For those that are skeptical about nano machinery, google 'ribosome' and be amazed.
Nano machinery is real, it exists, it powers the world, it's called biology.
The problem is that we seem to have a hard time translating our hard won knowledge of mechanical engineering to the nano scale, I expect some time will pass before the SF visions will become a reality but I think eventually we will get there. Figure another 50 to 100 years or so?
As long as we keep thinking in terms of conventional mechanics (wheels, gears, levers, wires and so on) it's going to be slow progress because we are trying to push down the size limit on our macroscopic ingredients without a real appreciation of what you can do with molecules 'as is'.
I think at some point the 'genetic synthesis' crowd is going to meet up with the 'nano technology' crowd and that's when we'll see some real action. The first little bits of progress in that direction have already been made.
5 to 10 years? 500 to 1000 years? 5000 to 10000 years? 5 to 10 months?
What are we basing our estimates on?
That is the author's real point. If we don't have a good enough handle on the principles necessary to make this into an engineering problem, there just isn't any sense in trying to guess when these technologies will arrive.
Unless you have some facts or data you're keeping from us.
Based on the slow progress of the last 30 year in 'hard nanotech' and the quite amazing strides in the last couple of years in our understanding of and our ability to synthesize 3 dimensional structures at the nano scale (including DNA), and to harness biological components to do different work.
The two fields are complementary I think, one is pushing from the top down trying to make our existing structures smaller (with all the trouble inherent in that, for instance, at the nanoscale there is no such thing as 'smooth', imagine making a bearing out of sandpaper or whatever you can come up with as the nearest equivalent) the other is coming from the bottom up as we learn how our biological structures are formed and what mechanisms are available both as tools and as a working example to study.
Sooner or later (but most likely a little later) there will be a meeting of minds, where new structures will be produced by re-using or adapting the tools that biology has already given us.
I think initially that will be a purely materials science oriented endeavor, mostly concentrating on the medical applications, or a direct application of synthetic molecules (such as artificial DNA) but it's possible there will be more 'mechanically interesting' applications as well.
Nanotech is a lot more than just moving parts, plenty of it would be classed as 'structural engineering'.
The biological part is the one that is moving fastest at the moment, I think at the present speed that two fields will meet and spawn a third field within a very short time, if that hasn't already happened.
For now this meeting place of technology and biology has been named 'synthetic biology' and there is very good progress there, I expect true nanomachinery to be a spin-off of this new field.
Synthetic Biology has made good progress in those four years. See, for example, Craig Venture claiming to create first truly synthetic life-form. If you are in the field, you will start noticing that terms are overlapping. So, your idea of synthetic biology may be related to nano-machines while my idea of synthetic biology may be related to genetic engineering.
Realize that a plasmid (what they are creating and injecting) is usually a loop of ~4000 base pairs while your DNA consists of ~3 billion base pairs. That's not much shell code, especially when we haven't found a shell to run.
"It also depends on what your time scale is. I’m willing to bet, at some nebulous point in the future, long after Drexler and I are dead, someone may eventually develop a technology sort of vaguely like what he imagines."
so really, this guy agrees that it's happening, but he disagrees with the timeline. but he didn't actually provide any information or argument about the rate of progress of miniaturization.
I read it as that this guy disagrees that it's happening. But taking the path of "never say never" he agrees that it might happen far in the future. But there is no sign that it's happening. Not even any sign that it might happen. No progress in 50 years. Zero. That was his point.
I wasn't too keen on the pessimism in his article; I can agree, in principle, with what he was saying but the point of emergent fields is in how unpredictable they are - this may lead to unrealistic expectations (which can often times drive innovation in the long run, TBH).
I've been accused of being too much of a stickler on this, but while the ribosome and other biological machinery are at the nanoscale, but they are NOT nanotechnology. If you use the Drexlerian definition, nanotech much be engineered, which ribosome is definitely not. We can certainly learn from natural systems, but that doesn't make them nanotech.
Also, nanosystems fundamentally are at a scale when quantum properties become important, and classical, mechanical rules start breaking down, which makes things really hard.
Having said that, there are some great examples of nanosystems around, from nanoparticles, to polymer drug delivery systems to materials for PET scans and eventually scalable manufacturing of quantum dots etc. The problem is, that breakthrough is always round the corner. Nano's going to be big, but (a) it is not one technology and (b) we need to be patient cause at those scales things are hard.
How would you feel about protein folding if it was at the direction of a completely artificial genome creating proteins that do not normally occur in nature to exact specifications?
Still, I would argue the biological machinery would count as proof of concept. Evidently it is possible to create self-replicating nano machines. So it is not ridiculous to think about it, as the article claims.
Artificial carbon nanotechnology has been used by man since we had fire. It's called "soot". We still don't understand all of the physics of soot formation. The vagaries of burning and soot formation can result in some incredibly tiny particles. This is why smoke is so often used as a drug delivery method and why smoking fish and meat works so well: soot particles can get small enough to work their way between and sometimes through cell membranes.
We recently discovered structures in soot we now call "Buckyballs" and "Buckytubes".
I don't know why, but there's a fascination with everyone outside the materials science/chemistry/physics world that nanotechnoloy is only about little tiny moving things inside a larger system. Yes, there are bio-molecular robots that are supposed to be used for things like drug delivery and whatnot.
But nanotechnology encompasses such a broad array of other disciplines that are primarily concerned with operating at the nano-scale. For example, anyone doing research that involves the electromagnetic spectrum (light) would be considered doing research in "nanotechnology". Those people would include: solar cell manufacturers, laser physicists, radiation specialists, electrical engineers...and so on
Hell, since nano is just a length description, ANY profession could be considered "nanotechnology". "Yes, we our car can travel 26 822 400 000 nanometers per second, its amazing"
And that sounds like what the author of the OP said here:
"I’m willing to bet, at some nebulous point in the future, long after Drexler and I are dead, someone may eventually develop a technology sort of vaguely like what he imagines."
It's been a while since I read Drexler's books, but if I remember correctly he alluded to more mechanical processes than biological ones. The article you linked to is purely biology and not really what Drexler was arguing for.
No, it's not purely biology. It is using biological elements to synthesize artificial constructs with exact mechanical dimensions.
By analogy, when a carpenter takes a tree and cuts the tree into boards that's engineering, not biology even if the raw materials were provided by biology.
And when you then use those boards to build a house, that's engineering too.
Evolution took billions of years to engineer the ribosome (and more importantly to engineer the whole supporting ecosystem where ribosome can do what it needs to do). I doubt humans will be able to do that in next 100 years.
Granted a similar argument can be made for many other comparisons (such as camera or airplanes) but at nano-scales physics of environment plays a tremendous role. The beauty of evolution was that it is an inside-out distributed process so the physics of environment actually helped engineering ribosome.
I wonder if humans will fall back to mechanisms such as guided evolution to create other "nano-machines". Lots of effort has started being put into that direction. Just google "artificial enzymes"
> Evolution took billions of years to engineer the ribosome
I see this argument a lot, but it doesn't really hold water. Evolution wasn't trying to engineer the ribosome. In fact, the only thing evolution tries to do is replicate. The biological forms that have come out of that are completely random.
Humans, on the other hand, consciously make efforts at engineering. Thousands of monkeys will never (for all practical purposes) manage to write a Shakespearean work, but Shakespeare did.
100 years ago we barely had phones. Almost no one had a car. There were no televisions and movies didn't have sound. Human flight was relegated to hot air balloons. On the other hand, total Internet bandwidth has increased from about 10 terabytes a month to 10,000,000 terabytes a month in the last 15 years. Trying to guess the limits of what we'll be able to do in the next 100 years is unfathomable to me.
> Evolution wasn't trying to engineer the ribosome.
Evolution isn't purposeless. Sure, it is a random process but it has a purpose. And that's of survival of the organism (or if you like Dawkins, genes that encode the organism).
> The biological forms that have come out of that are completely random.
Wow, biological forms are NOT random. They are what they are because of the environment they are habituated in. The exact implementation (animal) may not be deterministic but if you set the right fitness function, evolution will produce what you wanted it to produce. Evolution is like a blind tinkerer who knows he needs to fix but doesn't know where to hit the hammer to fix it. He only gets feedback once he has hit the hammer. From the feedback, he can definitely avoid hitting at the wrong spot again and again. I wouldn't call this process random.
My original argument is about engineering macro v/s nano machines. Evolution is great at doing nano work because it is an inside out process.
PS: Thousands of moneys _will_ produce a Shakespearean work if their survival depended on it! (Just kidding)
> Evolution isn't purposeless. Sure, it is a random process but it has a purpose. And that's of survival of the organism (or if you like Dawkins, genes that encode the organism).
Is survival a "purpose" or is it just the consequence that is selected for? (Survival isn't a property like length or color.)
You're right. What I meant to say was that biological forms are not the end goal of evolution, as opposed to engineered products which generally are the goal of human production.
Thousands of monkeys will never (for all practical purposes) manage to write a Shakespearean work, but Shakespeare did.
I always wonder why people choose monkeys for that analogy, seeing as we're more or less descended from a recent ancestor. At some point there were only thousands of that ancestor, but they managed to produce something capable of writing Shakespeare within a few million years! Maybe a better animal to choose would be octopi...
It's a false analogy with regard to evolution. Purely random input is extraordinarily unlikely to match any specific output. But if you bring in inheritance, mutation, and environmental pressures then you can produce just about anything, given sufficient time.
I like this perspective, I prefer optimism over critical pessimism. This stance is really what drives innovation - not being restrained by perceived or even real limitations.
Real limitations aren't "real" though, in the sense that some sort of boundary or new knowledge extends the demarcating line between it being a "real" limitation and an "unknown" limitation. A matter of degree I suppose. I'm certainly not going to argue that humans will suddenly begin sprouting wings on their backs, but people at some point in history - I'm sure - claimed human flight to be an impossibility, a gift only birds and gods possessed. That is, until the airplane was created, which extended that "limitation".
Many etiological examples can be drawn, but can you not agree that saying "real limitations will always restrain you" is too absolutist?
"Evolution took billions of years to engineer the ribosome (and more importantly to engineer the whole supporting ecosystem where ribosome can do what it needs to do)."
This isn't supported by observational evidence. Every known life-form on Earth, even the Archaea, makes use of the same fundamental nucleotide transcription machinery (which includes the ribosome). Every indication is that the ribosome is one of the oldest cellular structures, which pegs the upper limit for the amount of time it took to arise via evolution at a few hundred million years.
In the billions of years since life began, evolution never took a life form, put it on the moon, and brought it back. I doubt mankind can do it "before this decade is out".
Practical nanotech will inevitably harness the forces of life that already exist; why wouldn't it? Saying it's not real Drexlerian nanotech may be true, but it will also be a very uninteresting thing to say. Who cares? What's happening and what's real is what matters, and what's really happening has definitely started and isn't stopping anytime soon.
I hate to complain about headlines, but I feel this needs saying:
Headlines such as the one this author has chosen strike me as unpleasantly Digg-like. Exhortations to "stop talking about it and start laughing at it" are anti-intellectual; it's a call to glib derision rather than reasoned rejection.
Yeah it's just like the whole "Fail" meme. It's indicative of how demanding people are becoming that they no longer can be bothered to write a sentence or even conjugate a verb to express their disgust with some flaw that catches their attention for a microsecond.
The name is all about the funding. There are many 'nano'-technology grants, which focus on devices that are tens of thousands, or even millions of nanometers large. Micro-techology, however, doesn't get the same funding from the NSF et al. MEMS (http://en.wikipedia.org/wiki/Microelectromechanical_systems) has come a long way in a very short time, and, research-dollars willing, will shrink further, so the author is wrong when he says nanotechnology will take another century.
a lot of machine learning is parametric statistics...
and while I agree the term is really loaded, ML is different from statistics in its approach and the discoveries it made. For example, ML showed that some techniques are much more effective at scale (large datasets) than suspected by most statisticians, as recognized by "hard core" statisticians like Wasserman.
Very true. In the life sciences it was 'biotechnology' in 2000, 'nano-technology' in 2005, and now it is 'synthetic biology'; but they're all the same and have changed names to get the funding.
I think those terms are in part to indicate a progress of understanding as well. It's not all nonsense, it's just that when everybody else uses the next available buzz-word if you don't follow suit you will be left without funding.
The people that write the proposals are going to try to maximize their chances, think of it as marketing.
Nano-technology was terribly oversold so it lost it's glamor, but the concept is real, synthetic biology is real as a concept and only just now beginning to make some headway, eventually, long after it has lost it's glamor too it will be a mainstay of industry.
Just like the humble transistor is no longer glamorous, it's 'just' technology now. But in 1947 it was a miracle.
Buzzwords ideally should lose their power because the tech behind them becomes commonplace, when they're oversold we have a problem (looking at you, AI).
I really wish that the author would have, when complaining about "Drexler's Book", been clear about which one he was talking about.
Also, "Mechanical objects on microscales do not exist" is gratuitously false, we even have really simple mechanical objects (like cantilevers) on the 100nm scale now.
Not that I necessarily agree with this particular summary in everything, but the point of summarizing something it to make it easier to digest the core of the argument before criticizing it.
For that to work you have to have a really good understanding of the subject matter though, and I think he does Drexler great disservice when he says "He seems to lack the imagination, and of course, the physics to figure out what a real nanosized doodad might look like.".
That to me is the hallmark of someone with an ax to grind rather than an objective reviewing of a book written quite a while ago.
Of course Drexler didn't have the physics to figure out what a real nanobot might look like, nobody did, and nobody does.
Drexler could not have easily foreseen some of the obstacles a direct translation of mechanical concepts would encounter, but there are more ways to skin a cat, and biology seems to have found at least one of them.
Now it's up to us to find either another path or to harness biology to produce those structures that we can not produce by direct means and to build up a library of tested components to do our 'nano engineering' with.
I see Drexler more as a visionary than as a 'hard applicable science' guy and that's the way to approach his book. If you're looking for a 'ready to build' nano bot or a hard treatise on nano engineering you're not going to find it, if we had had that at that time then we wouldn't be where we are today.
Incidentally, I think that the 'proponents' of nano tech have done a very bad job at raising expectations (Drexler among them), I do believe that long term nano technology will become a reality and will become a mainstay of our technological arsenal. Just like electronics, which were practically non-existent one hundred years ago are today.
i understand what summary is for. what i don't "understand" is why he chose as the target a book that's 24 years old in a fast moving field. actually i do understand: it was a strawman!
This is a non-moving "field." It is nothing but a marketing word for third rate research on things which used to be called "tribology" and "chemistry."
That's actually huge problem around the whole 'nanotech' sphere.
Everybody seems to have a different definition for everything, usually the definitions tend to absurdity depending on whether the person is a skeptic or a 'true believer'.
It's very boring because in the end the terminology should be secondary to the goals and how we go about achieving them.
Just skimming through those pictures, but most of the fairly complex mechanical parts, at least the ones which have scales, seem to be of the order of 10-100um. Not sure what would actually constitute nanotech there, but certainly the minimum feature size (spacing of teeth on a gear, etc) is probably <1um, which would count in my book.
Flying cars have long been a symbol of the possible technological height of human transportation. But we're not exactly surprised by their absence. Similarly, can't Drexler's '90s idealism be regarded as the possible peak of nanotech? Does the fact we don't have tiny toasters in our bloodstream really negate the acheivement of nanotechnology in medicine and materials science? Personally, I don't think so. This article is too deliberately inflammatory, and doesn't properly look into anything beyond what is now an ancient thesis - as far as fast-moving research areas are concerned. This makes his dismissal of nanotech as sweeping as Drexler's predictions.
Flying cars are possible today, but not economically viable.
I know at least one startup building a flying emergency vehicle for urban areas. But this one just looking like a van, but actually has many things common with helicopters.
I think every Russian HNer laughed out loud when they saw this title. Nanotech here in Russia is a synonym of enormous and inefficient government spending on vaporware since they started considering it a "National project". To be fair, I have no idea how true it is, but it does look very much so, especially with frequent appearance in news but no details about actual results.
This is same as saying to someone working on EINACS and early computers that they are a laughing stock and nothing but huge calculators. Reminds me of an article by Scott Aaronson : "Whats taking so long Mr. Babbage" : http://scottaaronson.com/blog/?p=446
As someone who's done a fair amount of work in materials science and "nanotechnology", I have to admit there are some valid points in here. Drexler-type nanobots are still just as far away as when he wrote his book, and a large amount of the "nanotechnology" work out there is called such mainly to get funding.
But the author here overstates his case quite a bit. "Millitech"? There are many microscale mechanical systems out there, both in the lab and in industrial applications. Accelerometers are one example; the actuated mirrors in DLP projectors are another. If there's no microtechnology out there, I know a lot of MEMS engineers who are going to be surprised to here it.
A lot of what's called nanotech now could have been called physical chemistry thirty years ago, especially a lot of the "let's put nanoparticles in it!" projects out there. But there's some pretty real nanotech out there too, even if it doesn't qualify under the self-replicating nanobots definition. One example I can think of is the work being done on nanoscale printing--not lithography, but actual physical printing of materials into nanoscale patterns. And there's some pretty cool controlled nanoparticle engineering work out there too.
Why don't we have anything that can manipulate objects at the nano-scale? Couldn't we use atomic force microscopes to push some atoms together and at least make some interesting molecules (even if this process couldn't be scaled up)?
Also what ever happened to Feynman's idea of building half size remote controlled arms, which we then use to build half sized of that remote controlled arms, all the way down to a very small scale? Has anyone ever tried that?
We do have things that can manipulate objects at the nano-scale. They're called catalysts, and when they're made out of amino acids they're called enzymes, and you and I are built out of them.
Yes, you can assemble molecules with AFM. The problem with this idea is that either (a) the molecules don't stick together at temperatures much above absolute zero, or (b) they do, in which case they might be stable at useful temperatures, in which case it's better to figure out how to get them to self-assemble or to find a catalyst/enzyme that helps them self-assemble. Because your nifty molecule just isn't that nifty if it doesn't exist in quantities of order 1 mole, and 1 mole is... a lot. You just won't believe how vastly mind-bogglingly big 10^23 is.
All of which boils down to this: This essay is great, but the meat of it really is in the first paragraph. Nanotech is a euphemism for chemistry and biochemistry. It is 100% possible -- indeed, nanotech is older than any human-invented technology by many orders of magnitude, by definition, because humans are built out of "nanotech" -- but it's all just chemistry. The real question is why the word nanotechnology is so much more fashionable in grant-writing circles than the old words.
That's kind of the point of the article, materials do not behave on a nano scale the same way they behave at our macro scale. You absolutely cannot scale down, you have to design from the bottom up instead.
The author thinks this is impossible because progress is slow. That's not really an argument, but I'll agree that "just you wait and see!" isn't one either.
However, biology shows us that it's not impossible to do self-replicating "machinery" so I think we'll get there one day, but it might probably be a lot more organic and a lot less mechanical than we currently envision when we say nanotech.
When I was a child, I read a SciFi story about one guy, who built such an arm. Then using this arm under the microscope he built a set of microscopic tools, etc.
Later in the story he fighting with bacteria and viruses, but because of the feedback of his microscopic arm, he was injured and almost killed by these tiny creatures.
That is one of the authors points -- no one has done self replication on a macro scale and nanotech just assumes self replication at the molecular level
This is really a straw man argument. If you go read Drexler's blog(e.g. this series: http://metamodern.com/2009/12/19/molecular-manufacturing-whe...), he is very explicit that the path forward is studying natural self-assembly from DNA and then tearing that apart, finding how it works, and modularizing/modifying it for our own purposes. None of this appears to be a "crazy idea" or something out of humanity's reach til long after we're all dead, as this author claims
Drexler's egotistical to be sure. It doesn't make him wrong
I think it attacks a Drexler that never existed; the chapter in Nanosystems on ways forward outlined similar ideas to that more recent post. Just because the bulk of Nanosystems explores a class of systems that's easier to analyze and currently impossible to build, he wasn't ever saying "Let's try to build this class of systems! Right now!" The OP was a total caricature.
That's not really true, we self-replicate all the time (thanks mom and dad). What we don't know how to do is design things that self-replicate from scratch, but we do already have plenty of existence proofs.
Also, self-replication is actually easier at the molecular level because all the parts are perfectly regular.
And we can think too, and you can see where is AI in that moment. :)
And ducks can fly with flexible wings.
And the Sun is a thermonuclear reactor.
You can find demonstrations of this type of things everywhere, that doesn't means that we can make them work with our materials, techniques and knowledge.
Nanotech seems to my like AI, always there, at just two steps... that's the type of problems that smell to vaporware from miles.
Somebody said that the reason there is no progress in AI is that once a problem domain is understood well enough that there are working applications in it, nobody calls it AI any longer.
Now if only we could re-program women to make iPhones. The late-night trips to the gas station and gravel store because wife has another petroleum and sand-craving would totally be worth it.
We know how to train the brain of a human being to learn physics (well not always), but that doesn't means that we understand how that brain really works.
The life if full of examples of things we know how to use but that we don't understand. Sometimes it's just impossible to control that things. Think in the stock market or economics. Chemist has been studying this type of things for centuries and you can see the result: there are chemical products everywhere you look.
What the post attack is the concept of creating nanomachines that doesn't degrade (Cosmic rays! oxidation! temperature!!) and do exactly what we want.
And about the "artificial DNA", it's something so simple... Really! the problem is in the emergent properties of DNA, proteins and environmental effects, this is where you loose control. Even nature has problems with that, life forms die, had cancer...
Self-replicating nanotech is only one kind. All you really need to be able to do is build nanomachines; whether they're built by identical nanomachines or by some big contraption doesn't matter to the end result.
"coating one millionth of a millimetre thick – 500 times thinner than a human hair – can be applied to virtually any surface to protect it against water, dirt, bacteria, heat and UV radiation."
You did not fund molecular nanotech and let billions of funding go to buying new buildings at universities and for chemistry research. Molecular nanotech has been excluded from funding since the 2003 start of major funding. When you buy your Ford SUV, do you send your complaints about it to Ferrari when you did not buy a race car?
http://nextbigfuture.com/2010/09/eric-drexler-ralph-merkle-o...
"Imagining self replicating nanobots or nano machines is ridiculous." The author does not seem to understand that all of cells of life are nano machines. If you insist on a definition of a machine being a collection of gears and bolts, then you must expand your definition.
Yep, and we can make those using bacteria machinery. But, it is actually pretty different when compared to some of the possibilities set out by nanotech folk. Viruses passively work in a system where their structure is intrinsic to the system. Most nano-techs can be better explained as catalysts, surfactants, anti-bodies, etc. Technologies and phenomenon long studied in existing fields.
I agree that it seems unlikely that we'll achieve nanobots that are more complicated than biological cells. An example of an unrealistic idea would be a remote controlled (or even autonomous) submarine diving through blood vessels, broadcasting what it sees to the outside world in HD quality.
Still, lot's of interesting things should be possible...
Such devices are already possible because the scale needed is mm not nano. For example the devices to do angioplasties are pretty big because blood vessels are pretty big. There is also a lot of commercial and research work being done on "remote control" treatment delivery, esp. for cancer treatment. (By remote control I mean externally guided as opposed to joystick control.)
Nano machinery is real, it exists, it powers the world, it's called biology.
The problem is that we seem to have a hard time translating our hard won knowledge of mechanical engineering to the nano scale, I expect some time will pass before the SF visions will become a reality but I think eventually we will get there. Figure another 50 to 100 years or so?
As long as we keep thinking in terms of conventional mechanics (wheels, gears, levers, wires and so on) it's going to be slow progress because we are trying to push down the size limit on our macroscopic ingredients without a real appreciation of what you can do with molecules 'as is'.
I think at some point the 'genetic synthesis' crowd is going to meet up with the 'nano technology' crowd and that's when we'll see some real action. The first little bits of progress in that direction have already been made.