The Future of Programmable Matter

something we talked about earlier this month. Yeah, on December third, we published a podcast called self or podcast episode that terminology always wakes me out, anyway, it was called self We configuring modular Robots go and they went, and they did, go, they did, They went everywhere kind of thing they do, right. Yeah.

The concept of that is the the a robot that's made up of smaller components, each of which is its own robot, right, And so it's sort of an aggregate like a transformer, except actually better than a trans former because the transformer can only turn into two shapes, maybe three for some of them. Sure, sure, but it's basically either a car or robot or a gun or a

robot or etcetera. Right, And so a self reconfiguring modular robot, on the other hand, would be able to reshape itself into any number of different shapes because it's made of identical individual components that can adjust their position and functionality with relationship to all the other ones. Right, And we talked about a couple of different types. We talked about

M I, T, S M blocks. We talked about Harvard's kilobots. Now, the kilobots are not U modular robots in the same sense, but they demonstrate the the swarm behavior that would be necessary, the idea that each individual robot would be able to move into the correct position to form whatever shape is needed. Uh, they're more of a tool to test out that kind of technology, that sort of artificial intelligence that will be

a necessary component it for a true self reconfiguring modular robot. Right, because anything that large isn't ultimately going to be as useful as as something with parts that are much smaller that the smaller the parts, the more useful this kind of stuff is going to be. Yeah, And of course, as we've mentioned plenty of times on the show, miniaturization is hard, especially when you need to power all these elements.

But let's be optimists for a bit and imagine it's going to be weird not for me, And imagine how small you could make these modular robotic elements. That was, you do not have a very powerful imagination time bandits reference. Um no, I want to I want to talk about how this applies to a subject known as programmable matter. All right, so let's actually talk what is programmable matter. Well, programmable matter is matter. It's so it has substance as weight.

You can hold it in your hands, but it can alter its own physical characteristics in a non random way based on user input or pre programmed behavior. So yeah, where did this idea come from? Well, let's turn back the time. Watch the hands of the clock spin backwards. Uh, now it was it was the idea itself. The term was proposed by Tomaso to Foley and Norman Margolis to

M I T. Computer scientists. Now, their concept of programmable matter was really more about tiny computers that could perform parallel processing and communicate with one another, that could simulate the physics of real matter. But we're talking about a simulation.

They weren't replicating. These These small computers in this proposal weren't meant to take on physical shapes, but rather be able to communicate in a way that would allow them to process huge amounts of information very rap Okay, But of course subsequently people really expanded on this idea, right, We ended up hearing about another approach where they said, what if we took the same concept, but instead of

just talking about computers, we apply it to robotics. So each computer is itself a tiny robot that has some ability to move around, to communicate, to bind with fellow robots, and that they would be able to have a tiny processor inside them to to take on commands and then enact those commands. And then you could, in theory, have this massive robots that could create actual physical structures, larger macro structures. So kind of similar to the modular robots

we talked about before, but on an even smaller scale. Okay, but what people are really thinking about today is something kind of amazing picture. You've got a little blob of gray junk. It's just it looks like putty or goop or jam or something. It's just this kind of amorphous material jam. But you give it data and so the it is actually made of tons of tiny spherical elements that are all joined together and all individually have computing capability.

Somewhere deep inside each of those tiny little particles is a circuit and it takes the data you put in and says, okay, I need to be over here now in relation to this other particle, and they all do that at the same time to form a shape. And the shape is whatever you tell it, and then if you are tired of that shape, you can tell it to make a totally different shape and it will pretty creepy. Huh. Yeah, yeah,

I mean it's creepy and great. Like this is sort of if you want to get really crazy with it. It's sort of the T one thousand idea. Yeah, it's it's T one thousand meets silly putty, right. Uh. To be more realistic, it's sort of like not a fully functioning autonomous robot running around in the in the southern California area, but it's something that can change its own

shape at will based on what you want. Yeah. So you could think of it kind of as as a bunch of uh, you know, modules that can form various structures, but the modules themselves are so small that your your list of possible structures you can make is in the ideal implementation practically unlimited. Yeah. Imagine there as small as grains of sand, as if sand could make anything out of itself. It's almost as though this is some kind

of automatically sculpting clay. Yeah, that's a good name for it, but you didn't come up with it first, now, I did not, not at all. Actually, a one cool name for a particular implementation of this idea of programma will matter is Claytronics. And you might have heard this term before. It's a popular and cool idea and futurism circles. But what's behind the actual Claytronics idea. Well, I mean it came out of Carnegie, Melon and Intel, along with a

few other partnerships. There is an actual project. There's a great website where they kind of talk about what Claytronics is all about and how they hope to achieve it, which is is pretty neat um. The project's purpose is too and I quote it combines modular robotics, systems, nanotechnology, and computer science to create the dynamic three dimensional display of electronic information. So think of this as an alternative

to a computer monitor, right you are? You can actually use this to visualize data in a physical form, So you could you could in theory. Use this to do something as simple as compare the market share of two different operating systems and have two physical UH representations pop up, and maybe there's a large sphere in a small sphere to demonstrate what the two different UH companies are operating systems you're looking at how they compare against one another.

Or you can do something much more complex, like what we're talking about, the idea of being able to make any particular physical shape and then reshape it just by you know, running a program. Right. And so this technology is obviously based on the power of the individual particle,

right that's moving about to create these shapes. And in the field of claytronics, this is the claytronic atom, the catom it right, and the catom, like a lot of the units we're talking about, is modular in that you can have that connect with other atoms to make these larger shapes, these macro size shapes, right. They don't have to be specialized. They're all the same thing. Yeah, Yeah, And that's not always the case with programmable matter. There

are other implementations. You know, this was one approach. There's another one Cornell's Creative Machines Lab is working on a lot of different ways of looking at programmable matter. So I gotta keep in mind that we're talking about, uh, something that's in its infancy here, so we're still experimenting with different approaches to how to get to this kind

of uh future where we have programmable matter. So one of the things that Cornell's Creative Machines Lab is looking at is uh using these tiny building blocks that they call Vauxell's. These are physical objects that they know Vauxell's sort of like pixels. A pixel is an individual unit of display on your display. Why am I thinking it sounds like an alien from Hitchhiker's Guide to the Galaxy

Theogans yea with their terrible poetry. Uh, yeah, Vauxell's in this case, Yeah, it makes sense because of a single Vauxel would be a single unit that, together with other vox can make up this physical shape. Just as a single pixel is kind of useless. You have to have lots of pixels so that you can actually represent whatever visual uh image you want to to have on your

screen like that terminology. Yeah, that's kind of neat. It makes it It makes it easier to explain to people, right, Like think of it as just as an individual unit. It's kind of like an again, just like Claytronics goes with the catom, where you think of the atom as the individual unit. This is very similar. Uh. Now, these building blocks could come in different varieties, so you could have vaux holes that are soft or hard. They could

come in different colors. Some might be conductive, some might be insulators, some could have robotic elements like sensors or actuators. So you would get an assembler of some type that would print out those building blocks whichever ones you needed for whatever it was you were playing on building, and then you would be able to put it together to make the thing you need it to make. So I

think of it more like tinker toys or legos. You would get the different blocks you need, and you look at the instructions and then you can build the actual thing. So it's a little different from claytronics, where you would theoretically be able to change that shape over and over. I'm not exactly certain that the vauxells would have that same capability and may not, but it's again kind of a stepping stone toward that programmable matter where you would

be able to change things on the fly. One of the other things I liked about the voxhel approach is how they describe building a three D object, that it's a digital method as opposed to an analog method. So analog you would consider that to be a continuous shape, so you know, as curves, as edges, um, and but it's all continuous. Digital is there's either matter there or there's not. It's a one or zero. So it's sort of the difference between you know, a real image and

a pixelated image on the screen. Yeah. Yeah, so you would if you were trying to create something, you would say, all right, where where to do the individual vox need to be? And where do they absolutely not need to be? Where do they need to be absent? And uh, you know.

I've also likened it to the idea of We've talked about this with sculpting, where you cut everything that isn't the thing you want out, like, cut everything that doesn't look like David needs to be cut out of that block of granted, until you're left with David or marble. I guess I should say, at any rate, what what do I know? I'm no artist, um, but anyway, it's kind of neat because it's it's that one in zero approach. But there are other projects at Cornell that also relate

to programmable matter. And my favorite one out of all of them, and keep in mind their lots, is called jamming granular materials. Yeah, it's not fish, it's not it's not widespread panic, it's nothing like that. It's not that kind of jamming. Uh. This this is actually looking at the way, it sounds like a medical condition, Like you end up in the emergency room because you've got a case of jamming granular materials. I'm sorry, miss, he's got

jamming granular materials. Materials. Well. They The idea about this is is grain, very fine grains of of matter can behave like a fluid if they are loosely packed into a container. So imagine that you've got a jar that is maybe half filled with very fine dry sand, and you move the jar around and the sand flows almost like a fluid. It's similar more like a fluid than a solid. Right. But then let's say that you have twice as much sand packed in there, so it's actually

really tightly uh packed inside that jar. Then it is rigid. Even if that jar were made out of a flexible material like plastic, you would have this rigid structure inside of it because you had jammed the the grand granules

so tightly together. They're the same sort of idea has been proposed for types of these individual units for programmable or where you have some mechanism connected that can either compact or then relax and allowed this sort of granular material to either become rigid or soft, to change different shapes. So you would have the units and they would all be connected to each other in specific ways previously like

they wouldn't disconnect and reconnect necessarily. They might all be connected permanently in one in one type of type of structure, but based upon these little links between them made out of this granular stuff there, they take on specific shapes because they're either rigid or they're soft, and it doesn't matter. The other approach reason they go with this approach is that doesn't matter what the temperature is. They don't need to change the temperature to change that rigidity. It's just

whether it's compact or whether it's loose. So I thought that was pretty neat the the image they had imagine kind of a pyramid where each it's it's a it's a just the outline of pyramid, the structure of a pyramid without the actual phy egal sides. And each line is a little soft plastic container of sand. So when it's all rigid, it stands up like a pyramid, and when it's soft, it's just it's just a little pile.

So uh, it's it's again just showing that there are multiple approaches to this idea, and there are a lot of other universities and companies and research centers looking into this, including companies like Autodesk Research or Whitesides Group Research, m I T of course looking into at Harvard. We mentioned also there are tons of different approaches and no one yet has the you know, the way to do this,

but definitely not. But by coming in it from so many different angles, it's awesome, right because it means that we have we're not putting all of our eggs in one basket, and it may mean that we ultimately find different ways to do the same sort of output, which gives us a lot more flexibility. You're gonna have like the Blu Ray HD DVD wars all over again, but about how you turn putty into a hammer. I'll go, I'll go to c e S one year and and the week before CS one will have pulled out of

the show because it's clear that it's the loser. The first time I went to they will still be Putty Hammers HD DVD pulled out of c S. The first year I ever went, there was this one enormous blank spot on the c show floor where they were supposed to be. People were all taking naps. I guess, yeah, yeah, I think right, Well, I think we should look at what are the possible uses for something like claytronics or

other similar forms of programmable matter. Yeah. Yeah, We've talked about this a little bit, but let's let's go deeper into it. Okay, Well, I wanted to gather a few different things here. One of them would be the tangible three D representations generated by data, and we sort of talked about these earlier. So one thing is you could simply use three D object to represent numbers and data. But the other one would be that you could sort of have what I would think of like a Google

Docs of three D objects. So imagine you've got a saved document that is controlled by a digital file on your computer, but it's represented by a three D object, a real tangible three D object you can hold in your hand and you can make edits to it with your hands. Yeah, it's sort of like instead of having to, uh, you know, if you if you get a three D printer and you print out a prototype, if you want to make changes to that prototype, you have to print

another one from the ground up. You've got to go back in and redesign it digitally, redesign the digital thing. And and then well, I mean, and you'd still be redesigning the digital thing, but you'd be doing it physically with your You could just push is I wanted an indentt here? So instead of going and redesigning it on the computer, you just press it in and then those changes would be reflected because we're telling thing about at two way Street here. With a three D printer, there's

a one way flow of information. It goes from the computer to the printer, which then creates the three dimensional object. With this implementation, it's more like an interactive display. You create the thing, but then you can actually manipulate the thing you have on the display. In this case, the display is a three dimensional object and that gets reflected in the file that you originally created. So that's it's

it's kind of uh. It cuts out another step in prototyping, which could mean that the prototyping process, which has already been sped up phenomenally because of three D printing, becomes even faster. Right, And so this would make a big difference, I think, to people who are like engineers, but also to people who are just designers, people making art and or or even designing products like smartphones and things like that. I mean industry. Yeah, Johnny Ives, get a hold of

this thing, forget it? Game over. Another are interesting use for something like this would be a kind of utility putty that that phrase appeared in my brain, I want I wonder if I've read that somewhere before. I feel like that maybe a term that has been used by someone before. But anyway, just now it's yours, right, patent um No, Well, wherever this idea comes from, the utility putty would be if you imagine a kind of a small puddle of this beige or gray, it really doesn't

matter what color it is. Let's call it pink. Uh, this small puddle of pink putty sitting on your kitchen counter. All right, I'm with you so far. You're making dinner now, and so you need a ladle and a bowl and cutting board and a knife. So you call them up on your kitchen utensils app on your computer, on your phone, and the puddle of putty as symbols itself into these shapes and becomes rigid. You use them like you would to prepare dinner. Then when you're done with them, they

dissolve all back into the putty. Now, this could be a massive space saver in the home. Imagine if you like only had to call furniture into existence on demand when you needed it, and when you didn't need that piece of furniture, you could dissolve it. Yeah, or think about how useful something like this might be for portability purposes. So you're going on a camping trip, and if you ever been on a camping trip, you take the tent along with you. It is a large, unpleasant, unwieldy object.

It's difficult to assemble or yeah, a whole bunches of different parts. Let's say instead you go out camping and instead you just bring along a bucket of utility coup. I got a bucket a tent right here. Yeah, you so you just pour the bucket out on the ground, queue up the sort of shelter design on your phone, and it assembles itself. You know, Joe, I have to

say that tent design is beyond the pail. Oh oh, now we may be able to offer some pretty relevant criticisms of an idea like this in the next section, because you you immediately start thinking like, wait a minute, where does the power for it to form that shape come from? In the woods? And stuff like that. But uh, just a couple other ideas. How about the three D facts. This is something that Claytronics people I think talked about.

It goes like this, So you want to send a copy of a three dimensional object to somebody far away, Well, you could just submerge it into a tub of these particles like atoms or whatever they are. You've got a bucket of them, and you dip it underneath them. And then of course, once it's submerged, the particles can sense all of the contours of the object and then it can recreate that shape at the other side of the line. And here's the most sci fi of all, but it's

something I have seen talked about. The idea of sort of the ultimate telepresence instead of just sending your voice, instead of just sending your face, you send a real physical replication of your body and movements across the telephone wire. I think a giant pink blobby. Jonathan Strickland is the stuff of my nightmares. Come here, I love you, Teach me to love. Yeah, even without that particular image in

my head, that one struck me as being creepy. I'd like to apologize to everyone has not had that image in their heads. Jonathan hugging his loved ones from far away and a goopy, goopy mess and melting on them halfway through because the power goes out, Jonathan, And this is creepy to me. Alright, alright, well, okay, So other than the fact that we're all creeped up by this, what are some of the challenges that this sort of stuff faces. We Joe mentioned a second ago up up power.

That's that's a huge one. Um. Obviously, this face is a lot of challenges, right, or otherwise we'd have it right. Well, man, interesting, this is very difficult to create, right Manajorizing power is one of those things that we haven't really cracked yet, right. We we can miniaturize circuitry to astonishing degrees, but manaturizing

things like batteries is a lot more problematic. Um, assuming that you are using batteries and you're using something that is relying upon a chemical reaction to produce electrons for electricity, that eventually that's going to run out, You're gonna have to recharge. How do you do that? Like, what's the what's the mechanism? Right? What is the actual uh, method

of delivery and method of execution? Because if these elements are going to be forming any shapes that they're going to move about and take on specific forms, they obviously need to have some form of power to do that. They can't just you know, do that without any outside influence. There has to be something acting on them. Uh. And it's I go with electricity being the most likely. It's

not necessarily the only one. We'll talk about the possible possibility of going a different route in the second, but it's still something that we have to consider like how do we get that to how do we crack that code? Okay? But also how do you get them to do what

you're trying to get them to do? You need these particles all moving around each other, and any time that you have motion in a robot, moving parts are a complication because they can break and fail and do things that you didn't expect them to do, like totally not work right. So ideally you'd want these particles to be able to um move around relative to each other without having external moving parts, right, right, And in fact, a lot of work has gone into researching different ways to

do that. Even the killer bots, the Harvard killer botts, which look like you may remember I mentioned it before, it looks kind of like a quarter sized circuit um that's sitting on top of a bunch of a little spider legs. Well, those individual legs don't actually have any movement to them. They don't they don't pivot or move in any way like that action figure. Yeah, but there's there is a little vibrating motor that's part of the circuit board, and the way the motor vibrates determines which

way the little thing skitters across the floor. So there are a lot of different approaches that are looking into, not necessarily just that implementation, but this very this very idea you're proposing, Joe, the idea of a unit that can move around relative to other units without itself having any moving parts. Yeah. Yeah, I've seen magnets proposed as another thing. By by flipping the polarity of a magnet back and forth, you could get things to move around

each other. And there's some that are talking about electrostatic forces. Uh. In fact, that's another interesting element that we can talk about the fact that uh, sticking these things together a chance you've moved them into position, they need to form up and be solid, right, didn't the M I T M blocks which are definitely bigger than than you know,

particle size. Sure, yeah, they're like like the size of blocks that that you would give a little kid, like the once had the letters, right, but they I think had a momentum based movement system, right, like they had magnets, and then they would move by having a little flywheel inside that Basically you know, did the move you do when you're on the swing set as a kid and you kick your legs forward. It's a build up, build up momentum right right. It's it's like the dancing toaster

in Ghostbusters too. Yeah, yeah, it actually does. It does move like that like it's you hear it worrying and then there's a sudden jolt and it and it leaps, because what's happening is the fly will start spinning at an incredible speed, higher and higher. Yes, it's just like your love lifts me. Uh, Ghostbusters too is a terrible movie. I don't know why we're referencing when it really is

you don't love Prince Vigo. When it breaks as in B R A K E. And then that momentum gets shifted into motion for the block, which allows it to leap over or onto other sub objects. So super smart. I don't know if that would be doable at a very very small scale. Could you create cranes of sand that use momentum to move around? I don't think they've got those little electronic gyroscopes, and I still don't really

understand how they work. But sure, I'm sure that if you cause and Adams electrons to spin around and then you make it stop real quick, then it can totally bounce across the room. I think, believe Lauren, I think you'd probably be looking at other forces rather than momentum on something that's small of a scale. Why are you killing my dreams? Chat? You know you don't love Prince Vigo. I don't care what you think you guys, chasm, get

off my podcast room. Okay, okay. What are some of the other challenges with programmable matter, like the kind we've been talking Well, like we were saying, making them stick together? Right, that's hard, Like, how do you how do you do that? Yeah? What once they're in the right position relative to one another, how do you get them to lock up and be solid? Yeah? I mean, magnets are nice, but obviously, like all of these other issues, it gets harder and harder than more.

You miniat your eyes right. Uh. There's also the idea of using things like sockets or clamps, so this would be a physical thing on each individual unit that would allow it to connect to its neighbors. Uh. The clectronics projects exploring electrostatic latches, so you could have an external force applied to this stuff to make it stick to whatever shape it's supposed to be in. But that also gets complicated, particularly as you add more and more individual

elements to the overall object. Um, you'd also need some form of processor inside each unit, at least on some level. It wouldn't necessarily seems like the easiest part almost almost It all depends on how complicated the processor is and exactly how small we're talking, Yeah, because we have been

able to vastly miniature eyes computing. Definitely, the individual components have gotten incredibly tiny, Like a transistor on a microprocessor might be as small as like fourteen nanometers in width, which is that's it's all impossible for me to even imagine. It's so tiny. We're getting down to a very small scale with that. And depending upon how complex the processor needs to be for each individual unit, you can get

away with a pretty simple one. I mean, if all it needs to know is where it's relative position needs to be, uh compared to every other unit inside that that group of units, whatever it is catams or vauxhols or whatever, then it may not need to be that complex, right, you might not need that much processing. But if you're talking about adding other elements to it, like sensors or actuators, well then it starts to get a little more complicated.

The processor might need to be a little larger, there might need to be some other elements there, and that's

where you start running into some some other challenges. Another thing I want to point out is that the Claytronics project specifically calls out Moore's law as being an enabler for this kind of approach the idea right there, basing this on the assumption that Moore's law will continue to hold true, right, that we will be able to continue make being more powerful computational units things like you know,

processors at smaller and smaller scales. But if we ever do hit that kind of that wall where we suddenly plateau because we cannot physically make things smaller and still make them work, then that could become an issue. Or it may not mean that it's impossible to make programmable matter. It may just limit how small the individual units can

be before we can't make them effective. So we might still have programmable matter, but maybe we're talking about larger grains, so we have uh slightly you know, lower resolution for your macro sized object once they're all connected together. Um. Now, of course, it could turn out like people have been predicting the ends of into Moore's law shortly after Gordon

Moore made the observation. Uh So, it may very well be that it's premature to even say that such a thing is gonna plateau, because we might find other alternative means to keep moving that processing power forward. Just in a totally different way than what we have been doing. So I don't mean to say that that you know, this is untenable. It may be that we have find we find another way of doing it um, And there are plenty of engineers who are way smarter than I

am who are working on that. So it's just something I wanted to bring up another side of the kind of Morris law argument, which is in essence a physical property problem. Um, we have the the physical properties of particle interaction because they do different things at a macro scale, the scale that you and I and all of us run around on and the scale at which say, atoms and nanoparticles interact with each other. Right, you're talking about

quantum effects. Now we're talking about these you'll never get your hands quantum entangles with, like a baseball or something. I had an issue of quantum entanglement in college that nearly got me suspended. But at any rate, yeah, it was whatever that meant. It sounded creepy. Look, look, I had a love of chili cheese fries. There was an incident. I'm not allowed to talk about it more than that. Gosh, you guys just assume the worst. No, I'm sorry, you're

a sweet man. Thank you? At any rate? Uh no, I was specifically trying to lure you down a dark and depressing road. Um at any rate, No good. What what Lauren was saying is absolutely correct. We get into these quantum effects which can cause some real issues. Now, granted, we're talking super small here, right below the micro scale

down to the nano scale. But if you are actually going through the ideal implementation of this idea of programmable matter where you're able to make nano particles be part of this, you have to take quantum effects into account because you've got all sorts of weird things that can happen in that scale. Now, my guests would be as crazy as it sounds to us to have, you know, say, programmable matter particles that are the size of grains of

sand or something. This wouldn't really be much of an issue even on that scale, right, I mean, I think we'd have to be talking about even smaller than that. Yeah, yeah, exactly. I mean you have to get like, before we got down to say, transistors at the hundred nanometer scale, we didn't have to worry so much about something called electron tunneling, which is where an electron. Can you know there's a you know, you can't ever be certain exactly where an

electron is. There's kind of a yeah, there's like a zone, and the electron could in theory exist at any point within that zone. I'm oversimplified. Their their probability distribution exactly. Yeah. So let's say that you make an electron gate and the with the thickness of the gate is less than what that probability region you know, covers, the probability region actually overlaps it. So the electron should be on one

side of the gate. But this probability probability distribution states that there's a chance, maybe a small one, but there's a chance that could be on the other side of that gate. Well, if there's a chance, that means sometimes the electron is yeah, which means that the electron is has passed through. It's as if the electron has tunneled through that gate, even though it has not physically done.

So the gate never opened. The electron is able to pass over because it's that it probabilistically there was a chance for it. Yeah. It also it also means it makes your computer less reliable. It means you could get computer errors so now we're probably I'm very skeptical that we're ever going to get to a point where we can deal with matter on a scale that's small to make it programmable. But that's only based upon my very

limited understanding of nanotechnology and physics. It maybe that one day in the future we can control matter at that degree,

but it would really surprise me. Okay, but hey, speaking of controlling matter um and and physical and quantum properties of things, if we're going to reach that point anyway, there's a possibility of using chemical or physical interactions to create programmable material um It's it's the kind of stuff that a lot of researchers who are thinking about for d printing are being inspired by, like a protein folding and DNA stuff and biological interactions on that intracellular level

um and and in parallel crystallization patterns as well. Those kind of processes happen all around us all the time, and if we could harness them, it could be just as effective as building we little robots. Right. In fact, Jean Marie Lenn proposed even before we got to the term programmable matter, I mean this predates that genre. Lenn proposed developing synthetic molecules that could follow the rules of self organization, which would be determined by chemists by the

actual construction of those molecules. Once you know the structure of the molecules and how they interact with each other, you know what form the macro object will eventually take form to form these more complex structures. And Lenn called it informed matter. Uh so there's a lot of theoretical work in nanotechnology that follows that particular philosophy, just like what Lauren was saying. Yeah, so yeah, it's it's and

it is much more similar to four D printing. Uh if you've seen that video off forward, thinking that that would be closer to what we're talking about here. Um okay, So I want to round out this episode with some nay saying, all right, I've been I've been holding it in all episode. Uh okay, do we need to worry if if we're starting to make this this programmable putty stuff, do we need to worry about the gray goo apocalypse? Uh? No, No,

because because we're not self replicating, right. No, Well, I mean we don't need to worry about it for a couple of reasons. Uh. Number one, I am very very skeptical of the gray goo apocalypse, even if you're actually talking about molecular simblers, which I am skeptical of to begin with. But also this is not the same thing

as a molecular assembler. A molecular assembler is something that at the you know, molecular level, would be building new particles, making new molecules, and eventually assembling matter out of them. And the terrible idea is that what if they start making copies of themselves and turn the whole world into molecular assemblers. You know, they just eat through everything like using it as just using that as the raw material

to build more. Yeah. The difference between programmable materials and replicators is that replicators are kind of like carpenters, right car these are these are the things that make other objects. They themselves don't form that object, whereas programmable matter is just stuff that can conform to whatever shape or or you know, substance you need to perform a specific task, but itself is that thing. It's not making a chair,

It itself becomes a chair. Okay, so it's not melting down your old chair to to make sare I can't really imagine it would be all that dangerous. I don't know. Maybe if you ate a bunch of it, keep it up Toddler's hands. Okay, do we need to worry about people creating an army of T one thousand, so you know that one's more realistic. You know. That's response to that question. I don't know, because I I mean, all

of this, if it's realizable, is pretty far away. But what seems much more likely to me is the ability to UH to sort of as symbol and then dissolve rigid shapes. It will not so much the ability to create complex, moving and thinking object. However, we did talk about the possibility of using it as a form of visualization for telepresence, so and as I noted at that time, I think that's one of the more out there kind

of sci fi. So it may be there. It may be that we could in that same sci fi future where we could make this as a telepresence, you could make something akin to the T one thousand, although maybe it wouldn't have its own intelligence. It would be operating under some other UH commands that were remotely. It would be a remote controlled T one as opposed to a self aware, self actualized AI would have to advance a great deal as well, and in this time, if we were to create an actual t one be as smart

as Robert Patrick and as suave, suave, suave guy. But let us, let's let's all make a pact that we're just not gonna pair that kind of AI power with that kind of programmable material. I can, I can almost guarantee I won't excellent. Okay, what what about one last one? What about someone hacking your claytronics and I don't know, like making them attack you or like stick solely in

the shape of Dolf Longren space or something. I mean, you know, maybe I think of it in the same way that uh, like a hacker could hack into your system and cause your display to show whatever the hacker wanted. If you're if you're thinking of claytronics or programmable matter as a type of display technology, there's no reason that the same thing couldn't hold true. I'm trying to think what it could I mean, like, so again, we're assuming it's not going to be able to create a robot

that walks over to you and kills you. It might be able to form itself into the shape of a knife or something. Yeah, well, I mean, I don't know, Like if if you if you've got a chair, a Clatronics chair, and oh like it dissolves the chair while you're sitting on it. I see what I imagine, Pokey, what I imagine is trolling people. So like when a Star Wars nerd like myself decides that they want to make their own lightsaber hilt using this stuff instead it

forms the hand giving the vulcan salute. If that would be the kind of like, no, it's not even the rate starting man, it was stir Wars, that stern. That would be exactly how I would sound. I think we need to go back and start this podcast over with you doing that voice. Yeah. In fact, if if we can time travel and do the entire run of it, and that's that's when our podcast numbers plum. Okay, well, you know, actually I can kind of see that that

might be a concern you. You might want to I don't know, that might be a good case for if you were to ever have a sort of utility putty, to have it be a stand alone system that's not connected to the Internet. Yeah, I mean I can't. It's hard for me to imagine that simply because Right now, the trend is to connect more and more of our stuff to the Internet, and it's to have fewer and fewer stand alone systems that are isolated from the Internet.

But it could be that because of things like Internet security, that we start to see a trend reversing. That it could happen. We could start to see things like self contain networks that don't they either have limited or no connectivity to the Internet. Overall, that could become a trend and at least the for you know, the near future. I don't know that that's sustainable forever. But um, I did have one more naysayer question. But I think that

we've I think that we've mostly covered this one. But but I'm going to go ahead and say it out loud. Um okay. Based on the way that three D printing technology is moving these days, um becoming cheaper and and more advanced and using multiple materials and and electronics that you can print right into your your three D printed stuff.

You know, just at what point, like like, I can't imagine a universe in which programmable materials become cheap and plentiful enough to replace the kind of advances that we're seeing in three D printing these days. That's a that's that's a very fair criticism. I think the one big benefit obviously we've talked about is that in the ideal implementation, programmable material can take on any shape, so it replaces practically anything that that material is able, like anything that

any shape that material can make. It replaces whatever that would have been right. So in your kitchen example, it could replace all your kitchen accessories, which at least anything that's not necessarily like a blender or a toaster or something like Oh sure, sure, you know it's because you don't need a whisk twenty four hours a day. If you do, you're doing something that I'm not entirely sure I want to know about. It goes back to my college days and the quantum entanglement. But at any rate.

But but no, no, I mean, it could be a potato peeler, or it could be right, right or right. That's that's the benefit because obviously, with traditional three D printing, you are printing a set object and that's what that that's what's going to be until you throw it away or it breaks right, it's it's gonna be that thing.

It's not gonna suddenly morph into a different shape. However, if this programmable matter is prohibitively expensive, so that the buy in cost is so high that it doesn't make sense to ever get it, even though it can theoretically be you know, quote unquote whatever you need it to be, it would make much more sense to go the three D printed route, which is going to be less expensive. So economics plays a role. I mean, that's one of those things where we talk about the possibilities, and this

plays across all areas of science and technology. We talk about the things that are possible versus the things that are probable. We might prove in the lab that a certain approach is possible, but if it's not economically feasible, it's not plausible, right right, And and I certainly don't want to say that that everyone should stop researching this kind of thing immediately because it's useless, because A I would never say that about science, because all science is

useful in one way or another. Um, even if it's just the learning process that we go through to reach a conclusion that something is is less than feasible. Um. However, Yeah, No, I mean I just wanted to to throw a little bit of reality and in this I mean, I mean, it would be so awesome if we actually created this, Like we didn't even get into the industrial uses for this kind of stuff, Like if if you could create a bridge that you could reprogram on the fly in

case of natural disaster or something like that. Or I mean, think about if you had I don't know, airplane wings that could adjust themselves in shape to be maximum efficiency at different stages of flight. Um. Yeah, I mean, they're they're all kinds of different things that you could do

that we can just sit here and imagine. Part of the most interesting thing about something like this Claytronics programmable matter future is that we can't really predict all the ways that this kind of technology would change our lives. It seems like it could be one of those things that's just a total flop, you know, like, oh, we thought this would be really cool, but you know, who cares, or it could be completely revolutionary and I really can't

tell which. Right, We're again, we're in the infancy, so there's no way of telling yet, right, But we're learning tons about artificial intelligence, swarm intelligence, modularity, robotics, computer science. There's so much that we're learning, chemical and physical property exactly. Yeah, all of this information is useful. Yeah, so the quantum entanglement with chili cheese fries, we're learning so much whisks, and I don't like to talk about it. Actually I'm

mandated not to um. But at any rate, no, it's it's really cool that we want Joe, look you that you're whisking your chili cheese fries until they're properly emulsified. It's close enough, alright. So at any rate, Uh, yeah, we're we're learning so much that even if ultimately we never see an implementation that approaches the ideal one, uh, we're gonna benefit in other ways that we can't even anticipate right now. And that's that's just the way science is.

And in fact, that's the why that's why I'm always very passionate about advocating for science, because you know, to have to advocate science to say that there's some in the goal and that that in goal is going to be valuable so that you can justify the funding of it is one of the banes of science. I mean, it's a it's a reality in our world. But it's good to keep in mind. But but it shouldn't be a limiting factor doing awesome research. Yeah, people want promises.

I mean they want short term gains, things they can see. Well, I mean it's if you're if you're ultimately having to justify giving money to someone that I can understand. You know, I understand both sides, right, I understand the psychology that goes into both sides. Uh. That's why I'm really hoping we can get to that Star Trek economy we talked about a few episodes ago, so that we no longer have to worry about money and we can just end up, uh,

pursuing scientific endeavors without any limitations. Yeah, jumpsuits and with little communicators. Never forget the jumpsuit. You know, I can't. I've tried, all right. So that wraps up this discussion about claytronics and programmable matter. If you guys have any questions or suggestions for future topics, let us know. Send us an email the addresses f W Thinking at how Stuff Works dot com, or drop us a line on Twitter, Google Plus or Facebook. At Twitter and Google Plus, we

are f W Thinking. Just search f W Thinking and Facebook. We'll pop right up. Leave us a message and we'll talk to you again. Really soon. For more on this topic in the future of technology, I'll visit forward thinking dot com h brought to you by Toyota. Let's Go Places