You are not logged in.
Pages: 1
Hi Everyone,
When reading this forum every now and then someone says, "and nanotechnology robots will solve THAT!"
I just want to say that independent nano-technology robots are a long, long way off.
--- Rant mode on. ---
I read "Engines of Creation" by Eric Drexler, and it felt more like a religious tract than something with real science.
There are severe problems with the independent robots view of nano-tech that most people ignore.
- POWER: I've hand people tell me about nano-bots that swim thru my blood stream zapping viruses. Or little machines that burrow thru the rock finding useful metals, etc. "Where do they get their power?" I ask. "From your bodies heat!" they tell me. For a heat engine to work you need a heat differential. And one side of this microscopic bot is the same temperature as the other. Don't people know about the laws of Thermodynamics?
Others talk about putting an antenna on them and receiving power from radio waves. I ask them how an antenna a few atoms long are going to absorb a wavelength measured in meters. A shrug; they don't know what I'm talking about.
Other people talk about using a system to convert light into mechanical motion and using a rack and pinion system and clockwork springs to store energy. When I ask them to sketch out how this system would work I get shrugs. Light (at least that found underground or in the body) is not powerful enough to knock atoms about. It gets absorbed into the electron shell. So a system to mechanically covert this to motion in a rack and pinion is not trivial. Ah, well, at least the second system is not physically impossible in theory.
A person said that the nano-bots could take in sugars (from our bloodstream) and burn them producing wastes which are dumped back into the bloodstream. Ok, this could actually work. You will need enzymes to act as catalysts, ways of storing this power and systems to clean out the gunk produced by burning these substances. All of these systems push up the size of the nanomachines (which limits their mobility and usefulness) but again this system is possible.
- SMALL IS STICKY: People talk about the nanobots grabbing this atom here, that one there and assembling useful stuff. The problem is that those atoms are stuck together. Really tightly in some places and it takes a lot of power to wrench them apart when you are few hundred atoms in size. And when you grab one atom from a molecule both halves now want to stick to things. The piece you broke it away from, other stuff drifting around, your machine. When you put it on a half finished robot that robot has 'sticky' surfaces all over. What will stop some thing else coming along and attaching itself on your half done machine? Now you are not just building a machine you are cleaning random, chaotic debris off of it as you try to build it.
- SMALL IS DIRTY: On the atomic scale you have all sorts of elements and molecules everywhere and many of these have sticky surfaces. Glom some of them on your machine in inconvinient places and it stops working. When I ask enthusiasts for nano-technology about this problem I get suggestions like windshield wipers to brush off this stuff, or suggestions that free radicals and 'sticky' atoms don't exist in large enough numbers to be a problem. (Even tho your machine is creating such things locally.) Others talk about smaller nano-bots crawling over the bigger ones removing bad atoms.
To clean off this sticky crap is NOT an easy process. It will take a fair bit of brain power to know how to approach this particular jam, or that molecule's edge, wedged in there.
- PROCESSING POWER: Where do the brains for these robots go? People talk about using rods in carbon nano-tubes to store information. (If the rod is up 3 atoms it means something different than if it is up 4 atoms.) No reason rod logic can't work, but tho very small to us, it must be gigantic on the atomic scale. Nano-bots will have to deal with a very, very complex dirty world. Much of it will be dark and if you say, well put lights on it, the whole question of power comes up again. (And then you need the processing power to figure out that visual information in a world of difraction and interference rings.)
These machines will need a TONNE of processing power, to figure out what is in their environment (sensors, pattern recognition), will have to make plans in a chaotic environment, will have to move over potentially sticky, magnetic or electrostatic terrain. They will have to coordinate with other nano-bots for big projects, etc. And as the processing power goes up the nano-bots get larger and larger. Their power requirements go up. There are ever more surfaces that random stuff can stick to.
I've had people say that each nano-bot will have low processing power, but they all will find each other, link up into a super network, figure everything out, program each tiny bit of themselves with detailed instructions and then each bit will diassemble and go out do its thing.
Ignoring how difficult all this is I ask a very simple question. I ask how they will link together and exchange info? "There will be holes so that the rod logic can push rods into each other." Well, what if there is dirt or stuff stuck to the side of one so there is not a tight fit? "No problem, we will just put windshield wipers on them!" What will keep stray atoms from getting into the holes? "No problem, we will just put little powered hatches on them!" etc.
Let's just say I'm more frightened by the iron death of the universe than I am by "grey goo".
I ask that anytime someone wants to fix a tough terrafoming problem with self replicating nano-robots, they either:
A) Discuss how the above problems would be solved.
B) Admit that this is for the FAR future.
or
C) At least wait until we have experience with swarms of insect sized robots doing some useful, productive work before you magically solve problem with robots that have an enviroment that is orders of magnitude more complex.
---- Rant mode off. ----
For an interesting story about insect sized robots you might want to check our James P. Hogan's "Bug Park". It is a lot of fun & has some good lessons for what happens when you shrink scales.
Regards, Rick.
Offline
These are probably silly and impossible ideas for power generation, but I'll throw them out there anyways.
1. Why not just implant a charging station in a person. Robots come in to fill up their batteries. Course can a battery be made that small? What kind of charge will it hold if it can? Then you'd have a temperature differential, just plug yourself into an outlet, or maybe make power like one of those wristwatches that generates power when you move around?
2. Just park on an arterial wall and stick out a turbine. Let your blood flow generate some power.
3. Maybe have them migrate to the surface of the skin and run your heat differential that way.
Feel free to tear these ideas to shreads. I don't care, and they're probably dumb anyways.
Offline
These are probably silly and impossible ideas for power generation, but I'll throw them out there anyways.
1. Why not just implant a charging station in a person. Robots come in to fill up their batteries. Course can a battery be made that small? What kind of charge will it hold if it can? Then you'd have a temperature differential, just plug yourself into an outlet, or maybe make power like one of those wristwatches that generates power when you move around?
2. Just park on an arterial wall and stick out a turbine. Let your blood flow generate some power.
3. Maybe have them migrate to the surface of the skin and run your heat differential that way.
Feel free to tear these ideas to shreads. I don't care, and they're probably dumb anyways.
Hi X, everyone.
Actually all of these are smarter than some of the standard answers.
Chargers and batteries:
You would not want to have one charger (too hard for nano-bots to find) but billions spread thru out the body. Being immobile they can be a lot bigger and have more proccessing power. When they spot a nano-bot floating by they grab it some how and charge it up.
Very small batteries can be made, but you only get so much energy out of each pair of atoms that react. Even if you have thousands of atoms, the nano-bot won't be able to swim about and think all the time. It would have to drift, waiting for just the right thing to bump against it and then a processor clicks over, some rod logic slides, and maybe it is in the right place to spend its current and do something.
Immobile Turbine:
This is actually practical I think. By not being mobile your 'bot is in a much simpler environment. (It requires FAR less thinking to sit still than go wandering.) By being immobile it can be larger than the nano-bots that are supposed to fit thru the smallest capillaries or even pass thru cell walls. With a constant power source from your turnbine it can be constantly thinking and thus always ready to do real work.
We won't get any of the nonsense about nano-machines going into every living cell and snipping out bad DNA and replacing it with good DNA. (How would they know if they miss a cell somewhere for one thing.) But little factories studded on blood vessel walls could actually do some good. They could use chemicals to communicate with each other and likely do some real work. Imagine dispensing drugs only in the areas that need it. Very good idea X.
Migrate to Skin:
Moving in water at the nano-level is like trying to swim in cement. It takes a LOT of energy. I don't see how the little bots would be able to store enough power to move any distance to say nothing of trying to squeeze around skin cells. Also, the bot is so small that the heat differential is tiny. (One side is at 36.7 degrees and the other side is at 36.69999999382 degrees C.) We won't be able to get any useful energy in any sort of reasonable time. Lastly the skin constantly sheds skin cells. I think this is your least practical offering.
Warm regards, Rick.
Offline
Nanotech again? Here's one for you. One will occasionally hear about a remarkable machine that somebody has in mind, in which a lever (yes, a lever) is made using a single molecular bond as the lever arm, the lever itself consisting of three atims or something like that.
The news that molecular bonds spontaneously break from time to time seems to come as a shock to some of the true believers.
Offline
Nanotech again? Here's one for you. One will occasionally hear about a remarkable machine that somebody has in mind, in which a lever (yes, a lever) is made using a single molecular bond as the lever arm, the lever itself consisting of three atims or something like that.
The news that molecular bonds spontaneously break from time to time seems to come as a shock to some of the true believers.
Hi Joseph.
LOL !
That is a new one to me.
Warm regards, Rick.
Offline
Yeah, this is basically Smalley's explaination. Simply put, 'nanotech' is relegated to geometric organic tranformations (namely DNA/RNA). No one has discovered a way yet to make nanotech on the scale that Drexler talks about. It's simply not mechanically possible, as far as we know. (Always have to say "as far as we know" with science.) This makes any self-replicating or self-propogating nanotech pretty much currently impossible or infesible. If you ever read Smalley's arguments about nanotech he basically points out that nanotech is organics, nothing more (in fact, I would argue that if we could have these magnificant self-replicating intelligent forms then evolution would've brung it to us a long long time ago).
However, this doesn't mean we cannot have a small compact ready made solution to "make stuff" on Mars. It just means nanotech isn't the way we'd go about doing it. We already have prototype Rapid Prototypers (see www.reprap.org) we already have knowledge with smelting technology and other industrial tech on the small scale (see gingery machines). All that is required is a meld of the two technologies on a compact industrial level, and you can conceptually make anything with it. It's super high tech stuff, though, and no one that I know of has attempted to meld the two technologies together into one coherent technology that can... well... make stuff.
So instead of obsessing about the nano level, perhaps an inspection of the micro level is in order.
Some useful links while MER are active. [url=http://marsrovers.jpl.nasa.gov/home/index.html]Offical site[/url] [url=http://www.nasa.gov/multimedia/nasatv/MM_NTV_Web.html]NASA TV[/url] [url=http://www.jpl.nasa.gov/mer2004/]JPL MER2004[/url] [url=http://www.spaceflightnow.com/mars/mera/statustextonly.html]Text feed[/url]
--------
The amount of solar radiation reaching the surface of the earth totals some 3.9 million exajoules a year.
Offline
BTW, it should be stressed that robot swarms do not actually require much intelligence at all. Ants can build highly complex structures and we pretty much have their neurology mapped (we can make antfarms in computers). Insect colonies have very little intelligence for each individual, but the colonies as a whole have their own sort of neurology that has a higher order of computing power.
So on the microlevel I do think robots could be utilized to do things. Say we need a road paved or something, I think a bulldozing robot could do that all by itself. Most tasks can be simplified if you know what you're doing. Want to build a wall? Make robots that lay bricks in a line, one on top of the next. Trivial problems, really.
Some useful links while MER are active. [url=http://marsrovers.jpl.nasa.gov/home/index.html]Offical site[/url] [url=http://www.nasa.gov/multimedia/nasatv/MM_NTV_Web.html]NASA TV[/url] [url=http://www.jpl.nasa.gov/mer2004/]JPL MER2004[/url] [url=http://www.spaceflightnow.com/mars/mera/statustextonly.html]Text feed[/url]
--------
The amount of solar radiation reaching the surface of the earth totals some 3.9 million exajoules a year.
Offline
Hi Josh, everyone.
Very cool link Josh. You can also make a 3D fabber using carbon monoxide reacting with many metals to make carbonyls. (A fluid with the metal surrounded by CO.) Using lasers the carbonyl can be decomposed into almost any 3D shape allowing metal parts to be made of various alloys.
With fabbers making plastic parts and a carbonyl lathe creating metal parts, this will jump start industrial development on Mars.
Many thanks!
Warm regards, Rick.
Offline
Rick, I think you'd be a lot less sure of your position after reading Drexler's Nanosystems ...
http://www.e-drexler.com/d/06/00/Nanosystems/toc.html
and perhaps also the rebuttal to Smalley ...
http://www.imm.org/SciAmDebate2/smalley.html
"Heavier-than-air flying machines are impossible."
-- Lord Kelvin, president, Royal Society, 1895.
Fan of [url=http://www.red-oasis.com/]Red Oasis[/url]
Offline
Smalley's contribution may be "on shakey ground" but as far as I know he was the only one giving these really interesting counter arguments. I think that, even if he turns out to be completely wrong, it is important to have argument/counter argument with regards to nanotechnology. Smalley is unfortunately dead, so we don't have someone out there making these arguments as far as I know, and basically his whole point has been essentially dismissed without anyone showing how it can be done. For example, the sticky fingers problem is dismissed because Smalley showed how it was as problem in a certain domain, and nanotechnologists dismiss it because he didn't show how this was a problem in all domains.
We have interesting little CGI animations, but I don't see anyone going out and making them happen. The closest thing I have seen, as of yet, are DNA based pyramids which serve almost no functional purpose.
I think the obsession with nanotechnology (particularly self-replicating and programmable nanotechnology) has taken acedemia away from attempting to research larger (micro) level universal constucters, which would serve the same essential purpose.
Some useful links while MER are active. [url=http://marsrovers.jpl.nasa.gov/home/index.html]Offical site[/url] [url=http://www.nasa.gov/multimedia/nasatv/MM_NTV_Web.html]NASA TV[/url] [url=http://www.jpl.nasa.gov/mer2004/]JPL MER2004[/url] [url=http://www.spaceflightnow.com/mars/mera/statustextonly.html]Text feed[/url]
--------
The amount of solar radiation reaching the surface of the earth totals some 3.9 million exajoules a year.
Offline
Hi Everyone,
When reading this forum every now and then someone says, "and nanotechnology robots will solve THAT!"I just want to say that independent nano-technology robots are a long, long way off.
--- Rant mode on. ---
I read "Engines of Creation" by Eric Drexler, and it felt more like a religious tract than something with real science.
There are severe problems with the independent robots view of nano-tech that most people ignore.
- POWER: I've hand people tell me about nano-bots that swim thru my blood stream zapping viruses. Or little machines that burrow thru the rock finding useful metals, etc. "Where do they get their power?" I ask. "From your bodies heat!" they tell me. For a heat engine to work you need a heat differential. And one side of this microscopic bot is the same temperature as the other. Don't people know about the laws of Thermodynamics?
Just a rhetorical question:
How do crystals assemble themselves?
The don't require a power source to get the right atoms or molecules in the right place so that the parts assemble themselves into the right crystal lattice.
Buckytubes don't seem to need some robot with an independent power source to put the molecules in the right place for assembly. Crystals rely on radom motion of the molecules in the environment its in and these molecules only stick in certain places. Statistically with enough atoms and molecules moving about, you will eventually get the right sort of molecules in the right place so that they will stick. For a nanomachine think of it as a self-assembling puzzle. You have to make sure that one and only one sort of molecule will stick in the right place and at the opposite end you make sure only another sort of molecule will stick there. The trick is to get the right sort of molecule assembled in the first place, then you limit the sorts of molecules that are available in its environment so the wrong sorts of molecules won't be around to gum up the works.
On a larger scale you will need a power source and some device to move the molecules in the right places, but the initial assembly of the smallest parts is statistical.
Others talk about putting an antenna on them and receiving power from radio waves. I ask them how an antenna a few atoms long are going to absorb a wavelength measured in meters. A shrug; they don't know what I'm talking about.
Nanotechnology is going to start with biology. The first step is to reverse-engineer the structures we find in biology such as the living cell, then we try to build the simplest possible cell from scratch and see if it reproduces, and then we add things to it, we get the cells to assemble certain things to it. Now normally biology is accidental, but human directed biology is deliberate, we can create cells designed to produce certain materials, and maybe using biology as a template we can assemble nanotechnology that does not rely on traditional biology.
Other people talk about using a system to convert light into mechanical motion and using a rack and pinion system and clockwork springs to store energy. When I ask them to sketch out how this system would work I get shrugs. Light (at least that found underground or in the body) is not powerful enough to knock atoms about. It gets absorbed into the electron shell. So a system to mechanically covert this to motion in a rack and pinion is not trivial. Ah, well, at least the second system is not physically impossible in theory.
A person said that the nano-bots could take in sugars (from our bloodstream) and burn them producing wastes which are dumped back into the bloodstream. Ok, this could actually work. You will need enzymes to act as catalysts, ways of storing this power and systems to clean out the gunk produced by burning these substances. All of these systems push up the size of the nanomachines (which limits their mobility and usefulness) but again this system is possible.
- SMALL IS STICKY: People talk about the nanobots grabbing this atom here, that one there and assembling useful stuff. The problem is that those atoms are stuck together. Really tightly in some places and it takes a lot of power to wrench them apart when you are few hundred atoms in size. And when you grab one atom from a molecule both halves now want to stick to things. The piece you broke it away from, other stuff drifting around, your machine. When you put it on a half finished robot that robot has 'sticky' surfaces all over. What will stop some thing else coming along and attaching itself on your half done machine? Now you are not just building a machine you are cleaning random, chaotic debris off of it as you try to build it.
- SMALL IS DIRTY: On the atomic scale you have all sorts of elements and molecules everywhere and many of these have sticky surfaces. Glom some of them on your machine in inconvinient places and it stops working. When I ask enthusiasts for nano-technology about this problem I get suggestions like windshield wipers to brush off this stuff, or suggestions that free radicals and 'sticky' atoms don't exist in large enough numbers to be a problem. (Even tho your machine is creating such things locally.) Others talk about smaller nano-bots crawling over the bigger ones removing bad atoms.
To clean off this sticky crap is NOT an easy process. It will take a fair bit of brain power to know how to approach this particular jam, or that molecule's edge, wedged in there.
Hydrogen atoms generally serves to make a surface non-sticky.
If you put hydrogen atoms on surfaces where another atom may attach, you prevent the molecule from growing further, mostly. Oxygen can still oxidize of course.
- PROCESSING POWER: Where do the brains for these robots go? People talk about using rods in carbon nano-tubes to store information. (If the rod is up 3 atoms it means something different than if it is up 4 atoms.) No reason rod logic can't work, but tho very small to us, it must be gigantic on the atomic scale. Nano-bots will have to deal with a very, very complex dirty world. Much of it will be dark and if you say, well put lights on it, the whole question of power comes up again. (And then you need the processing power to figure out that visual information in a world of difraction and interference rings.)
Your obviously not going to get a nanomachine that consists of three atoms, there things are enourmously complex.
These machines will need a TONNE of processing power, to figure out what is in their environment (sensors, pattern recognition), will have to make plans in a chaotic environment, will have to move over potentially sticky, magnetic or electrostatic terrain. They will have to coordinate with other nano-bots for big projects, etc. And as the processing power goes up the nano-bots get larger and larger. Their power requirements go up. There are ever more surfaces that random stuff can stick to.
I've had people say that each nano-bot will have low processing power, but they all will find each other, link up into a super network, figure everything out, program each tiny bit of themselves with detailed instructions and then each bit will diassemble and go out do its thing.Ignoring how difficult all this is I ask a very simple question. I ask how they will link together and exchange info? "There will be holes so that the rod logic can push rods into each other." Well, what if there is dirt or stuff stuck to the side of one so there is not a tight fit? "No problem, we will just put windshield wipers on them!" What will keep stray atoms from getting into the holes? "No problem, we will just put little powered hatches on them!" etc.
Let's just say I'm more frightened by the iron death of the universe than I am by "grey goo".
I ask that anytime someone wants to fix a tough terrafoming problem with self replicating nano-robots, they either:
A) Discuss how the above problems would be solved.
B) Admit that this is for the FAR future.
or
C) At least wait until we have experience with swarms of insect sized robots doing some useful, productive work before you magically solve problem with robots that have an enviroment that is orders of magnitude more complex.
---- Rant mode off. ----
For an interesting story about insect sized robots you might want to check our James P. Hogan's "Bug Park". It is a lot of fun & has some good lessons for what happens when you shrink scales.
Regards, Rick.
If I had all the answers to nanotechnology, I'd be rich! I don't, but I ask you not to underestimate the limits of human ingenuity or the ability of humans to surpass themselves with their own technology.
Offline
Ever consider a machine with wheels and gears made up of atoms, it exists in a medium with a bunch of molecules bouncing about. Most of the time when a molecule hits the machine, nothing happens, but when it hits the machine in certain places, the wheel spins and it does something. The wheel only spins on one direction, there are gears that prevent it from moving backwards. These are random atoms and molecules moving about, but each impact when it hits in the right place causes the machine to advance, and that is the power supply. You don't need heat differentials, the molecules aren't treated statistically at this scale. When a molecule hits in the right place, the machine advances further and the gears and the pullies direct this mechnical motion to do some useful work, maybe operate a mechanical computer for instance.
Offline
<discussing powering nano-robots>
How do crystals assemble themselves?
Crystals form when you have a saturated solution in a liquid that is cooling or evaporating. The gibbs free energy is lower for the atoms to come out of solution. For some substances, they intermolecular forces favor a regular alignment of atoms or molecules, so a crystal forms.
However, the energy for putting the atoms into solution has already been placed into the system so the power requirements for forming the crystal have already been met.
Buckytubes don't seem to need some robot with an independent power source to put the molecules in the right place for assembly. ...
On a larger scale you will need a power source and some device to move the molecules in the right places, but the initial assembly of the smallest parts is statistical.
In an oxygen poor flame, carbon will form bucky-tube like structures, but they are far from perfect. They are sea shell like spirals, glommed onto flat graphite sheets, with several atom specks of diamond crystal structure. All in all it makes a fairly good abrasive. This is what soot is. However, to form perfect buckytubes is far from easy, requiring precise conditions. And you still need carbon atoms in a gas or plasma which is a very energy rich source.
Again, you are starting with a high energy state. If you have lots of energy you can 'reverse entropy' for a while and create order. But the energy was put in first.
Ever consider a machine with wheels and gears made up of atoms, it exists in a medium with a bunch of molecules bouncing about. Most of the time when a molecule hits the machine, nothing happens, but when it hits the machine in certain places, the wheel spins and it does something. The wheel only spins on one direction, there are gears that prevent it from moving backwards. These are random atoms and molecules moving about, but each impact when it hits in the right place causes the machine to advance, and that is the power supply. You don't need heat differentials, the molecules aren't treated statistically at this scale. When a molecule hits in the right place, the machine advances further and the gears and the pullies direct this mechanical motion to do some useful work, maybe operate a mechanical computer for instance.
Actually I have considered an idea like this but it won't work. No machine can be built that will convert random heat into useful work with out a heat differential. (2nd law of thermodynamics.) See this URL:
http://en.wikipedia.org/wiki/Laws_of_thermodynamics
also
http://web.lemoyne.edu/~giunta/perrin.html
and
http://www.kilty.com/pmotion.htm
(You may want to look at "perpetual motion of the third kind" in the third link above, as your machine will use quantum effects to bring the warm reservoir down to absolute zero in a finite number of steps.)
Assuming the second law does not apply to your machine, I can see some problems: first assume that the first atom that hits your piston pushes it in and you get some work. OK what drives the piston back? Do you have a working fluid on the other side of that piston? Then your random molecular motion has to overcome the pressure of that working fluid. If there is a vacuum on the other side of the piston then this problem does not apply, but you had to do work to put the vacuum there in the first place.
Or let us say that you have a bunch of paddles (like a paddle wheel) sticking out the side of your machine. If you have a current going past it no problem, there is energy to be extracted there. But if the fluid your paddle wheel is stagnant with only random thermal motion, then you have as many atoms bumping against the back of your paddles than the front. If we make the paddles really small, then randomly we might get more hitting the front than the back. But we now have to over come the energy needed to lift the ratchet, turn the whole paddle on its axle and to overcome the friction of the rod in the bucky tube (or spring or whatever) you are using to store the power.
If you assume frictionless surfaces than perpetual motion is possible but you won't be able to extract useful work (in the physics sense of work) from the system.
Nanotechnology is going to start with biology. ...
Agreed. But in Drexler's "Engines of Creation" he is talking about tiny machines that go inside cells. He talks about machines that dismantle whole planets, get all the information in the planet, and then put the planet back together the way it was. And so on. The smaller the little robots get the more powerful they become and the harder the problems that I brought up are to solve.
<Talking about how chaotic / sticky the nano-tech environment is.>
Hydrogen atoms generally serves to make a surface non-sticky.
If you put hydrogen atoms on surfaces where another atom may attach, you prevent the molecule from growing further, mostly. Oxygen can still oxidize of course.
Hydrogen tends to donate an electron to fill an outer electron shell. Thus it makes the non-metals less sticky. It won't help if you have a proton donor on your robot. For that matter, even if hydrogen perfectly fits the bill, you still have to perfectly keep all surfaces clean until you can find a hydrogen atom to fit into this slot. Not easy. Furthermore the most common source of hydrogen is water. If you grab one it its hydrogens you leave behind a hydroxyl ion which is powerfully reactive. This powerful base will steal hydrogens from anywhere it can, and your machine is covered with hydrogen and near by...
Even a non-sticky atom in a molecule will sometimes react. Chemical reactions are about equilibriums and while a surface may be pretty darn stable, sometimes you will get a reaction with it. Such things destabilize your machine. That is why cells spend so much molecular machinery to repairing themselves.
Note that biological systems also are troubled by the powerful electrostatic forces holding the atoms they need together. They use an incredibly exotic array of enzymes and catalysts to help them. Most fans of nano-tech seem to assume that they are assembling atoms like tinker toys in a vacuum.
... I ask that anytime someone wants to fix a tough terraforming problem with self replicating nano-robots, they either:
A) Discuss how the above problems would be solved.
B) Admit that this is for the FAR future. - OR -
C) At least wait until we have experience with swarms of insect sized robots doing some useful, productive work before you magically solve problem with robots that have an environment that is orders of magnitude more complex.
If I had all the answers to nanotechnology, I'd be rich! I don't, but I ask you not to underestimate the limits of human ingenuity or the ability of humans to surpass themselves with their own technology.
I am not saying that nano-technology is impossible. I am saying that it is DIFFICULT. I am mostly interested in terraforming problems that could be attempted within the next 500 years. Whenever someone says "Oh, we will just give Mars an oxygen atmosphere by building self-reproducing robots and having them take apart the soil. Lots of oxygen in Martian dirt!", then I nod my head and think, "Yah.... right."
Too often nano-technology is used as a magical, mental crutch to allow the person discussing it to ignore tough problems. My post was intended as a wake up call to those that think that it will be simple to solve any problem with nano-tech. Usually the problem they are discussing (e.g. giving Mars an oxygen atmosphere) is easier to solve than their nano-tech solution would be.
However, nano-tech is potentially a solution to some problems. If they suggest nano-tech (perhaps some sort of artificial enzyme to speed a chemical reaction) that discusses how to get around the problems I brought up then I would be DELIGHTED to hear from it. It is just such creative solutions that will allow human colonization of Mars.
Let me give an example of a nanotech solution that will work (not using tiny robots tho, sadly...)
You can dissolve many metals in carbon monoxide (CO) making what is known as a carbonyl. You can then fire a laser into the carbonyl tuned to the right frequency so that it will decompose the carbonyl and leave pure metal behind. The CO gas floats to the top of the tank.
Using finely focused and controlled lasers solid metal devices can be built up one layer of atoms at a time. Microscopic machines can be built of almost any shape. Using mixtures of different carbonyls and different lasers you can even make a variety of alloys.
How does this idea for making microscopic machines deal with the problems I raise above?
POWER: External. Not a problem.
CHAOTIC ENVIRONMENT: We prepare ahead a time a pure solution of say, iron carbonyl. Given that the environment is one simple substance this problem is solved.
DIRTY ENVIRONMENT: We prepare a very clean environment (at considerable cost of effort) ahead of time. Thus this is not a problem.
PROCESSING POWER: This is outside of our carbonyl lathe working surface so it is not an issue. We can make it as large as we want.
This is an example of a nano-technology problem that is easy (altho I have not hear of anyone that has actually built a carbonyl lathe). The machines that build nano-sized integrated circuits on a silicon wafer are likewise solved problems.
Note that the integrated circuit industry has spent billions of dollars and years of time to solve the very significant problems involved with chip foundries.
Warm regards, Rick
Offline
Well the nanotechnology that we know works is biology. Now if we can reverse engineer "nature", we could create synthetic biological cells. The cells would metabolize sugars or photosynthesize just like the natural cells found in nature, but these cells would be optimized. One difference would be this. In nature, cells have evolved over a billion years, each cell has genetic encoding in its DNA, much of this DNA is junk, it does nothing useful, but suppose we learned enough about the mechanisms under which a cell operates and were able to create our own optimized DNA with no junk, and created a minimal cell as small as physically possible and "programmed" its DNA to cause the cell to perform a certain task and then divide. I think there are alot of useful things a synthetic biological cell can accomplish, much of the problem in terraforming entails converting atmospheric gases, not building large complicated structures.
Another problem with biology is the tendency of cells to mutate or coply themselves imperfectly, this property has allowed for evolution, but if we were to design a cell from the bottom up, we might want to make this cell less mutable than a natural cell, perhaps a built in error coding system. We should proceed carefully with synthetic biology as whatever organisms we introduce may become part of the permanent biosphere, and they may persist long after their job is done, ot they might do their job too much, producing too much oxygen for instance.
Offline
Imagine a kind of "nanotechnology" that's not the way Eric Drexler envisioned it, it is not a bunch of molecular gogs or levers, this technology is much more familiar to us and it operates on the level of the living cell. These cells have all the basic functions as our own cells, they are made of the same basic building block of nature like ours as well, they require nutrients of the same sort that out cells require, they contains an aqueous environment inside themselves, and they must get rid of waste like our cells do as well, the only difference is that these cells are build from the bottom up, they aren't a direct product of a billion years of evolution, but something deliberately created by us through a complete understanding of how a natural cell works. Once we understand how these cells work, we can program their molecular sequences. e could program such a cell to divide exactly 100 times no more and no less for example, we can cause these cells to differentiate and produce different tissues such as muscle, bone, organ or brain, we can cause these cells to divide and produce many different sorts of brains including natural ones. The ultimate produce of this field of study would be something I would call "Living technology"
You could create a robot made out of living cells, this robot would need to eat, eliminate, would perhaps need to sleep, would need oxygen to breathe and live in the same environment that we do, it could even be as intelligent as a human, and may look like one as well, although this won't always be necessary, but here is where the similarity ends. As human looking as this living robot may be, it won't think like a human, because we don't want it to, we deliberately design it not to so that it will serve us humans. This is admittedly a scary notion that we can't tell our machines from ourselves except though close examination of its DNA. The DNA of a living machine would be short, deliberate, and purposeful, it would contain no "junk" as a natually evolved living creature would have, and such living technology could be tremendously useful in terraforming Mars. We could create them by the billions, they have the ability to reproduce, their bodies could be designed to withstand the martian environment, and as that environment changes, we create new living robots which are designed to operate in the new envirnment. Robots of this sort would carry a supply of oxygen internally, and breath infrequently much in the way of a whale underwater. Eventually these robots would need to take a breath and eat some food, but inbetween breaths it can do some work on the Martain surface or even in space without a space suit. A supply of oxygen, water, and food would have to be kept nearby at all times. the celluar adaptations to deal with these harsh environments would be somewhat unique, but certainly not impossible. The advantage of living technology is that these machines can repair themselves though a process called healing or regeneration. If a living robot loses a limb, it can grow a new one.
Certain types of living robots could look completely human, and could act as personal servants for the real humans, they are not human in the slightest however, their cellular biology is not designed to accomodate human viruses, so they can't be virus carriers. Imagine the most beautiful supermodel you've ever seen, and that's the shape of your personal living robot servant, it looks human, so human in fact that your temped to call it she rather than it. if you touch it, it feels warm, its skin has a human texture, it sweats and has hair, it is designed to look and feel human down to the smallest detail that a casual observer might make note of without powerful microscopes to examine its cellular biology. This robot doesn't think like a human do, all it wants to do is serve its human master, it has no concept of self and no concept of self-preservation except in the context of following its programming in the service of its master, as it was designed to do down to the cellular level. You could have sex with it, but you could not produce offspring by it as its not really human, its cells and yours are not compatible enough to produce offspring as its an entirely different creature and only looks human.
Imagine billions of these robots, the human looking once are designed to interact with real humans while the not so human looking ones are designed to operate in particular environments or do certain tasks that it specializes in. With such a mastery of biology, humans can be made to live a long time, we'll understand how a cell operates, so perhaps we'll be able to maintain them better than nature can, old age will be a thing of the past, if we lose limbs, we can grow new ones. Our technology can be maintained similarly, but it gets kind of eery not being able to tell some of the machines from the humans. These machines can be set upon the task of terraforming Mars, it may take 10,000 years, or perhaps less with optimized atmospheric converting living plants. The living robots will toil throughout the Solar System, mining asteroids and comets, building giant mirrors in space and propelling them through the solar system via living spaceships. If something requires more resources we just have more living robots. Humans need not work at all, and their participation is not required in the terraforming project except in telling their robots what they want. When labor is free the cost in only in the materials and energy of the Solar System used up.
Offline
Pages: 1