Scott Locklin on Nanotechnology and Drexler Sunday, Sep 5 2010
nanotechnology 12:42 pm
Some of you may have been following Scott Locklin’s “reality check” on nanotechnology, which was linked by CrunchGear and Hacker News.
My opinion of the post is that is confuses Drexlerian nanotech with nanotechnology “in general”, and makes many major errors, including denying the existence of micromachines and nano-sized elements that drive larger systems.
The article is also wrong because it claims that, in his book, Eric Drexler is just porting macroscale designs to the nano-world, but the entire work (Nanosystems) takes great pains to analyze the differences between the nanoscale and macroscale and introduce engineering innovations that could be a good starting point for true molecular manufacturing. Another error the article makes is suggesting that Drexler dismisses using biology as tools for nanomachines, which is ironic considering that Drexler advocates “molecular and biomolecular design and self-assembly” approaches to molecular nanotechnology, and often discusses the protein folding path on his blog.
Drexler posted a response to Locklin in the comments section:
Hi Scott,
In my view, molecular and biomolecular design and self-assembly are the most promising directions for lab research in atomically precise nanotechnology. There’s been enormous progress — complex, million-atom atomically-precise frameworks, etc. — but much of the work isn’t called “nanotechnology”, and this leaves many observers of the field confused about where it stands. I follow this topic in my blog, Metamodern.com.
Regarding the longer-term prospects for this branch of nanotechnology, there’s a publication that offers good starting point for serious discussion.
The technical analysis that I presented in my book Nanosystems: Molecular Machinery, Manufacturing, and Computation, (it’s based on my MIT dissertation) was examined in a report issued by the National Academy of Sciences, on “The Technical Feasibility of Site-Specific Chemistry for Large-Scale Manufacturing”. The report finds no show-stoppers. It notes uncertainties regarding potential system performance and “optimum research paths”, however, and closes with a call for funding experimental research.
This report was prepared by a scientific committee convened by the U.S. National Research Council in response to a request from Congress. It is based on the scientific literature, and on an NRC committee workshop with a range of invited experts and extensive follow-on discussion and evaluation.
I think that this report (and the Battelle/National Labs technology roadmap) deserves more attention from serious thinkers. It deflates a lot of mythology about a topic that just might be real and important.
If either of these publications has been mentioned above, I missed it.
In general, I think Locklin’s post is a very well-designed piece of flamebait, and I commend him for drawing attention to his post. Some group of people really love talking about nanotechnology, and they need some outlet, and this is a fine outlet of the week. Locklin is right that a lot of nanotechnology is just chemistry or materials science with a cool name slapped onto it, but certainly not all of it.
Funny quote from the comments thread: “any sufficiently advanced technology is indistinguishable from a rigged demo”.
From Locklin’s bio on his own site:
“I have a particular dislike of self-anointed ‘experts’…”
Reading his blog, I’m amused by the irony of this statement.
That creature gladly defended hypocrisy as a virtuous personality trait. He is an unsuccessful academic who decided to burn the entire field because he couldn’t play in it.
It’s just a lot of frustration, resentment and hate coming from a strange right-wing anti-everthing-humane selfish point of view.
Dude, that comment is freakin’ hilarious.
I do not like his presentation, but he makes a good point: the burden of proof is on the nanotechnology crowd. Only time and science will tell if Scott or Drexler is right.
One of Scott Dickling’s comments:
“Well, that’s a great example of how goddamned science fictiony all this is. Beyond the fact that this gizmo can hardly turn everyday objects found in its environment into copies of itself, and never will be able to. Remember; that’s why we’re supposed to be afraid of/want nanobots. They’ll give us godlike powers of creation, right?”
How the fuck is that well designed flamebait?
The man’s a troll and proud of it.
Pardon my ignorance, but don’t chipmakers work at the nanoscale?
Jay, it’s a well-designed flamebait because it got linked from prominent blogs. Getting linked isn’t easy — you try it. It had just enough technical detail to look vaguely legit. Maybe this isn’t an art you’re interested in, but I certainly am. Call it “memetic engineering”.
Panda, he’s talking about nanoscale machines manipulating things at the nanoscale, not macroscale machines building nanoscale patterns.
I understand what you mean.
And yeah I’m interested in it. I’m in Internet Marketing these days!
Burden of proof. Which direction does it go in?
Null hypothesis. What is it in this case?
I cannot remember the number of times I’ve had to correct someone talking about “nanotech” in the senses it got waylaid by, so much so that I finally adopted Drexler’s solution of dropping “nanotech” as a specific, and using Mechanosynthesis when talking about directed molecular manufacturing.
Despite Bush and Smalley’s derailment of the NNI, I have been amazed at the amount of progress we have nonetheless made in the last decade, but that pales to the progess we have made in just the last two years alone. I think we are beginning to approach the knee of the curve, and might even reach it next decade. While the “Drexlerian” vision might still be 15-25 years away, I fully expect to see mass production of “modular” molecular manufacturing, in which easy to make molecular “legos” like CNTs and Graphene are commonly used by the end of the decade, and possibly even a “crude” nanobot in the form of a computer controlled stem cell “swarm” being used to repair damage to the human body, or to modify it cosmetically.
So yeah, I think you are right,Michael.This was flamebait, and completely ignores the reality in hopes of garnering attention.
The burden of proof is on those who would show that a new theoretical concept can be transformed into technology. But the fact that burden remains to met is not the same as the fact that it will never be met. It may well be that Dr. Drexler’s theories will never work out. I think even he would have to admit that his theories remain speculative. But this is no insult. They also remain useful for provoking thought and new avenues of research.
As to whether current developments make his speculations more probable, I lack the training to know.
Panda:
There are no “burdens of proof” in science as there are in law. This concept adds nothing to the discussion.
Who has the burden of proof in proving that the laws of physics permit a building 140 stories tall to exist or permit a manned spacecraft to Mars to be built? Neither exists yet. Are we supposed to be totally agnostic as to whether the laws of physics permit them?
It law, one party has the burden of proof to provide evidence that something is likely to be true. When it does, it is said to have established a “prima fascie” case. The other side then has the burden to rebutt it, by carefully answering the propositions that the first party has asserted.
If you want to use this type of nomenclature, Drexler has established a prima fascie case (met the burden of proof) with the highly detailed arguments in Nanosystems. His opponents have the burden to rebutt by answering these propositions in detail. Instead they provide nothing but vague ad homenein attacks.
I agree that the legal concept of a burden of proof is not applicable–hence I will not argue whether Dr. Drexler has established a prima facie case. (But as a sidenote, meeting the burden of proof in law is not the same as establishing a prima facie case–which is technically only meeting the burden of production.)
My point is that science is a field that requires proof (evidence) before a proposition is accepted. You may reply, as Carl Sagan wrote, that the “Absence of evidence is not evidence of absence.” However, I am not saying that Dr. Drexler is wrong. I cannot, however, say he is right until evidence shows that he is right. As the Battelle Report. which Dr. Drexler himself cites. notes, there is as yet no confirmation that he is right or wrong. Further research is required to determine whether his theories can be implemented.
If you reply that all theories have a continuum of evidence and that we cannot have a definitive answer on this matter, then I disagree. In this case, Dr. Drexler can prove himself right in a very simple way: he can present to us a working model of one of his conceptual devices.
You are correct that an opponent has a burden to prove he is wrong, before we can say he is wrong. After all, if an opponent proves one of Dr. Drexler’s theoretical points wrong so that the whole framework collapses, then that will indeed also answer the question. But simply because an opponent has not proven Dr. Drexler wrong does not mean his theories can be implemented. We will not know that they can be implemented until they actually are.
Concerning the two proposition of (a) “Drexler is wrong” and (b) “Drexler is right”, BOTH still require proof. But the fact that one has not been proven does not prove the other right.
The fallacy of many of his detractors is that since no one has proven (b), then (a) must be true. Your fallacy is that since no one has proven (a), then (b) must be true.
Science does not work this way. Science, for a new idea about reality requires evidence. It does not say that the new idea is wrong–it withholds judgment. But it certainly does not say it is right!
Panda:
I agree with you. I think that our only disagreement may be regarding what counts as evidence and proof. Neither Drexler can prove with 100% certainty that he is right, nor can his detractors prove with 100% certainty that he is wrong. Biological cells are universal constructors across the domain of amino & nucleic acids, lipids and other biomolecules. Drexler hypothesizes the existence of universal constructors across the domain of all stable molecular configurations (or something close to this). He argues that computer simulations, the existence of biological systems and experiments with proximal probes are evidence that universal assemblers. Critics may argue that simulations are unreliable, biological systems can only create biomolecules and proximal probes only work in extreme conditions. Neither side has a burden. Everyone agrees that more research is needed. But Drexler argues that this additional research should be funded. His critics are essentially arguing that these ideas are so wrong that additional research is not warranted. That is an unusual position considering the possible impact of nanotechnology and that no one has found any theoretical impossibility.
As for proving his point by building an assembler, that is analogous to the following: An architect proposes to build a 140 story skyscraper. The architect seeks funding and encounters critics who argue that the laws of physics forbid the existence of 140 story buildings. The architect argues that 120 story buildings presently exist and that computer simulations and theoretical calculations show that 140 story buildings would be stable. The and the existence of 120 story buildings are an existence proof of the feasibility of 140 story buildings. The critics argue that only after the architect actually constructs a 140 story building will he have met his “burden of proof”; and only then will they consider funding for the project. Until then they will remain agnostic whether the laws of physics permit the existence of 140 story buildings just as they are agnostic about the existence of fairies, gods, aliens, etc. This is, to caricature it a bit, the position of nanotechnology critics.
I generally agree. But as a layperson who must choose between competing experts, I am very lost. I do not buy the ad hominems, but it hardly helps me evaluate Dr. Drexler’s ideas when scientists do not give it a fair go. Without some review I have to remain dubious. I hope that explains my personal hesitation. I do not blame Dr. Drexler for the lack of good review. It is certainly not his fault. But laypeople cannot just accept theories without seeing result or peer review. If you are someone with the training and luxury of time to evaluate his theories, then it is a different matter. So I agree with you that the evidence demanded is the relevant question. It may differ for you and me.
No, science does not require proof; in fact, nothing can be proven in science.
Since requires falsifiability.
In science, everything is a theory, until such a time that maybe it is falsified, and replaced by a more apt theory.
Drexler made a comment once concerning skepticism about nanotech, I think it might have been in ‘Engines of Creation’, not sure.
The gist of it was, “want to see something assembled by nanotechnology? Want to see nanomachines in action? Go look in the mirror. You are a product of nanotechnology, and the nanomachines are called DNA. Still think it’s ‘impossible’?”
Locklin, like so many others of his naysayer type, suffers from a failure of imagination.
If a goal is desirable, we will pursue it. If it is possible to achieve, no matter how long, laborious and circuitous the path, we will achieve it.
Even if we fail, failure teaches us something perhaps even more valuable: the limits of the possible. If we were to listen to sceptics, we would never know.
No amount of naysaying – or evangelism for that matter – will alter the outcome.
I think this is what is amazing about the human spirit. Epic goals are what give us purpose beyond the humdrum of everyday endeavour. If we listened to the Locklins of the world, we still be living in caves.
This goes as much for the ‘singularity’ as nanotechnology.
DMan:
You have a good point. After decades of whispering and bad talking (ie Smally never having read Drexler’s actual proposals and not knowing about non-aquatic enzymes) , nanocritics finally have a serious proponent in Richard Jones: http://www.softmachines.org/wordpress/?page_id=346
Jones argues that biology-like systems are the closest to molecular nanotech that is physically possible. Jones argues that Drexler is right to propose protein assemblers but wrong that protein assemblers will develop into “hard” diamondoid universal assemblers. Jones argues that small diamond clusters (like molecular gears) will spontaneously reconfigure into other shapes. He further argues that proteins are soft (being single covalent bond chains folded and held together by weak molecular forces). This softness lets them survive Brownian battering and to move during chemical reactions where bonds are made and broken. Hard diamond objects will not have these abilities.
So biology is not a complete existence proof of molecular nanotechnology. For me, the real questions are:
1. Whether protein like systems can create multiple covalent bonds in a lattice configuration (like diamond or fullerenes) or only single chain peptide bonds (as in existing biological cells).
2. Whether diamond and fullerene systems (produced by the above) are stable and can serve as logic gates, sensors and actuators.
I suspect that the answer to both questions is yes and that Drexlerian nanotech, or something approaching it for all effects and purposes, is possible. Professor Jones critiques are, however, valuable and should be studied, even if we disagree with his ultimate conclusion.
This is what scientific criticism should be. Rather than pontificating that something is impossible, show what possible flaws in the designs are. If no possible designs can overcome the objection, the proposal is impossible. By offering serious criticism, we speed up progress.
I can’t think of a stronger contrast with Drexler’s nanotechnology mirage than Craig Venter’s new venture:
His Corporate Strategy: The Scientific Method
http://www.nytimes.com/2010/09/05/business/05venter.html?_r=2&src=busln&pagewanted=all
Venter has illustrated Locklin’s advice of growing up and attempting to achieve real things.
The gist of it was, “want to see something assembled by nanotechnology? Want to see nanomachines in action? Go look in the mirror. You are a product of nanotechnology, and the nanomachines are called DNA. Still think it’s ‘impossible’?”
This is an argument for the existence of biological “wet” nanotechnology. It, in no way, implies that nanomechanical “dry” nanotechnology is possible. Craig Ventor and many others all over the world are working to develop biologically-based nanotechnology. There is no question that this is possible and that this technology will have many (but not all) of the capabilities that the “dry” nanotechnology is supposed to be capable of.
Abelard, my objection to arguments of the kind that Locklin makes is that his assertions are premature.
There are enough respectable, credentialled supporters of dry nanotech out there to take it as a serious possibility. Note – possibility, not certainty, but a goal worth pursuing for the potential enormous benefits.
People like Locklin seem to be saying that because they think it’s ‘impossible’ we shouldn’t even bother to investigate it. Or, after only a short time (relative to other scientific endeavours) of pursuing it, give up!
Should we listen to such people? Those who ‘talk’ more than they ‘do’? The history of science is littered with ‘impossible’ achievements that have transformed science and society.
Critics don’t ‘know’ that it’s impossible. They are basing their assertions on a framework of reference that may be in error.
In nanotech research we are in a continuous process of discovering unexpected – and largely unsuspected – phenomena and properties that manifest only at the nanoscale.
For instance, metals that are soft and ductile at the macroscale are hard at the nanoscale. Since many of the critics did not predict this (to my knowledge anyway) it seems rather arrogant for them to state they ‘know’ what’s impossible at the nanoscale.
The barely-understood Casimir force makes its influence felt at the nanoscale, and perhaps we may even learn how to manipulate it to achieve nanoscale goals.
Point is, we don’t know what we don’t know, but Smalley’s invective against Drexler, that he is an engineer and not a chemist, is ironic: some of the little-understood phenomena that manifest at this scale neither Drexler or Smalley are qualified to evaluate. They enter the domain of the physicist.
I think the problem is that all three frameworks of reference – engineering, chemistry and physics – are inadequate in isolation to evaluate (or achieve) dry nanotech.
I think the problem is that if it is indeed possible, perhaps unlike wet nanotech, it’s going to require a whole new discipline that unites the three, to make effective progress.
I’m reminded of the metaphor of the three blind men who encounter an elephant. One feels the trunk, decides it’s a snake, one feels the ear and decides it’s a giant leaf and one feels a leg and decides it’s a tree. If they just spoke to each other, they would ‘see’ the elephant.
If dry nanotech is to succeed I think it needs to see the elephant, so to speak. And I suspect ‘true’ nanotech will be quite different from what either side of the debate predicts.
I dont know if this was picked up already, but Scott Locklin is just regurgitating his own vomit from 2006. Check out http://lupoleboucher.livejournal.com/32527.html
Seems the reaction wasn’t big enough the first time