Nano-nonsense: 25 years of charlatanry (2010) | Hacker News Reader

Nano-nonsense: 25 years of charlatanry (2010)

scottlocklin.wordpress.com

87 pointsandsoitis57 comments

"Part 2 of this volume is an exercise in theoretical applied science, a mode of research which aims to describe technological possibilities as constrained not by present-day laboratory and factory techniques, but by physical law." -Eric Drexler, Nanosystems, Appendix A.

Dude was very clear about his aims, and it is a shitty kind of strawman argument to dunk on Drexler like Scott Locklin is doing here. Physicists think like this all the time when considering upper- and lower-bounds to things. Drexler just had the temerity to put nice graphics alongside his speculation. It's not his fault if people got inspired. The upside of his work (inspiring dreamers to get into materials science, physical chemistry, even molecular/cellular biology) far outweighs even the worst downsides haters like Locklin can invent.

> We don’t even have micromachines. Mechanical objects on microscales do not exist.

Huh? Aren’t micromirror devices made of millions of microscale mechanical objects?

The first DMD page that came up for me is a TI DLP4500, which has about an 11mm diagonal and is about 1000 mirrors wide, putting the individual mirrors, and their tiny hinges at around 10 microns. What definition of “microscale” is the author using?

Edit: I didn’t realize that this article is 12 years old, but it doesn’t matter. DMDs had been in use in consumer products for almost a decade when this was written.

Yes, this is a weird argument and unnecessary for the main target of the article. There were plenty of successful microactuators, sensors, full microrobots, etc even when this article was written (2010).

Yeah, I agree that it’s tangential to the main point. Still, it makes me question the author’s credibility on the subject. If you’re claiming something won’t exist, and you’re overlooking a large amount of technology that does exist, I can’t help but think that you might have overlooked something in your assessment of the state of the art.

It is obvious that the author is referring to the sort of technology that Drexler and company were positing - a technology explicitly claimed to be so unlike anything then- (and now) realized that it would be radically transformative. At the time of writing, nothing even approaching a proof-of-concept existed (DMD devices certainly are not that), and this has not changed in the decade+ since.

You can, of course, suppose that the author has overlooked something, but there are others who have equally - and perhaps more carefully, by your standard - made the same points, such as Richard Smalley, who is quoted here.

When he says "even" he presumably means micro in contrast to nano machines.

Look at the sentence in context. It is obvious that the author is referring to microscale machines, not nanoscale. And the statement he makes there is simply false.

My point is that the case against Drexler’s concept does not collapse because of this one mistake.

Right, which I already acknowledged.

Well, you did indeed nominally acknowledge that this is only one detail, but then you continued by using this one technical error to cast doubt on the article as a whole! What were we supposed to make of that? There are cases where one apparently small issue is devastating (“who shaves the barber?”) but this is not one of them.

MEMS [1] is, I believe, the name you're looking for. Lots of cool detail on the Wiki page.

[1] https://en.m.wikipedia.org/wiki/Microelectromechanical_syste...

Ah yes. DMDs were just the example that immediately popped into my mind, probably because they were one of the first that I read about in any depth.

Nanotechnology was actually envisioned by Richard Feynmann in his famed 1959 lecture "there's plenty of room at the bottom".

I don't presume anyone would call out Feynmann for not knowing his physics. He even won a nobel prize for it.

https://en.wikipedia.org/wiki/There%27s_Plenty_of_Room_at_th...

> Really “material science” was a sort of fancy label for the chemistry of things we use to build other things.

Wrong, much of what material scientists do relate to the physical properties of the materials.

I know a few material scientists, and while they do have a small part of the lab that looks pretty much like a chemistry lab, most of their methods and goals differ from what happens in a typical chemistry lab.

They're adjacent fields and get into things like polymers and there's a fair bit of overlap. But metallurgy, for example, deals with a lot of things that chemistry mostly doesn't related to mechanical properties at different size scaales for example.

This 'nonsense' seems a lot more feasible now. We are now able to direct the formation of covalent bonds between two molecules using an atomic force microscope to position said molecules close to each other such that a reaction occurs in the desired location[0]. They show how NaCl films may present a solution the fat/sticky finger problem.

It's still a long way from drexlerian nanotech, but being unable to controllably form covalent bonds through positioning was one of the big arguments against drexlerian nanotech.

[0]https://www.nature.com/articles/s41557-021-00773-4

The human that wrote this is actually a enormous assemblage of nanorobotics albeit of the carbon variety.

That’s one of the more powerful counterarguments against “Drexlerism.” Nanotech already exists. It’s called biology and it has already spent a billion years exploring the combinatorial space of what molecular machinery can do. Indeed it has evolved a lot of very exotic molecular machines that do seem to take advantage of quantum effects in various ways.

We are grey goo, and are currently working to build rockets to go out and continue converting the entire solar system and eventually the universe into grey goo.

How is that a counterargument to Drexlerism? The fact that it already exists would seem to make it more likely that human engineering will leverage it.

Yeah but the human engineering you are looking for is mostly called biotech.

I think the argument is that there just isn’t a novel field here. Drexler tried to carve one out where none exists to make himself and his followers look like they invented something when they did not.

Nanotech is really just gloss on biology, materials engineering, or chemistry, all of which are well aware of quantum scale effects and use them.

Biology has a pretty limited set of building blocks to work with.

Drexler specifically wanted assemblers that could work with aluminum and that could do some extra stuff with carbon that biology can't do.

> Biology has a pretty limited set of building blocks to work with.

Not that small: https://www.researchgate.net/figure/Periodic-representations...

Proteins can bind and use all kinds of other elements and the combinatorial space is enormous.

>We are grey goo, and are currently working to build rockets to go out and continue converting the entire solar system and eventually the universe into grey goo.

In fairness, we are green goo and it's not like the universe was doing anything better with it's atoms. (We're also not likely to succeed much within the solar system, and almost certainly not likely to succeed outside of it.)

Nature/evolution is working on a macroscopic scale, too, for billions of years. Still, there was and is a lot of "combinatorial space" for human engineers to explore, like inventing the wheel, for example, with which nature didn't bother.

This was written in 2010. Has the situation—hype?—around nanotechnology changed since then?

Yes, the buzzword bingo has progressed and "ML/AI" and "VR" have become better moneybait words than "nano-" was.

Somewhat. At this point nano, 2D materials, topological materials, and other buzzwords of the past are either mainstream or dying. AI/ML for automated experiment design in high-throughput screening is the new hot stuff in experimental science.

I remember slapping the word "nano" in to the title of a research grant that had previously been rejected after much back and forth.

Approved and funded instantly.

Now that would be an interesting history to read! Are some of these occurrences documented?

Its not interesting at all, there are buzzwords and buzzwords get you paid. The secret is knowing the words they need and maneuvering them into the title.

I dont doubt if you look at the blurbs for an annual intake of YC you will see the exact same effect.

I am personally working the very first reimaging of the traditional <insert market place here> with a distributed no code ai blockchain approach; that puts mobile, web and API customers first.

Please stop contacting me for funding rounds. Due to overwhelming demand I have had to close funding. However, in exceptional circumstances or if you have a very large offering then I may reopen funding.

Some call it "research grant bullshit bingo" - you massage the grant contents till you hit the currently necessary bingo combination for getting your grant past gatekeepers of money.

In similar vein is padding grants with equipment (including software licenses) that is often secondary to the needs or sometimes even was already available, but which would in turn enable you to work more efficiently on 20 more projects and be passed to institution after grant ended, helping other people.

Grant review and acceptance is quite often a question of what's fashionable and thus might get unspoken extra points in evaluation.

Modern nanotechnology is a multi disciplinary approach using effects on the nanoscale for new technologies. Therefore people from different fields come together to synthesis, simulate, analyse etc. nanoparticles/systems. Of course in the media nanotechnology is this big catch all word, where everybody immediately thinks about self replicating nanomachines - this is still fiction. Eventhough nanotechnology has not fullfilled this science fiction promise of atom precise manipulation the result are still astonishing. Just to name a few: Computerchips, memorychips, displays, catalysts, composite materials, surface coatings…

I think the argument is that the parts that work are just rebranding of chemistry, physics, and materials engineering.

Always has been

Isn't that basically what "multi-disciplinary" means?

Additionally, scientific niches rebranding themselves is a thing that happens, for examples when Computer Science rebranded itself from Mathematics and Electrical Engineering.

The parts of chemistry that work are just rebrandings of physics and the parts of physics that work are just rebrandings of math. Nanotech is a subset of physics with connections to solid state physics, quantum mechanics, materials science, optics etc. I don't get why people say nanotech is charlantry when the semiconductors chip manufacturing industry is top down nanotech manufacturing.

My freshman year of undergrad I took a nanoscience course as an elective. Through that, i found out that the department was in the final stages of creating a Nanoscience major, and since the prerequisites lined up with my primary major (engineering) I decided to go for a double major.

If I could do it over, I would stay away. The subject matter was fascinating, but was also a discontinuous hodgepodge of nano-related topics from physics, chemistry, and biology.

Things may become more organized as the program matures, but looking back it felt more like a 4 year exploratory to prep for grad school than a Bachelor of Science degree.

Not my field at all but I read this piece and an interview with him [0]. I don't know, this guy comes across as an arrogant asshole to me. I struggled to finish the interview.

[0] https://niccolo.substack.com/p/the-troms-interviews-scott-lo...

What's the best nanotech achievement so far, though?

Other than computer chips and advances in microscale machining (not yet nanoscale, but it takes time to scale down the manufacturing capabilities)?

If we count microscale advances then projector displays, fully optical dynamic switching in MPLS networks, all sorts of sensors including accelerometers in phones etc.

You are stretching definitions. Fact is, we don't have swarms of self-repairing micro-robots doing things. That was the main advertised output of nanotechnology. We don't have it.

Does laymen overhyping a pie in the sky part of future, a "this is what we know is possible, but we're barely on first steps to get there" goal, disprove progress towards it or successful deployments?

Is R1/XCON accidentally restarting overhyping of AI leading to AI winter responsible for AI Winter? Does AI Winter disprove that a single program utilising classic AI techniques in middle of AI Winter paid back all ARPA/DARPA funding for AI?

No in all cases. The self-assembling nanobots aren't a short term goal, and pretty sure they were never seriously promised as short term goal. It's like complaining just after Lilienthal's death about lack of regular rocket service to Mars, because aviation enthusiast also espoused that it will be possible in the future.

Come on now, that's a straw man. If nanobots were neveer the short-term goal then maybe it would pay off to not ramble that enthusiastically about them (and a good amount of other just as impossible-sounding technology achievements)? Or at least not try and pass it as serious science? Or if not even that, not sell a book selling sci-fi to the masses while dressing it as serious science when it clearly isn't?

Obviously science very rarely makes huge leaps. It's also very obvious we need a good amount of progress in materials science before even the theoretically-possible notion of microbots becomes viable.

That's all quite fine. But I join the author of the article in mocking people who babble enthusiastically about the far future and sell it as science. And then even have the gall to use complete non-sequiturs like "you can't disprove me!".

So you see, I don't disagree with you as far as I can tell. But I still do disagree with the "scientist" that the author's article is referring to.

Did Drexler predict that we would have swarms of self-repairing micro-robots by 2022? I don't know the answer to that, I'm genuinely asking the question.

Like what's the timeline for nano-technology as predicted all those years ago and are we on that timeline?

I haven't seen the author of the OP article talk about timelines. It was about skeptics not being able to disprove that we will NOT have that at some point.

Which I found to be good criticism because many charlatans operate exactly like that: (a) make a nebulous claim, (b) get called out, (c) pull the "you can't disprove me" card while OBVIOUSLY a good scientific discussion should go the other way: the claimer should prove their claims.

That's why I liked the OP article where the author called Drexler a sci-fi writer.

Read the rest of his blog; you'll probably find your opinion reinforced.

However he's often correct when others are not; which many feel justifies some arrogance.

He most definitely is an arrogant asshole. The question is: is he right?

I dont know: both guys seem to enjoy playing the asshole for laughs. The picture at the top suggests that he could really complete the look by pairing a full suit with a mohawk.

> We don’t even have micromachines. Mechanical objects on microscales do not exist... I have already worked for a “millitech” company which was going to use tiny accelerometers to do sensing stuff in your cell phone.

I don't see any definition by which MEMS sensors are not 'micromachines'.

Drexler's book is a great piece of science fiction. The kind that goes and inspire people to try to achieve some very hard but technically possible goal, with great reward.

But it's not a science work at all.

(And then, this article seems to be outdated on all specific complaints.)

Very smart people easily fall for stuff that merely sounds smart but is actually nonsense.

It gets worse when you have a whole bunch of them on a bandwagon since how could so many smart people be wrong? The answer is: because they all think that.

So, he's saying it's just like quantum computing? duck

So no The Diamond Age anytime soon? That was a really fun read.

https://en.wikipedia.org/wiki/The_Diamond_Age

People joke about how hard it is to exit vim but I don’t think I’ve ever been able to exit nano without just closing the tab

This is about nanotech not about nano (editor).

I think the comment you're responding to was nano, (the joke.) ;)

Reply on news.ycombinator.com