. What's going on?
I see a historical parallel here. A hundred years ago, electricity was the
First World's hot new technology. It had already transformed the city, and
was about to be rolled out into the country to transform the farm. It was
being applied not only to save labor in known areas, lights and motors, but
to do new things hardly imagined before, such as project moving images.
At the same time, and based on these undeniable achievements, things were
being claimed for electricity that were wildly improbable or even outright
wrong. One magazine ad c.1911 shows a muscled male wearing the Supreme
Electric Belt, with the slogan: "I Am A Man Once More!" - code for curing
impotence. Electricity would solve all problems, end all poverty, make wars
obsolete, and vastly extend our life-span.
Sound familiar? So here's my first set of MarCom recommendations for MAM:
(1) Tone down your predictions. At the very least, isolate them in some
special-interest group or think tank, and clearly label them as
speculations. Avoid the present indicative active verb tense in these
statements: Is, Does, etc. - and the future tense as well: Will, Must. Make
it clear when speculation is speculation. Merkle's keynote at San Jose last
year ("Hello, doctor? My heart's stopped") came off like stand-up comedy.
(2) Sever all links with the corpsicle crowd. Perhaps they provide you with
moral and/or financial support; but they also have made, and continue to
make, the entire MAM a laughingstock among most mainstream scientists. While
Dr. Drexler's vision for nanotech extends to such miracles as corpse revival,
and he has to date been unafraid to express this, it seems time for him to
keep his longer-term projections to himself. They do him, and his MAM, no
good: quite the contrary. They make him sound like a preacher on a soapbox
in the park.
(3) Give the MAM a single voice. This might be the Foresight Institute; it
might not. But as long as a babble of voices arises, each purporting to
speak for the MAM, you are not going to be heard clearly and taken
seriously. I hereby nominate Chris Phoenix for World MAM President. He's
sober, eloquent, passionate, informed, historically knowledgeable. Chris,
you could do for the MAM generally what you've been able to do in my case:
turn obdurate, exasperated ridicule into sober second thought. Before I
encountered you, I'd have said that was impossible. The MAM needs your voice
front and center.
(4) Watch the little tricks that tick off mainstream scientists. One the
most irritating things the MAM can do to a working researcher is to post his
or her work on a MAM website, as if to say: "See! We're as legitimate as
these guys!" I ran into this at Nanotech Planet last year. Colleagues of
scientists cited by the IMM, FI, etc. were astounded that their colleagues'
research was being referenced to support aims with which the researcher
would almost certainly disagree. The MAM must get the researcher's
permission for every citation it makes. I know these data are in the public
domain; but the MAM's use of them to date reeks of special pleading. Again,
you do your own cause no good here.
(5) Get beyond the endless bloody calculations: Check in with reality: Do
your own research. Do it with as much rigor as the best science anywhere can
claim. Do what the great researchers do: attempt everything to disprove your
own cherished notions. Don't stop the instant you find a fact or two that
seems to support your views, and claim victory. Test your own research data
to destruction, or try to. Publish only those you aren't able to destruct.
This will require fundraising; but the more genuine R&D you do, the more
mainstream you'll become; the more grants you'll get; the more you'll
attract the respect and interest of mainstream science. Don't be afraid to
test your dreams.
(6) I say it again: the MAM must modify its claims. Many of these now border
on the outrageous. Think what you like; get together in SIGs and brainstorm
a future as wild as you want to. But the face you put toward the world must
be far more sober, even dull. As Tom Theis of IBM notes, almost no
innovation succeeds unless it restricts its attempted influence to one
business level up or down. Make a new and better car tire, and you may earn
a billion dollars. Announce that all wheeled vehicles are irrelevant because
you're going to come up with teleportation, and you'll hit the wall - and
get ridiculed at the same time.
(7) Tying in with the above: greatly shorten the feedback interval between
experiment and prediction. Saying "give us ten years" - or twenty, or a
hundred - is like saying "Keep the faith: God is on our side." Whether it's
true or not is, in practical terms, irrelevant. Such an approach can never
work. It's this insistence on long-term belief, in the absence of rigorously
demonstrated incremental progress in accumulated facts, that gets the MAM
laughed out of court most loudly. A single fact, viz. the recent
demonstration of the frictionless nanobearing, will (if properly used) get
the MAM more credibility than all the cries of "immortality-is-coming." One
sees this all the time in the business world. The most highly regarded
companies make careful forecasts, then meet or exceed them.
July 2, from Chris Phoenix to Bill Atkinson
Subject: Re: Dialog page...
Thanks for this reply. I'm not surprised, but I'm grateful that you're
trying to help.
Should we include this thread in the published dialog? I'm not sure
that it'll advance the discussion of the technical stuff, and it'll
contain a higher percentage of dirty laundry than the rest of the
discussion. I'm inclined to leave it out.
Many of your recommendations don't surprise me. They all seem to have
merit for geting MNT more accepted scientifically. Of course, some of
them are difficult, and two of them (4 and 7) appear to conflict.
Your recommendations also don't seem to take into account one crucial
factor: Much of the reason for talking about MNT at all is to talk about
its policy implications, and that can't be done without talking about
what an assembler/nanofactory would make possible. And that requires
talking at least ten years out, and talking not just about basic
research but about plausible extrapolations of that research. As we
talk about this, please keep in mind that the primary goal of many of us
is preparing for a potentially dangerous technology--though being seen
as scientifically plausible is important for several reasons.
It may be that it's too early to talk about what nanobots can do, before
we've shown that they're likely to work as claimed. But at this point,
enough work has been done to justify the concern--it's "might" or
(sometimes) "probably", not "will", but some of the concerns are
definitely worth talking about even in the "might" stage. So I think
that here too, the problem is one of communication. The question is, is
it possible to both improve the respectibility of MNT's science, and
push the activism, at the same time? Or is there some fundamental law
of communication that says this can't work? It clearly didn't work for
Drexler to tackle both simultaneously, but how much of that was
inevitable and how much was simply his particular methods?
There are at least three problems with trying to cozy up to the
scientists. First, the MAM has already pissed them off and made them
uncomfortable. Recently Drexler has tried (starting with the open
letter to Smalley) to force down the throats of several scientists the
fact that they're saying false things about MNT. (As I write this, he's
just sent a second open letter to Smalley.) Before that, apparently
citing scientists' work pissed them off (I'll get back to this one).
And of course the wilder claims and associations have made them want to
distance themselves from the whole movement. (I'm reminded of the early
history of the railroad, in which an engineer was ridiculed in front of
Congress simply by getting him to admit that trains might in the future
go as fast as 40 MPH.)
Second, "Science advances one funeral at a time." At this point, it
could take decades for scientists to change their tune. From a policy
point of view, we don't have decades.
Third, scientists working on non-MNT nanotech will be protective of
Given all of the above, it's really tempting to think that the overall
best strategy is to cut our losses and bypass the scientists. Of course
that's probably what got Drexler in trouble in the first place. Mind
you, he didn't bypass the science--just the scientists. But he was
probably in trouble the moment he published Engines. Anyway, I'm a bit
skeptical of the idea that the MAM can rehabilitate itself quickly by
following your advice. But that doesn't mean we shouldn't try.
Now, to your specific recommendations:
1) Tone down predictions. Trouble is that we're speaking in many
different contexts. Merkle was preaching to the converted--I guarantee
that to many in the audience, what he said was pretty much plausible as
stated. Futuristic of course. But nothing prevents it from working.
That said, yes, we should perhaps be more careful to speak to the lowest
common denominator when talking to mixed audiences. More about this in
2) Sever links with cryonics. This one is hard. Merkle is a Director
of Alcor! Basically, we'd have to sever links with Merkle--I don't
think that would be fair or appropriate. And beyond that, a lot of
nanotech people (including me) think that "curing severe frostbite" is
more accurate than "reviving corpses" (unless defibrillators also revive
corpses)--many of them are signed up, and IIRC Drexler recently signed
up as well (but don't quote me on this). We can't get everyone to stay
quiet about it.
If someone is signed up for Alcor, but doesn't make a big deal of it,
does that condemn them? What if they give a talk at an Alcor
conference? Does it matter if they structure it around science rather
than hype? And do you think cryonics will get any more respectable when
they manage to revive a frozen mammal?
I'm trying to imagine convincing every major person associated with MNT
to sever their association with cryonics. Won't happen. So I'm looking
for the minimum necessary. Is recantation necessary, or just a tasteful
silence? And what counts as silence?
Which leads us to 6) Modify and tone down claims. Yes, there's a
difference between brainstorming and making claims. And yes, we need to
keep that in mind. So... there's a continuum from physics application
(frictionless bearings) to extrapolation of capabilities (Nanosystems)
to preliminary design of possible products (Respirocytes, Utility Fog)
to claims of radical change (Engines) to good science fiction (David
Marusek) to bad science fiction (Kevin J. Anderson). Where do we need
to draw the line? You said something about labeling all speculation as
speculation. To some extent, that can be done. But not completely, not
with so many people doing so many kinds of writing.
What do you think of Robert Freitas' work? Respirocytes, Vasculoid,
Nanomedicine... all far-future stuff, all dependent on diamondoid
nanobots, but carefully worked out and conservatively engineered. He
gets his papers published in mainstream medical journals. For example,
here's the abstract for his Respirocytes paper, published in Artificial
Cells, Blood Substitutes, and Immobil. Biotech.
Molecular manufacturing promises precise control of matter at the atomic
and molecular level, allowing the construction of micron-scale machines
comprised of nanometer-scale components. Medical nanomachines will be
among the earliest applications. The artificial red blood cell or
"respirocyte" proposed here is a bloodborne spherical 1-micron
diamondoid 1000-atm pressure vessel with active pumping powered by
endogenous serum glucose, able to deliver 236 times more oxygen to the
tissues per unit volume than natural red cells and to manage carbonic
acidity. An onboard nanocomputer and numerous chemical and pressure
sensors enable complex device behaviors remotely reprogrammable by the
physician via externally applied acoustic signals. Primary applications
will include transfusable blood substitution; partial treatment for
anemia, perinatal/neonatal and lung disorders; enhancement of
cardiovascular/neurovascular procedures, tumor therapies and
diagnostics; prevention of asphyxia; artificial breathing; and a variety
of sports, veterinary, battlefield and other uses.
He uses all the wrong tenses, going straight from "Molecular
manufacturing promises..." to "Medical nanomachines will be..." to
""respirocyte" proposed here is..." But I hear that he's respected even
among people who disapprove of Drexler. What's he doing differently?
Maybe it's what you said: "But the face you put toward the world must
be far more sober, even dull." I think Freitas works very hard to
present just that kind of face. Proving once more that the medium is
I hadn't realized that sober and dull were actual virtues in the
scientific world. Rigorous, logical, well-researched--yes, of course.
But sober and dull? Yet now that you say it, I can see the truth of
it. But then how can one be eloquent, passionate, and dull at the same
3) Give MAM a single voice. The phrase "herding cats" comes to mind.
This may simply be impractical. I agree this is a problem. In fact, on
the science side, there's a variety of opinions even among people who
think nanobot/mechanochemical science is sound, about how to bootstrap
them, what they should be made of, and what their capabilities will be.
That ought to be OK--it shows that we're not religiously tied to a
particular idea. On the futurist/policy side, we have a credibility
gap, caused in part by a variety of opinions (some of them weird), and
in part by a failure of explanation.
I think the failure of explanation may be key here. There's really
nothing that ties together all the work that's been done into one
comprehensible body of understanding. Many, perhaps most, of the
nanobot opponents still think that shallow criticism is sufficient. We
(in general, not you and me) would advance the dialog immeasurably if we
could just start from an agreement that there is no simple way to
disprove nanobots. But judging by the simplistic arguments I see
everywhere, we're still a ways off from that. And those who try a
simple attack, and discover it's not that easy, generally only become
quiet--so we have little evidence to show that the attacks have failed.
Meanwhile, new people are delivering the same attacks, and we keep
giving the same arguments. And our arguments are taken out of context
and used as the basis for further attacks. It'd be nice if the
anti-nanobot side would start saying, "Hey, we have to be more
sophisticated here--these simple attacks aren't valid anymore." But
invalid or not, the attacks succeed in maintaining a level of suspicion
and distrust, and taking a lot of our time to answer them--which is
exactly what many anti-nanobot people want, and validity be damned.
Sorry--it gets frustrating after a while. Anyway, do you have any idea
how to shift the grounds of the debate so that simple pseudo-physics
attacks are generally recognized as invalid?
Also, thanks for your very flattering nomination of me as World MAM
President. I guess lack of sober second thought is exactly what I was
complaining about a couple of paragraphs ago. I hope you're right that
I could generate sober second thought in general, but I know my
limitations. There's a huge difference between conversation and
speechmaking. I'm an excellent tutor, but I'm not sure I'd be a good
teacher. And I know I'm no good as a demagogue, or arguing against
demagogues. I also suspect that you're quite unusual in your
willingness to change your opinions and admit you were wrong or
Speaking of mediums and messages and contexts... You said, "As Tom Theis
of IBM notes, almost no innovation succeeds unless it restricts its
attempted influence to one business level up or down." But MNT is not
an innovation. It's a paradigm that will, if all goes well, spawn
zillions of innovations. Compare it to polymer chemistry. It can be
viewed as an innovative way to do chemistry. Or a source of new
materials. Or an integral part of almost every product today. Or a
consumer of crude oil. Or a way of extending engineering capability.
Back when nylon was discovered, someone could have said, "Wow, imagine
all the things we can make with this! This will be huge! Industry
should pay attention to this, and figure out how to develop it and
invest in it!" How this would be received would depend on who was
saying it, and how, and to whom. I guess I'm drifting off topic. What
I started to say is that plastic's influence is not limited to "one
business level up or down." Nanotech won't be either. But I recognize
that this has to be said carefully--much more carefully than we've been
saying it--with attention to style, tone, speaker, audience, etc. But
at this point, we have many audiences listening to everything we say--we
can't tune it to everyone. And many of them are actively looking for
statements to use against us.
It was hard for Drexler to get it right, developing this amazing new
paradigm, and trying to talk about it as a scientist, activist,
educator, and maybe even booster all at once. It's easy to see how he
could set people's teeth on edge. But it's also hard to see how a
single-voice MAM can work. We have battles to fight on at least four
(4) Don't cite other people's research. This one surprises me. And I
suspect that the problem is more with the delivery than with the mere
fact of citing the research. For example, I can say, "The recent
pick-and-place of covalent silicon with AFM demonstrates that Smalley's
criticisms are too broad." Can't I? But it's a little more
questionable to say, "The recent .... demonstrates could be considered
an example of a mechanochemical process." And definitely bad to say,
"The recent .... shows that mechanochemistry works." But this
distinction is subtle, and probably a lot of MNT people don't get it.
The question is, if we do it right, will the scientists get it, and not
be annoyed? Or is the situation so bad that any MNT person making any
mention of non-MNT research is likely to be annoying?
" .... as if to say: "See! We're as legitimate as these guys!" " I
really don't think that was the intention. If that's how it comes
across, then we should probably be changing how we present it. Unless,
as I said, any mention whatsoever is likely to be annoying--in which
case, as always, the temptation to say "The hell with them" is very
strong. (But I do recognize that it's pointless to work for scientific
legitimacy while pissing off scientists.)
(5) Get beyond the endless bloody calculations: Check in with reality: Do your own research. Do it with as much rigor as the best science anywhere can claim. Do what the great researchers do: attempt everything to disprove your own cherished notions.
In many respectable fields, endless bloody calculations do count as
research. Everything from astrophysics to particle physics to
computational chemistry to complexity theory... and the best MNT workers
do their damnedest to make sure that they're applying the equations and
constants correctly, within the limits of the research that the
equations and constants were based on. Everything in Nanosystems and
Nanomedicine is tied to physical research. And most of the ties are not
brittle: if a constant is wrong by a factor of two, or maybe even ten,
the concept can still be made to work.
IMHO, saying that we are out of touch with reality because we haven't
done enough experiments is a blanket dismissal, not a serious
criticism. How can we get across the idea that yes, we really do know
what we're doing; we haven't simply slapped together equations that look
nice and support our views, but every part of our work is defensible and
just waiting for replication and criticism?
Don't stop the instant you find a fact or two that seems to support your views, and claim victory.
Is that how it looks? Then we have a PR problem, not a science
problem. We have not stopped. We have not claimed victory. The fact
or two is not to prove we're right, but to prove we're plausible.
Remember, a lot of people still think our claims are ridiculously
implausible. If we can show that buckytubes are frictionless and
silicon can be moved atom by atom, then at least some of our ideas are
There are two purposes to harping on these "fact or two". One is to
address specific criticisms of impossibility. This, as I said above,
ought to be valid. The other is to beg for serious consideration of the
rest of the material--something we have been consistently denied. This
latter is not good logic, but we sometimes get pretty frustrated at the
lack of worthy criticism. I think many scientists, to some degree, do
treat our work as plasticine time machines--despite the fact that we're
not building time machines, and not using plasticine. Being shut out of
the scientific process altogether--denied the criticism and/or
replication that is necessary for scientific testing of ideas--perhaps
we are too eager to use these scraps of validation. But note this:
we're not trying to use them to prove we're right. We're hoping that if
enough of our extrapolations turn out to be on target, people will start
taking us seriously. So we're not trying to claim victory--just to
establish ourselves as being in the running.
Test your own research data to destruction, or try to.
We can't prove a negative. As long as people want to believe nanobots
can't work, they will always be able to find grounds for criticism. And
the criticism we get now is at a level far more superficial than the
work we've already done! What's our incentive for doing more work? We
have already been denied participation in the scientific process.
Publish only those you aren't able to destruct. This will require fundraising; but the more genuine R&D you do, the more mainstream you'll become; the more grants you'll get; the more you'll attract the respect and interest of mainstream science. Don't be afraid to test your dreams.
This process would take too long for the activism side. By the time we
reestablished validity, recruited lab researchers, got funding, did
research, published results, and repeated this cycle a few times, it
would be 2015 if not 2030. Meanwhile, some country pulls a nanobot
If it seems like we're dreaming, or afraid to test our predictions, then
again we have a PR problem.
7) "Tying in with the above: greatly shorten the feedback interval
between experiment and prediction." So, make predictions that can be
demonstrated quickly, then demonstrate them.
Well, the predictions have already been made. We can't very well
retract them. (And again, we need to use them for the activism.)
I think that even without experimental verification, the work would hold
up pretty well under the scientific process. I think you're asking us
to bootstrap a whole field ourselves. That's too big a job. Science
can't be done in a vacuum. We've already done a hell of a lot of work
that is ready for review--has been ready for a decade. How much do we
have to do? I just don't believe that lack of experimental verification
is the root of the problem. People who want to dismiss us would find it
just as easy no matter how much experimental work we had done.
You said, "A single fact, viz. the recent demonstration of the
frictionless nanobearing, will (if properly used) get the MAM more
credibility than all the cries of "immortality-is-coming." "
Given what you said in 4) about how it's annoying to researchers to be
cited in connection with MA's, how can we use these facts? We've been
trying--but that just gets scientists annoyed at us.
At some point, we have to make the leap from science to policy. Going
back to the continuum from physics to bad sci-fi, policy has to be
extrapolated at least from device capabilities, with some reference to
possible products to make the discussion concrete and comprehensible.
Science ought to reach at least to the applied-physics level, and to
some extent, the device capabilities as well. But there's all sorts of
distinctions that must be made and carefully preserved, or we'll be seen
as saying "Science predicts Utility Fog". Hmmm... I wonder if that's
why the scientists are so uptight at us? I never thought of that, but
it makes sense.
So there you have it. We think we've been doing science, but the
scientists won't play with us. We think our policy work is reasonable
too, but that may have made the scientists more uncomfortable. We have
been trying to do too much, and a lot of confusions have arisen, and we
have not been good at clearing them up. Ironically, by making people
uncomfortable, we have attracted much unfounded criticism of our
science--when the problem is in our PR. And at this point, however we
attempt to fix it, we will run into egos and personalities on both
You have given advice on how to play by the rules (of the scientific
establishment) and become mainstream researchers. I'd like to suggest
two other goals, and see if either of them is approachable.
First, get it recognized that the basic diamondoid mechanochemistry
nanobot theory is not flaky, brittle, or fraudulent--that all the
obvious criticisms have been answered, and it will take serious work to
challenge it. If this can be accomplished, it'll lead straight to the
sober second thought.
Second, make clear the distinction between the science, the activism,
and the wild-eyed extrapolation. None of these things is going to go
away, but they must be separated. How can they be separated when some
prominent individuals engage in two or three of them?
Keep in mind that we are speaking to several different audiences:
scientists, science writers, journalists, the government, fringe groups
(even if we stop speaking to them, they'll keep listening). How can we
direct and sculpt our messages so that one group doesn't blame us for
what we say to another? Even within one group, as I said, answers to
one question will be taken out of context:
Critic: "Machines can't self-replicate."
Drexler: "Biology is an existence proof of self-replication."
Critic: "So you're claiming you're just like biology--ridiculous!"
Drexler: "No, we're actually quite different from biology."
Critic: "So you reject biology."
It often feels exactly like debating creationists--every time you start
to prove a point they change the subject, and they have an endless
supply of irrelevant and shallow criticisms.
So, are we hopeless? Or can we rehabilitate ourselves in time to do
July 3, from Bill Atkinson to Chris Phoenix
Subject: Re: Dialog page...
Yes, this is the dirtiest of laundry; and for that very reason we must run it. Go
into your response if you must, and take away some of its edge: but leave it in,
its essence and its emotion - even its despair. Because this is also paydirt.
Shrinks live for these moments of revelation.
You and I have danced around each other, feinted some, and both landed some blows.
Now here it is: the core, the nub, the essence. We mustn't omit it. This is what
we've been leading up to. It's the breakthrough! No, it doesn't "advance the
discussion of the technical stuff" - at least not directly. But it does, I think,
represent the biggest present chance of advancing the MAM as a whole: of making it
a true discipline rather than an interesting intellectual movement, like Moral
So what happened in our last two exchanges? It was like cutting a diamond: a tap
in the right place, and voila. I didn't abate any of my criticisms of the MAM. But
I rephrased them so that they were coming from a professional communications
advisor. I left off the razor edge of real opposition and delivered my assessment
of why MAM is, to mainstream science, beyond the pale. I asked and then answered
the question, which you hadn't asked explicitly but which lay (and still lies)
latent in everything the MAM says and does: "Why doesn't the scientific
establishment take us seriously?"
Now consider your response. You almost got it; then you shied away. You construed
my comments as more attacks, to be defended against. And you mount a defence
that's up to your usual brilliance. But work with me here; understand that in that
last communication, I was actually on your side. Say you had a disease, and I was
a doctor giving you my diagnosis: telling you how I saw your problem, suggesting
how you could solve it. Would your response be: No, you're wrong, it's not like
that? Or would it be: Jeez, I suspected that's how it was? Possibly both: some
denial, some willingness to listen. And that's what you did.
I was, and am, trying to help here. To see and say clearly what MAM must do to
take the first step toward reconciliation with the establishment. Sure, by your
lights I'm still an outsider, and being unfair and unduly harsh in at least some
of my assessments - maybe most. But you ARE outside the Establishment, and I gave
you a pretty good synopsis of why. It may not be fair, but those are the
conceptions and preconceptions you folks are going to have to address if you want
to come back into the scientific fold.
Because you do have to come back. You can still be mavericks and dreamers; God
knows the mainstream needs those, and always will. Dick Feynman, Bucky Fuller,
Karl Proteus Steinmetz. Hell, Albert Einstein! But it needs rebels WITHIN the
system, not outside throwing rocks and huddling in a cave, telling one another how
put-upon they are. Perhaps you are put-upon by a smug Establishment. That's not
the issue. The issue is, as you yourself see clearly, perception. And the passive
kind: not how you see, but how you are seen. And brother, that is bad.
To specifics. Maybe it's some kind of intellectual compromise for the MAM
officially to surrender its love affair with the corpsicle crowd. ("Cryonics", by
the way, is a legitimate science, and deals with the physics of the supercold:
quantum fluids, low-temp superconductivity, &c. Using the term as a euphemism for
the corpsicle cult is an attempt to blur reality with Newspeak.)
And maybe it's a compromise for you to rein in your predictions to the point where
the published ones are testable within five years max. But sure as there are
little green apples, unless the MAM makes up its mind to compromise to some
degree, you're headed for the dustbin of history along with the people who plugged
the Supreme Electric Belt. The purer you are, the more intent you are to follow
the quirks and whims of Drexler, Merkle and the boys, the more you'll wander in
the wilderness. Upon which 2050 won't see the MA, but may see a Ph.D. dissertation
or two on an interesting start-of-century cult.
But I have to consider, and you must too, whether exile is exactly what you want.
Whether, like the splinter churches, you define yourselves by what you're not.
Whether the MAM's entire raison d'etre is to be the City on the Hill, possessed of
arcane and secret knowledge and content to let the rich and powerful sinners go to
hell at their own speed. It's exactly this attitude - equal parts
self-righteousness, arrogance, shyness, and the fear that actual experiment will
disprove the truths you now comfortably take on faith - that characterizes the
MAM. Unless that changes, the MAM will remain in splendid isolation. You really
think you can achieve the MA in attic labs, unbeholden to the Establishment? We'll
see a backyard moon rocket before that happens.
Yet it's all unnecessary. It's not too late to turn things around! Yes, you may
split the Movement: in fact you will. You must. Let the nutbars stay in their
hilltop fortress. You, and thinking moderates like you, can make a sufficiently
good case to induce a critical mass of liberal- minded mainstreamers to rethink
their opposition. It just requires speaking their language.
I'm not advising you to sell your souls here. Keep thinking outside the box;
without free-thinkers, science quickly becomes a herd of highly qualified sheep,
replicating one another's experiments and marking time till each is promoted from
the bench (yuk!) to administration, where the real power and money are. Mainstream
science needs people like you to keep it honest.
Compromise is the only way you'll get the acceptance, and thus the grant money, to
pursue your dreams in reality rather than on the blackboard. It frustrates you
that the US Government won't make policy based on the great potential of MA. But
as Charlie Schultz once wrote, great potentials are a dime a dozen. No legislator
in his right mind is going to buy a dream when it takes money from material
reality: ie. nanotech in his bailiwick, done by labs making real experiments and
discovering real things.
Policy implications? Do a thought experiment with me. My name is K. Eric Atkinson
and I have done an immense amount of mathematics demonstrating categorically that
humans can transmute matter by mental energy. I have a large following; I have
sold tens of thousands of books describing exactly how this can and will be done;
I know of no show-stoppers, though I am constantly attacked by the ill-informed.
So let's make policy here, Mr and Ms Representative! Think of the transforming
implications of this revolutionary new science paradigm! If we don't do it, the
Ain't gonna happen, is it? Realistically, now? Not without material demonstration:
not without advancing prediction, supposition and speculation to the level of
productive experiment. Yes yes yes, science needs calculations; I've never denied
that. But while necessary, quants are not in themselves sufficient. I've talked
myself blue in the face on this one, and I'll go on doing so till I get my point
across. The MAM must, repeat must, come away from being a strictly blackboard
activity. For as long as it remains so, it will also be a fringe activity.
You don't have science yet: you have the glimmerings of science. You need data.
Perhaps more importantly, you need to demonstrate to the world that you have data:
or are at least in search of data. Only then do you have a hope of being regarded
You cite engineers who were derided for forecasting ridiculous speeds of 40 mph.
But you make a false syllogism here: Great scientists were derided, the MAM is
derided, thus the MAM is great science. Nope: logically possible, but not
logically certain. Goofballs are derided too, and justly. And They Laughed At Me
For Saying The Moon is Made Of Green Cheese.
It is not possible to push nanobot activism and gain respectability. Or it is, but
only if the activism is hidden: if the face you present to the world is, well,
dull. Keep the flame in your hearts, but keep it off the public talks and NIH
grant applications! Drexler wasn't in trouble the moment he published Engines. But
he was in endless trouble as time went on and he kept embroidering Engines rather
than insisting on the performance of corroborating experiments.
Heck, I know what makes the man tick. He has the rare strength of mind to keep
before him day to day the fact of his own mortality, and to try (as no one since
Gilgamesh has done) to oppose his will and his mind to it. That, I think, is his
ultimate motivation. But I fear it's slowly turned his mind to mush. The MA IS
possible! Physical immortality WILL BE the result!!! Yah. Show me. Even Freitas,
detail man though he is, is still just doing calcs and spinning tales till a real
invention is demonstrated. Enough chat, already.
I will say it till it sinks in, Chris: calcs by themselves are not enough, no
matter how sophistimacated they are. In law, one is innocent until proven guilty.
In nanotech - again, this may not be fair, but it's reality - the MA is not
possible until it's been demonstrated. This is, I grant you, 25% science, 25%
technology, and 50% [+ delta-P] perception. Tom Theis's pronouncement about
business levels need not apply to something transforming; but that's true only in
retrospect. At the time of transformation, even the most revolutionary technology
must come across (for sordid marketing reasons) as an agent of only incremental
change. Iron? Just bronze with an attitude. Petroleum? Merely a hay-less horse.
Logically, the skeptics don't have to disprove your arguments, or show errors in
your math. They can say Deus Non Vult and it's a valid argument. They can say they
were asleep when you made your summa, or watching Legally Blonde II, and it's a
valid argument. The only thing that will convince them, whatever their arguments,
is to produce the technologies that their arguments deny. That is the one
unanswerable thing that the MAM can do. To do this, it needs lab facilities and
money. And to do THAT, it needs to abate the wildness of its claims to access
federal funding. That, old son, is the practical essence of my MarCom advice.
More later - gotta sleep
July 14, from Chris Phoenix to Bill Atkinson
Subject: Re: Dialog page...
You read my response as defense against your attack. It wasn't. I knew
that you were not attacking but trying to help. And I was trying, not
to defend, but to show real-life problems that we have to figure out how
to work around. Please re-read my comments in that light. I was
looking ahead, defining current problems so we could solve them--not
defending what we had done.
You and I have danced around each other, feinted some, and both landed some blows. Now here it is: the core, the nub, the essence. We mustn't omit it.
I'm including it unedited (so far), including this letter. The whole
thing should be published on Rocky's site today or tomorrow.
This is one of the cores, essences, nubs. I agree that this is a
crucial topic. It's not the only one, not when people are still getting
published for saying that thermal noise means nanobots can't work.
(That's not a personal dig--I was thinking of an article in Wired, but
having written it, I'll leave it in.)
I didn't abate any of my criticisms of the MAM. But I rephrased them so that they were coming from a professional communications advisor. I left off the razor edge of real opposition and delivered my assessment of why MAM is, to mainstream science, beyond the pale.
And that made all the difference. If you as a prizewinning science
writer say that our science is based on impossibilities, then obviously
we have to defend. But if you say that our presentation needs
work--there's nothing to defend, only to improve.
I was, and am, trying to help here. To see and say clearly what MAM must do to take the first step toward reconciliation with the establishment. Sure, by your lights I'm still an outsider, and being unfair and unduly harsh in at least some of my assessments - maybe most.
Believe me, I appreciate your attempts to help. And I think there's a
good chance that they will be productive. And don't worry about being
unfair and harsh--on anything but the science. There, you'll find a
very tough opponent. Everywhere else, you'll find a willing student or
Because you do have to come back. You can still be mavericks and dreamers; God knows the mainstream needs those, and always will. Dick Feynman, Bucky Fuller, Karl Proteus Steinmetz. Hell, Albert Einstein! But it needs rebels WITHIN the system,
Yep. How's this for timing: the night before you sent this, I was at a
friend's house listening to her teen-ager tell about challenging the
teacher in class, and then complain about getting a bad grade on an
assignment where he didn't follow the rules. At that moment,
enlightenment struck, and I told him, "No, it wasn't unfair. You can
either challenge the system or break the rules--but not both." And then
I read this.
To specifics. Maybe it's some kind of intellectual compromise for the MAM officially to surrender its love affair with the corpsicle crowd. ("Cryonics", by the way, is a legitimate science, and deals with the physics of the supercold: quantum fluids, low-temp superconductivity, &c. Using the term as a euphemism for the corpsicle cult is an attempt to blur reality with Newspeak.)
You're thinking of "cryogenics" for the science. "Cryonics" refers only
to freezing people. My Webster's Ninth New Collegiate gives a first-use
date of 1967 for the latter word. There's no Newspeak, and I don't
think there's any attempt to confuse.
It's not love affair, and it's not about intellectual compromise. It's
that many of the people who have studied MNT have gone and signed up for
cryonics, and some have become active in the field. (It's a field, not
just a cult--and yes, they do actual lab research and publish
interesting and useful results.)
And maybe it's a compromise for you to rein in your predictions to the point where the published ones are testable within five years max. But sure as there are little green apples, unless the MAM makes up its mind to compromise to some degree, you're headed for the dustbin of history along with the people who plugged the Supreme Electric Belt.
Do you have any knowledge of how we're perceived in other countries? A
few years ago, I forget exactly when, I heard a speaker at a Foresight
conference talking about MNT in Japan. She said that they had really
bought into it, and once another breakthrough or two was achieved, they
were going to launch a crash program. I wonder if that's still true, or
if it ever was. It wouldn't surprise me much.
Certainly, if we want to get standard science funding in the US, we have
to change our ways. Or more precisely, it sounds like at least some of
us have to, and apparently they have to disavow the rest. But there are
other sources of funding, some of them likely to be more efficient and
more forward-looking and policy-oriented. So I haven't given up
completely on the idea of blowing off the U.S. scientific
establishment. And in any case, I'm less concerned about funding than I
am about a simple acknowledgement that the ideas are worthwhile. Of
course, getting funded is a good way, probably the best way, to
The purer you are, the more intent you are to follow the quirks and whims of Drexler, Merkle and the boys, the more you'll wander in the wilderness.
You seem to be setting me up personally as a mainstream type MNT
spokesperson. Surely you weren't seriously suggesting that I should
make myself president of a worldwide MNT group? I'm not well suited,
and I'd almost certainly fail. And politics and funding are not my
goals in life. I'm much better off as Director of Research for a MNT
policy research and education group--which is where I am already.
But your advice will not be wasted. Any that seems even potentially
workable, I will pass on to others who can make better use of it. I do
have a lot of contacts among "Drexler, Merkle, and the boys", and many
of them are not as incorrigible as they might seem.
Upon which 2050 won't see the MA, but may see a Ph.D. dissertation or two on an interesting start-of-century cult.
If MA's aren't worked on till 2040, by 2060 they'll be science fair
projects. Just like we've written Babbage Machine simulators. The
dissertations will be on how a promising technology was abandoned for a
long time. (But I'm just saying that to annoy you.)
But I have to consider, and you must too, whether exile is exactly what you want. .... It's exactly this attitude - equal parts self-righteousness, arrogance, shyness, ...
This sounds like Objectivism, the Ayn Rand fringe of
economics/politics (not the same as libertarian--even farther in that
general direction). And yes, you'll find some Objectivists among the
Extropians, cryonics people, and Foresight people. The worldviews are
uncomfortably close. But here and in the next few paragraphs, I get the
feeling that you're talking about the fringe of the members more than
the leaders and researchers. To some extent, the members set the tone
of the groups, and this may be unfortunate. (Note that CRN is not a
membership organization.) But don't judge the motivations of the
leaders by the actions of the followers. Foresight, the group I know,
is quite loose and unstructured. Anyone can be a member, which means
our members come from all over. Those who are shy (nerdy/technophile),
arrogant (attracted to powerful tech), and self-righteous (attracted to
the world-changing meme) will of course be attracted. But they
self-assemble--they are not led. There is no cult here, and they are
showing what they bring, not what they reflect.
... and the fear that actual experiment will disprove the truths you now comfortably take on faith
Your other characterizations ring true to some extent. This one
doesn't. AFAIK, most people in Foresight are quite confident about what
experiment will show. We have no reason to fear any experiment. In
fact, we want to see experiments--because experiments mean progress, and
we know this.
that characterizes the MAM. Unless that changes, the MAM will remain in splendid isolation. You really think you can achieve the MA in attic labs, unbeholden to the Establishment? We'll see a backyard moon rocket before that happens.
No, of course we don't think that. We're hoping that some large facet
of the Establishment, with a lot of resources, will recognize the
potential and fund it. We are not separatist--just unpopular, and
sometimes, yes, arrogant.
Yet it's all unnecessary. It's not too late to turn things around! Yes, you may split the Movement: in fact you will. You must.
How does this fit with your advice that the Movement must speak with one
voice? Will the journalists and scientists flock to the first
reasonable voice, so that the fringe will be ignored? Or will it simply
increase the cacophany?
Let the nutbars stay in their hilltop fortress. You, and thinking moderates like you, can make a sufficiently good case to induce a critical mass of liberal- minded mainstreamers to rethink their opposition. It just requires speaking their language.
Moderate? You think I'm a moderate? I guess there must be something to
this style-makes-the-difference stuff. I remain agnostic about some of
the stuff in Engines, but I'm pretty firmly convinced that the stuff in
Nanosystems will work more or less as claimed. Obviously, I'm open to
disproof and discussion. If that makes me a moderate, so be it. But
then a lot of the people you call "nutbars" are also moderates!
Compromise is the only way you'll get the acceptance, and thus the grant money, to pursue your dreams in reality rather than on the blackboard. It frustrates you that the US Government won't make policy based on the great potential of MA.
But I've been realizing recently that the US Congress is way too busy to
deal directly with new stuff. It has to be lobbied in, or
pork-barrelled in, or otherwise slipped in through the channels that
have evolved to protect our representatives from complete meltdown.
But Congress is only one small part of the Government. And that's only
one government in the world. I want to look at multiple options.
Congress-mediated science funding is one way to go, but I'm not
convinced it can happen quickly enough.
Policy implications? Do a thought experiment with me. My name is K. Eric Atkinson and I have done an immense amount of mathematics demonstrating categorically that humans can transmute matter by mental energy. .... If we don't do it, the Russians will!
Ain't gonna happen, is it?
Two words: Distance viewing. Sure, it's probably completely
bogus. But when the Soviets started trying it, the US (SRI)
started investigating it too. So don't get too pessimistic.
You don't have science yet: you have the glimmerings of science. You need data. Perhaps more importantly, you need to demonstrate to the world that you have data: or are at least in search of data. Only then do you have a hope of being regarded as respectable.
In other words, it's another PR problem. There is actual data all
through Nanosystems. Bond lengths and energies are well known. The
molecular modeling software is based on experiment. Many of the
calculations are based on physical law, and are as reliable as geometry.
Of course, Nanosystems as a whole doesn't prove that diamondoid nanobots
can absolutely definitely be built. But it certainly demonstrates that
they're plausible--yes, grounded in experimental results.
I'm a little surprised that there'd be any question about whether we're
in search of data. Haven't we been saying we want funding and
experiments and further study? Asking to be considered as a serious
science (which should always want more testing)? Apparently, we've
missed a fundamental problem in communication. Are you saying that
those in the establishment belive we *don't* want data? Is this just
your impression of us, or have a lot of scientists made this complaint?
You cite engineers who were derided for forecasting ridiculous speeds of 40 mph. But you make a false syllogism here: Great scientists were derided, the MAM is derided, thus the MAM is great science. Nope: logically possible, but not logically certain. Goofballs are derided too, and justly. And They Laughed At Me For Saying The Moon is Made Of Green Cheese.
I wasn't trying to prove anything here--simply observing that derision
is not proof of bogosity. And engineers who project amazing
capabilities can expect to be derided, even if their projections are
completely reasonable from an engineering point of view.
It is not possible to push nanobot activism and gain respectability. Or it is, but only if the activism is hidden: if the face you present to the world is, well, dull.
Like I said: we can challenge the system, or break the rules, but not
both. We've been trying to do both.
Keep the flame in your hearts, but keep it off the public talks and NIH grant applications! Drexler wasn't in trouble the moment he published Engines. But he was in endless trouble as time went on and he kept embroidering Engines rather than insisting on the performance of corroborating experiments.
Nanosystems is *not* embroidering Engines. It is far more limited in
scope. The whole point of Nanosystems was to establish in detail the
theoretical feasibility of a very limited and preliminary kind of
molecular nanotech. This is not embroidery; it's support. Sure, it's
theoretical support, not quite as good as experiment, but the only kind
of support one can accomplish without funding.
While doing the work that went into Nanosystems, Eric was also calling
for experiments. Sorry if the scientists missed that. It's probably a
context/communication problem again, since he was often wearing his
activist hat while making that call.
Heck, I know what makes the man tick. He has the rare strength of mind to keep before him day to day the fact of his own mortality, and to try (as no one since Gilgamesh has done) to oppose his will and his mind to it. That, I think, is his ultimate motivation.
Again, I think you're reacting to a composite character. If this were
Drexler's motivation, he would have signed up for cryonics long ago.
But I fear it's slowly turned his mind to mush. The MA IS possible! Physical immortality WILL BE the result!!! Yah. Show me.
Where does Drexler say this? How much time has he spent saying it? You
may be confusing him with Merkle. I think Drexler has spent most of his
time talking about other consequences of nanotech. In Engines of
Creation, for example--one of his most speculative works--he spends less
than 20 pages talking about cryonics, including biological background,
feasibility arguments, and social implications.
Even Freitas, detail man though he is, is still just doing calcs and spinning tales till a real invention is demonstrated. Enough chat, already.
Mike tells me that Freitas is well respected even outside the nanobot
crowd. Doing calcs? Yes. Tabulating useful data? Yes. Spinning
tales? Have you actually read Nanomedicine? Calling it "tales" is
quite unfair. It does contain several nanomachine-type inventions (many
of which are presented, not as proposals, but simply as exploration of
likely capabilities). It also contains many sections that do not depend
at all on MNT, including physics calculations, extensive and
well-researched collections of biomedical and anatomical facts, and
discussions of nanoscale design issues that will apply quite as much to
implant medicine and conventional nanotech as they will to MNT.
Put it this way. When I want to look up a physics formula and how to
use it, I reach first for Nanomedicine--if it's not in there, I fall
back on my college physics textbook.
I will say it till it sinks in, Chris: calcs by themselves are not enough, no matter how sophistimacated they are.
We agree. But see a few paragraphs below...
Tom Theis's pronouncement about business levels need not apply to something transforming; but that's true only in retrospect. At the time of transformation, even the most revolutionary technology must come across (for sordid marketing reasons) as an agent of only incremental change. Iron? Just bronze with an attitude. Petroleum? Merely a hay-less horse.
A very, very good point. I think we've been missing this. We know that
MNT *could* (if the theory holds) be a transforming technology. And so
we expect that if we can demonstrate this, it will be treated as such.
But what we don't realize is that transforming technologies have never
been treated in any special way, because they were only recognized after
There is in fact no standard for how to treat transforming
technologies. I understand that even nuclear weapons were not viewed as
transforming until after people saw and reflected on what actually
happened in Hiroshima and Nagasaki. So it's probably a fundamental
error to expect any technology to be treated as transforming.
They can say they were asleep when you made your summa, or watching Legally Blonde II, and it's a valid argument. The only thing that will convince them, whatever their arguments, is to produce the technologies that their arguments deny.
So here's the question. We have a chicken-and-egg problem here. We
can't produce the technologies without funding. We can't get funding
without convincing the skeptics. And we can't convince the skeptics
without producing the technologies. How can we break this deadlock?
Science has some room for speculation. Again: we can challenge the
system as long as we don't break the rules. So one way out of the
deadlock is to produce some boring papers that convince people that the
topic can be interesting. After they start looking at it, they might
produce some papers, and eventually a new field is started.
One problem is that this will likely take too long. "Science advances
one funeral at a time." We could expect it to take another 20-30 years
for the basic paradigm of MNT to be accepted in science, and then a few
more decades for useful mechanochemical results to be published, and
then for the molecular machines to be invented... From a policy point
of view, this sucks. Because long before then, some country will have
started a Manhattan Project and will likely be able to revolutionize
their armed forces almost overnight.
Another problem is that we've already broken the rules. How can we
recover from that? Will the system forgive and forget, or have too many
people already lent their reputation to reactions against MNT?
I asked above: will a moderate voice be accepted, or merely increase the
cacophany? To this, I add other questions. Will the NNI's definition
of nanotech be taken as that moderate voice, dooming all of MNT to
fringe status regardless of what we do? (Is Smalley already the
President of the World Nanotech Association?) How much will it be
necessary to exclude from the "moderate" writing: obviously Engines, but
what about Nanosystems?
That is the one unanswerable thing that the MAM can do. To do this, it needs lab facilities and money. And to do THAT, it needs to abate the wildness of its claims to access federal funding. That, old son, is the practical essence of my MarCom advice.
Note that Nanosystems was supposed to do this. There aren't many wild
claims in Nanosystems. Nothing about cryonics; nothing even about
health, or social issues. Nanosystems was supposed to be the sober,
dull, meet-them-on-their-own-ground science.
So, why didn't it work, and how much of it can be repackaged so that it
will work? This is not rhetorical. You will help us immensely if you
can tell us how to present the work in Nanosystems so that it will be
taken (as it should be) as the first step toward dull, sober, scientific
analysis of MNT.