• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed

  • Powered by FeedBlitz

« Technology Works | Main | Tomorrow's Gold Mine »

April 05, 2005


Feed You can follow this conversation by subscribing to the comment feed for this post.

jim moore

I don’t think that you have to wait for a nanofactory to do a Design Wiki. You can do it with today’s rapid prototyping machines. There might be a real business opportunity renting time on today’s expensive machines to hobbyist then mail them the fabricated object.

This could allow us to baby step our way to responsible nanofactories.


People always talk about how nanofactories can make more nanofactories, or how prototype fabricators could be enhanced to make more fabricators. But they don't talk so much about what else the nanofactories could make.

What can a two pound desktop nanofactory make? Can it make food? Can it make a house? A car? Can it make clothing? Can it make a TV set? Bookshelves? A reclining chair? (I'm looking around my living room!) Can it make solar panels for my roof so I can live off the grid? A robot to climb up there and install them for me?

It's easy to just wave hands and say it can make anything. But let's try to be realistic; suppose it makes diamondoid with enough variety to replicate itself. What else can it make that is useful? I have visions of this thing popping out unlimited numbers of six inch tall diamond figurines. But I don't need a lot of hunks of diamond cluttering my desk.

Mike Deering

Hal, check this out:


Chris Phoenix, CRN

Hal, good question. The answer is that any large-scale diamondoid product will be mostly empty space or inert mass. Therefore, inflate them with air or water, or unfold them like a pop-up book.

With materials 100 times as strong as today's, you could build a 1-pound hang glider (with or without a motor), a quarter-pounder bicycle or chair, etc. I think you could do a five-pound car; the engine and furnishings would weigh almost nothing, and the structural materials could be made a lot lighter once solid sheets of metal are replaced by nanoscale trusses.

It's easy to design a nanofactory that can make products bigger than itself. So a five-pound car would take 2.5 hours.


Brett Bellmore

I've been dubious all along about that inflating/unfolding concept. Sure, you could do that, but it would contribute immensely to the design complexity of a lot of products. I figure the nanofactories will probably be built large enough to accomidate uncompressed products.

For one thing, manufacturing low density products in a low density form would allow the assemblers in the factory to operate at a lower duty cycle, easing cooling requirements.

I also don't see the nanofactory, given it's requirements for power, cooling, reagent feeds, as a "desktop" device. It's the sort of appliance that would be plumbed into your home systems, not free standing.

The way I envision it, you'd have unit tiles, each complete in itself, and capable of extruding product right out to it's edge. To make larger products, you expand the array, perhaps automatically, and they cooperate.

A small unit, for instance, might BE your desktop. Larger units could tile the floor, or the wall. When not actively manufacturing product, the nanofactory would extrude a surface to present to the world, anything from marble to wood, carpet to a computer display.

In fact, one could envision a tiled surface on the wall, normally configured as a 3d display, which would display the products you were considering, and when you directed it to manufacture one, you'd see that object move towards the surface, and seamlessly emerge, as though the surface of the nanofactory were a hole into another world, and not an array of nanoscale assemblers.

Tom Craver

Hal -

I don't see any reason why, beyond a first few versions, a nanofactory would need to be limited to diamond. But suppose it were only able to work in carbon - diamond, graphite, nanotubes, etc.

Obviously you can make stuff like dishes, glasses, bowls, window glass.

Maybe a weave of diamond and carbon nanotubes could be used to make simple tools able to stand repeated impacts - saws, hammers, knives, axeheads.

Solar collectors - Chris has proposed thermionic generation of electricity. Another approach might be to have micro-scale chemical factories directly using focused sunlight and heat to drive chemical reactions that produce fuel, or simple foods (sugar, starch) to feed to animals. (The hardest part is splitting water for hydrogen. I found a web site that suggests this might be possible at ~1700C in a solar furnace, with a membrane permeable to hydrogen but not oxygen - it may be possible to make that membrane out of diamond. With vaccuum insulation to keep heat in the reaction system, and recycling of heat from extracted hydrogen and oxygen, this might be made fairly efficient and compact.)

Edison's first lightbulb used a carbon filament - assuming we can make carbon conductors, and reinforce a thin carbon filament with diamond to make it survive the repeated thermal shocks longer, a lightbulb isn't out of the question.

A carbon internal combustion engine is probably not feasible, but a diamond steam engine might be - burn hydrogen and oxygen into or under water in a vacuum insulated boiler to produce steam.

A flexible cloth-like material might be made of tiny rings of diamond, linked together like chainmail. If you make layers of it thick enough, it shouldn't be transparent, so clothes might be made of it. Make a tent - or make bags, fill them with dirt, and pile them up to make walls - a house. Weave it tight enough use several layers to make a bag large and light enough to inflate with hot air - go flying. Or weave the rings into long strands - thread, string, rope.

Perhaps a mix of diamond and graphite (non-conductor and conductor), finely patterned, could act like a semi-conductor, for electronics. If not, perhaps micro-scale vacuum tubes would be possible. And of course, there's Drexler's rod-logic computers, and analog mechanical schemes using gears and such.

Chris Phoenix, CRN

Brett, what I'm talking about here is a much simpler unfolding concept than my convergent-assembly nanofactory required. I agree that would have been difficult to design for. But here, I'm just talking about macro-scale unfolding, nothing more complex than the way a vinyl pool toy can be folded into a relatively small box.

I really like your image of integrating a display into the surface of a planar assembly/extrusion factory, so that you can see the product as it's forming. One problem: the user might get confused and try to remove the product while it was still half-formed!

A primitive nanofactory would indeed need to be plumbed-in. But a more advanced design, using mills for the chemistry, could be air-cooled in a kg-scale version.

A planar-architecture nanofactory should be able to extrude at 1 cm/s or better (scaling analysis says this is independent of block size). Even if your product is quite sparse, the synthesizers will probably be the limiting factor, not the extrusion. And moving small blocks for meters through the nanofactory might add a non-trivial energy load.

Collapsibility will be a significant advantage in large products. Don't lock up your bicycle--stick it in your pocket.

Yes, you could build a planar nanofactory that could extrude a car--but you'd have to keep a car-sized space clear in front of it. If you extruded a large brick in your kitchen, carried it outside, and attached it to your garden hose, then you'd only need space for the car in your driveway.

If products were collapsible, then you could store an airplane, a motorhome, several cars, and a yacht in a corner of your garage, inflating them as needed. (Ballast for the yacht might be a problem, unless it used active trim.)


Brett Bellmore

"And moving small blocks for meters through the nanofactory might add a non-trivial energy load."

Possibly would. That's why I envision the separate tiles being self-contained, aside from data interchange with the rest of the array. Each tile would be served by it's own assemblers and chemical mill.

michael vassar

Mike: I imagine that you will get everyone to agree on how to use nanotechnology a few seconds after you can get them to agree on evolution, morality, and which culture is "best".

Karl Gallagher

you can get them to agree on evolution, morality, and which culture is "best"

I'd hate seeing an attempt on that. Reminds me of the start-up CEO I worked for who kept muttering "There's more than one way to get unanimity on your board of directors" as he arranged the shareholder vote.

Tom Craver

CRN keeps mentioning the idea of "approved products". But consider what that would really turn out to mean - every product would have to pass through a gauntlet of approval influenced by every interest group that can muster the power to get representation of their ideas into whatever agency or board does the approvals. It would be an explicitly political process, and is a prescription for paralysis.

Any food product design would have to be reviewed by those who want to protect the livelihood of traditional farmers, those who want to end use of animals for food, those who fear artificial food, those concerned about bad dietary habits, those who have an already approved product and don't want any competition and claim to have patents/designs that bear on the decision, those who want to make sure the product has no unexpected side effects, contains no molecules that might count as recreational drugs, isn't part of a plot to poison people, etc., etc. Even groups that have no direct interest will weigh in on every design, trading their support to other groups for their support on other designs.

Getting a new design approved will be like getting a new law passed - and the (presumably government) agency tasked with approval will have power that is the envy of Congress. In fact, Congress may just decide to take over the job themselves. :-)

One result will be a thriving black market in unapproved (illegal) designs produced by amateurs, designs stolen while still undergoing approval, designs that claim to meet some common demand (e.g. cancer cures) but really don't do anything. And of course, designs for anything illegal - drugs, weapons, etc.

All the best and worst new designs will be available on the black market, and many (if not most) people will become scofflaws - not merely breaking the law, but considering laws controlling designs a farce. Likely there will also be laws against "promoting unapproved products" - i.e. communicating what is good and what is bad - leaving consumers to take their chances.

Janessa Ravenwood

Tom: what you said (I agree).

I don't worry about coopertive international regulation of the CRN variety - that's a dream of CRN's that stands no chance whatsoever of becoming reality. The situation you described sounds about like what I (and others here) have been envisioning as actually coming about. The REAL question isn't "How do we prevent nano-anarchy?" but "What do we do about nano-anarchy WHEN it arrives?" Note that I'm using the term "nano-anarchy" as "MNT freely available outside of strict regulations."

Chris Phoenix, CRN

Well then, Janessa, what *do* we do about nano-anarchy when it arrives? Assume for the moment that you're right, and there's no chance of putting together a structure that can prevent uncontrolled nanofactories *and* that would preserve even a modicum of liberty and privacy. Then we seem to have a situation in which an individual, with a modest amount of effort, can kill tens of thousands of people, in their own society or elsewhere.

How can societies and technologies worldwide be structured so that 1) few people take it into their heads to do that, or 2) those that do can be preempted while still preserving the above-mentioned liberty and privacy, or 3) some factor makes it significantly harder to deploy nano-built weapons than it currently appears that it will be?

This is not a rhetorical question. It's very practical, and meant to be fair, not argumentative. If there's a better way to look at the problem, please suggest it.


jim moore

Biotech is a much more immediate threat than nanotech. Many of the dangers that CRN is concerned with are already present and growing with the rapid developments in biotech. If we are unlucky enough we will not have to deal with irresponsible uses of nanofactories.

Janessa Ravenwood

Chris: I understand, it's indeed a valid question and believe me, If I had "The Answer" I'd be happy to give it to you (how’s “42” grab you?). Here's the problem - I don't and I'm not sure there is one. I personally see the Wild West coming back, so to speak. File-sharing has already primed people and society as a whole for this sort of behavior (and the new P2P's just get more efficient and are deliberately harder to police with each new generation of design). Essentially I think we’re going to see some “rapid social evolution” coming up, some good, some not, and people will have to start taking personal security much more seriously. Updating your body with “virus-scanners” may become a matter of life and death, not just suffering lost files like if your PC gets a nasty virus.

And don’t count on lawmakers to be rational about this. Just about without fail, Congress’s reaction to any new form of revolutionary technology that is even slightly threatening is either “Ban it!” or “Strict Regulations!”, both of which will just encourage the growth of the nano-underground. In other words, plan on government making the situation worse, not better. I actually see this coming down to the actions of individuals and organizations to make the real difference – though the outcome of this could be either good or bad, or a mixture of both (likely). While all this is going on, Congress will be busy trying to talk the problem to death and the feds will be making mostly ineffectual raids on the nano-underground and boasting that they’re “winning the war on illegal nanotech.” Sound familiar?

Chris Phoenix, CRN

It's not just virus-scanners, it's bullet-scanners and land mine scanners and shigawire scanners (very thin strong wire from Dune)...

I hope you're wrong about virus scanning being a life or death issue. People can have quite perverse responses to actual life-threatening risk.

If I didn't know you were so pro-American, I'd start to think you were criticizing our government. :-) Yes, the "War on X" rallying cry is certainly overused (which is not to say X isn't bad, just that declaring war on it isn't always the best approach), and its use seems correlated with attempts to curtail our rights and liberties. But not too much, because then they might win the war and they'd lose a campaign issue.

But do you really see "the feds" surviving as an institution in the kind of wild-west scenario you're picturing? I'm not sure any government as globally unpopular as ours would last longer than it took someone to innovate a single unexpected weapon (and build a million copies).


Tom Craver


I think it's summed up in that Franklin quote about those who give up a little essential liberty to obtain a little temporary safety deserving (and getting) neither in the long run. If we pick liberty in the first place, we can be certain that we will never have total safety. As today we have to be alert and aware of computer viruses and scams, and take steps to protect ourselves from them, so we'd have to behave if we chose liberty - individually, and at the level of government.

Unfortunately, I'm not very optimistic that things will turn out that way - US citizens have become awfully compliant, dependent and safety-oriented. Your "approved products" approach will likely be a big hit with politicians and the elite opinion shapers, each of whom will envision himself setting the agenda for the approval agency.

Karl Gallagher

Chris, that's a fair question. As one of the folks who thinks the CRN goal is completely impractical I'll take a swing at it.

The approach I visualize is complete decentralization for defense. This means extending the 2nd Amendment to personal-scale nanoweapons (or equivalent for other countries). Everyone would be responsible for maintaining a basic level of defense against threats. Not having an acceptably up to date level of "virus protection" would be treated the same way as having an active case of TB is, i.e. "this person is a negligent hazard to his neighbors." This shouldn't be much stress on individuals since they'd already want to have their own protections in place. In current terms, we'd not only have strong punishment for spammers, but also have short prison terms for anyone negligent enough to let his PC get turned into a spam-zombie. The basic nano-gear for households would resist just about any nano-weapons. We'd want our gear to resist entropy in the form of disease and broken stuff. Weapons are devices for increasing entropy so they have to overcome the basic life-improvement tech before running into any specific defenses.

This would be combined with the everyone-watches-everyone scenario in Brin's Transparent Society. Anybody planning an evil act would know that anybody could be "surfing" his neighborhood looking over his shoulder. That makes individuals the first line of defense on detecting attacks, whether infiltrated from abroad or created domestically. Once one is detected the alert can spread and local volunteers can counter-attack.

So there's no point trying to create "restricted" nanofactories. The restrictions would be broken quickly by government professionals and amateur hackers anyway. What keeps us safe is having a nano-spy sitting over every factory watching what comes out. If it looks scary, the neighbors pick up the phone and say "what the heck is that? And why are there 4096 of them?" If they don't get a good answer they sound the horn call of Buckland (or local equivalent). If it's a big fight they get reinforcements from the next town--and eventually the Army.

Would this be the same society we have today, preserved in the face of changing technology? Heck, no. The federal government would be much less relevant to daily life (and busy with external threats). We'd replace the modern casually obnoxious behavior with the "polite society" that concentrates on not giving anyone any motive to shoot you. Privacy would discussed only in history classes. Various groups would form local communities dominated by their religion/ideology/ethnicity and pressure "others" to move out (limited by the fact that the whole country can watch them doing it, so harsh behavior would be slapped down fast). The borders with nations not living by this system would be deep zones of observation to detect infiltration, inhabited by the expendable (see The Diamond Age for an example).

On the other hand "an it harm none, do as thou wilt" will be a general rule nationwide. You can say what you want, dress as you please, take any job you want, move anywhere in the country, organize your family as you please. So I consider it a free society, one I'd be happy to leave to my kids. Various other people would hate it. But I think it would be a stable solution.

So there's my vision of how to survive in "nano-anarchy" on the world scale. I don't claim it's the only or best solution, it's what I've come up with in thinking on the problem. The virtue is that it can handle a breakdown at any one point without losing the rest of the nation in the process, while still maintaining a reasonable free society. The vice is a minimum loss rate of people lost to the start of outside attacks and innocents who were mistaken for the next Tim McVey by their neighbors. It's a reasonable compromise between the technological constraints and my dynamist/libertarian beliefs.

I'd be happy to see more alternatives. A key factor I'd be judging them for is robustness. A good solution has to handle a failure without the whole system collapsing. That's my problem with the CRN's proposal for international regulation. One nation negotiating in bad faith, or one black marketeer distributing hacked nanofactories, destroys the benefits of the scheme. That's an unstable state, ready to collapse into a stable state. One stable state is universal dictatorship, something I wouldn't live to see. Another is the extinction of the human race. Let's keep working to find better ones.

Brett Bellmore

"The federal government would be much less relevant to daily life"

Problem is, the government wants to be relevant to our daily life. In fact, it insists on being relevant to it. Not being relevant to our daily life is a threat to it's very existance.

Aside from that, while Brin's survailance society might come about as a consequence of some trends, it's not going to be something people chose voluntarilly. People like their privacy. That might change in another generation or two, (When all the shy people have died off because they couldn't find any place private to make out.) but you need a system today's people would adopt.

So, how to do this?

First, we'll need a system of trusted sources for designs. There would have to be multiple sources, no single bottleneck, in order to avoid any bottleneck becoming censorous, encouraging a black market.

Second, we need a way to encourage use of trusted sources. I think you could do that by "poisoning" the underground distribution pathways with sabotoged designs, to the point where anyone who used them on a regular basis, and wasn't quite careful, would come to, not a bad end, but serious embarassment at the least.

So a 14 year old out to cause trouble surfs the web, and downloads the design for a "fully obedient T-2 Terminator", and runs one off. And it hauls him off to his parents, or if given sufficiently nasty orders, the police.

This will cause even the bad guys to band together in their own trusted source communities... Which makes them easier to catch.

I'll develop this idea more later, but I've got places to go.

Chris Phoenix, CRN

1) It's very hard to know what a nanofactory is spitting out. Probably impossible... unless the nanofactory is technologically limited (e.g. uses only transparent file formats). And even then, the difference between a harmless tool and a weapon will be largely software.

2) If you postulate the technical capability to watch everything, then it can be made very difficult for a nation to negotiate in bad faith.

3) The idea of buffer zones between regions is 2-dimensional thinking. Do you intend to prevent all space access and air traffic, and filter the oceans?

4) It'll be hard to tell where an attack is coming from. If the standard way to respond to an attack is with a counterattack, then one case of mistaken identity will set off a feud. This does not seem stable at all. "Armed = polite" only works if it's easy to see the source of attacks.

5) Out of pure convenience and self-preservation, people who were good at nanotech would have a strong incentive to manage others' use of nanotech. If my neighbor isn't protecting me, why should I let him have a nanofactory at all? This quickly regresses to a central authority.


Tom Craver

We need a list of potential malicious MM users:

1. Insane-genius supervillain
2. Governments
3. Terrorists
4. Greedy criminals
5. Emotionally driven criminals
6. Ego driven virus writers, system crackers, etc
7. (Who did I miss?)

Looking through the list and thinking about what motivates them, I'm thinking the danger of malicious uses of MM may be over-rated, EXCEPT for governments (war, draconian police states, etc).

E.g., terrorists attack civilians because they lack sufficient power to effectively attack those in power. That changes if you assume the ability to make specifically targeted anonymous attacks. Terrorists would likely shift to guerilla tactics - and if they succeed, they will need to negotiate for peace in order to safely take open control of their "liberated" territory.

Karl Gallagher

Number 7 is "well-meaning incompetents", the free-lance kind (as opposed to 2).

Some terrorists might take advantage of MNT to shift to guerilla tactics, depending on their motives. The Provisional IRA could be a good candidate for that. But these days "terrorist" tends to be a euphemism for "islamofascist militant" and that's somebody who considers killing infidel women and children a worthy act in itself. Al Qaeda would've used MNT on 9/11 without changing targets.

Tom Craver


Terrorism is actually an expression of impotence - it's a frustrated act of revenge, taken by someone who has no chance of getting his way by force and very little hope of winning through his terrorist acts.

Someone who obtains a highly effective nano-weapon, such as we're considering, no longer needs to be a terrorist, regardless of his motives - he can attack directly and effectively.

BTW - I strongly suspect that a major part of the image of islamic terrorists as motivated by killing infidels is hype - both from the terrorist leaders, and from our own leaders. The terrorists leaders use that image to inspire unity of purpose among the terrorists, while our leaders to the same, for us.

If some enemy occupied the US, in order to insure access to our natural resources (or perhaps to suppress our appetite for oil they wish to consume), would even radical red-neck Christian terrorists (oops, sorry, they're "freedom fighters" if they're on our side) really be primarily motivated by desire to kill infidels?

amanati mostafa

In The Name Of God
From:The Manager,
Optimal Industry Time Company,
Dear Company:
This is proposal cooperation and I want you to consider it very important
I am request models of Nano product your company.
If you accept this offer,I will appreciate your timely response.
Yours Sincerely,
Mostafa Amanati.
Optimal Industry Time(Iran)

The comments to this entry are closed.