 And I think we are live. All right guys, we got a special last second thing, let's say workshop today or webinar. And we got two cool guests on today. The title obviously you can see is kind of like a new blockchain protocol. What is a total chain protocol? So we have two of the co-founders of the total chain. We have Tufi, Saliba. And so Tufi's background, he calls himself a serial techno entrepreneur. His track record is in algorithms, machine learning and cryptography. Tufi has built and sold IP that ended up in Google, Intel, OpenTex, HP, and he is the CEO of Total Corp and Privacy Shell. Then we have Dan Taliber. Dan cut his teeth in building mission critical solutions for NASA and Fortune 500 companies. He creates solutions in problem spaces like distributed systems, programming language semantics and databases and overseas development and innovation as CTO of Privacy Shell. Guys, welcome to the show, how you doing? Thanks, Amir. It's been having us, exciting. Cool, so the reason why we have this show is Tufi, when was it? I don't know. Couple of months ago you introduced me to the total chain and I still don't know exactly how it works. So the whole reason why we have this is I actually want to kind of get more transparency in education on my behalf on exactly what the, what is it? Sure. So perhaps before we get to this, we need to talk about why we're doing this and how we got here. So Privacy Shell is a startup factory. We have about, we're like the co-founder of 22 startups. We're like the minority co-founders. And most of our startups, they have that theme of like privacy and security by design. So we use the blockchain a lot for a lot of reasons, not just fintech. And we like the blockchains for decentralization aspects. So we don't have to have someone in the middle doing certain thing and can be trustless and all of this stuff. So we got to the point where we felt that the scalability of the existing blockchains that are out there, they're not really, this throughput wasn't anywhere near of what we want, even one of the companies to benefit from what it needs to be. Yeah. And in the same time, the cost per transaction was so prohibitive for a lot of the things that we're working on as well. And when I say cost, I'm not talking about fees. I'm talking about the actual cost, the net net cost of humanity. When you look at Ethereum today, consumes more electricity than the country of Cyprus. Yes. Combination of Ethereum and Bitcoin, they're on a chart of like countries, they come about number 72. So they're more than 120 other countries. Okay. So we can't, I mean, the cost can't be hidden from the user for so long, but it can be hidden forever. So scalability cost. And then when you start looking into decentralization, some people, they say, oh yeah, it's fully decentralized. But then other people, they're like, well, when you have like miners, they can sit in one room, four or five of them that they can pretty much manipulate most of it. Or if all the concentrations of miners they're in a certain country that may one day not like it or whatnot. So that's debatable from that perspective, but we'll leave that aside. We'll let's assume that it's fully decentralized. And when you look at the actual users, how effective are they in the actual decentralization? And we decided to take on a quest of complete decentralization driven by users. And if we're able to achieve that, then we need to also do distributed computing. So then the actual computation to run the system, they're not like, you know, solving hash cash or whatnot like you do in proof of work. We need to innovate on that as well. So we came up with something that's called the proof of actual work. Okay. So when we dive into that, what is that actual? What's the difference between proof of actual work and then the proof of work? What are the major? Yeah. Sure. And before we actually talk about what proof of actual work I need to talk about how we got inspired with that and give credit to a lot of people. Yeah. Here started with, you know, Adam back in the 90s and then a lot of folks that the innovation on that and built so many systems that they can utilize that in so many different ways, you know, like Nick Sabo and then you get some other folks like Andrew Miller with the proof of useful work. So the useful work was like utilizing something that is needed by humanity and that is useful, you know, and running that so that it's not wasted computation for no reason, you know, just like. So we got inspired by all of these and felt, okay, so what if we can get the actual work that is needed? Is it distributed computing needed to run the system? Okay. The existing system so itself. So it's, so you need a lot of distributed computing to run a system to be global, to run a lot of, to have a high throughput. And if it's, you know, the actual system itself, then we call it proof of actual work because it's doing work distributed computing that is needed to end the security. And usually proof of work, it's like something that's going to need a lot of work but to confirm it, it's just like tiny little work. It's like anybody can confirm it just instantaneously. So similarly with proof of actual work, it's going to require a lot of computation work which is distributed computing to get to something that can be verified instantaneously. Now what is that thing that can be verified instantaneously? So think of that as the Merkle route. Now I'm assuming that most of the audience understand what's the Merkle tree and the Merkle route and if they can't, they probably can Google that. So I'm not going to get, stand a lot of time explaining what's a Merkle tree but it's basically a binary tree of hashes and each and every hash think of that as a transaction that happened. And then when you combine all of those, you get to the route. Now currently with existing blockchain, as much as people, they call it distributed ledger technology. I'm sure you've heard of that. It's actually replicated ledger technology. So I'm not sure why they call it distributed. It's not distributed ledger okay? So, but it's great that they call it distributed because it's very inspiring if you were here. Oh wow, it's distributed computing. From that perspective, it's great. So with the actual building the Merkle tree in a distributed way needed a lot of innovations, a lot of research, we started working on that. And we got to the point where we were able to have an algorithm that helped each and every device to contribute into building that Merkle tree with the replications of 32, not 9,800, 10,000 whatever because you need to have a lot of fault tolerance. So we call it pseudo random function in choosing who's doing what each and every time which depends on the previous Merkle route. So each and every implementation can make some changes to that whether you want the nodes to be an actual smartphone, ideally you want them to be smart phones because we know that globally, not a lot of people that have the ability to buy mining machines and the computers running 24 seven whatnot. And if that's not the case, so to build a truly centralized system, we need to look at what people can have for the foreseeable future. And if they have smartphones and perhaps we need to make it possible to run a smartphone. So we do have some implementations that are completely on smartphones. Some other implementations, they run on a smartphone with metering in the cloud and others, they run on a combination of both which we're going to be making announcements soon, certain partnerships with some of the best ledger based blockchains because we are ledger less. Okay, so when we tell people we're ledgerless blockchains and they look us up, they're like, what's ledgerless blockchain? They come back, they're like, I asked my CTO and he said, there's no such thing as ledgerless blockchain. I'm like, yeah, because we just called it yesterday. Maybe CTO depends on the memo. You're going to have to pause there and educate myself on that, but maybe you can simplify that for us and explain to us exactly how does that work? Sure. Think of each and every, so I need to take a step back before we actually get to the note. Let's think of what's the value that it's being exchanged and how we're managing that exchange in a decentralized way. Now, when you tell folks about, for example, Bitcoin, if somebody's like just want to learn what's Bitcoin or Ethereum or Dash or Zcash and tell them, okay, your phone is becoming your bank and your coins are sitting on your phone. And I'm sure you've heard of that several times, but in reality, you and I know the coins are not sitting on any device. The actual private key is sitting here. The actual coins, they're just like a number in a ledger. And that number in a ledger is actually series of numbers of when you got at first in what's plus, minus, plus, minus, plus, minus. It gets to, this is your wallet number and this is what it has in a series of things and adds up to, let's say, 15. You have 15 Bitcoin. If you were to go and send then 16 Bitcoin, whoever is managing that ledger will not allow. And if they do, the majority of them, they're not going to. And they're always assuming that the majority are going to be good. By the way, in any decentralized system, you have to have the assumption that the majority are going to do it. So if somebody's listening in the view of the world that the majority of the world is so negative, 90% of the world are bad, 90% are bad, you should choose a different field because this is not going to work. You have to have certain assumptions, okay? Yeah, there's actually, there's a cool, there's an interesting saying I heard before where it's like an optimist sees a color green, a pessimist sees a color red, but a truly wise person sees no color. Yeah, yeah, that's right. Yeah, we're kind of color blind, so. Which is very hard to kind of navigate the traffic sometimes, especially when there's construction because they don't have like those series of lights. So we started, I'd say it's like, when you need that value to be exchanged and there are certain machines that they have that ledger, those machines are being managed by what they're called miners and if you were to find a way to kind of separate them into so many different machines and it becomes like way to cumbersome, it's doable by the way, like there's some folks in Ethereum that are working on the sharding and all that. We don't know what the outcome of that, but it gives you some scalability. Every time we look at some scalability that is out there, we see that it's either off-chain, just I don't wanna say it's terrible or horrible, what not, but it's just like, if you have the blockchain and you do off-chain, maybe you're not doing it on a blockchain, so it's not as beneficial as it could be when it's on a blockchain, objectively speaking again. Otherwise, you don't wanna use a blockchain. So if you wanna do it on-chain and if it's decentralized, then you need to find a way to actually get those ledgers to grow or be sharded in a certain way and if it's sharded in a certain way, then it gets too cumbersome and may or may not work. We decided to take a step back and be like, okay, we're doing value exchange. So if we are doing value exchange, let's think of any object as an atomic structure. And if you think of an atomic structure that I wanna send it to you, if it's a file, let's say it's text file and I'm sending it to you, I can send it to millions of other people using HTTP, but each and every one of them, they can claim ownership over that. Now, can we have a protocol that can ensure that the ownership of that file belongs to only one person at any given time? Then we start working, okay, so what if this file has a unique number for the life of it? How can we ensure that it can have a unique number for the life of it? Then we start building something that is similar to the IPv6 and we call it the TodaT, T stands for tree. And I get to that while we decided to just stop at the T later or later. So the TodaT protocol is based on huge binary tree that it's like it's got 256 levels, okay? Any binary tree that has 256 levels has more points in it than estimated atoms in the universe multiplied by 10 and I don't know, 15 zeros next to it or whatnot, 16, something like that. Anyway, so you can have like a billion, billion universes represented on that tree if you really want to. So you never really have to build that tree anywhere, but every computer that ever existed knows how to navigate binary tree. You give it the proper instructions. So now the very bottom of that tree, the actual, or if you look at it this way, the actual leaves at the top of the tree, depending on what we look at it, computer scientists will look at the tree from the bottom, the leaves at the bottom, the roots on the top. Those leaves we call them quarks. Why we call them quarks? Because many quarks can form an atom, okay? Now, if you're sending an object from one place or another, it can only send an atom, you cannot send quarks. And those atomic structures are at level 32. So now they are at level 32, they can have two to the power 32 quarks inside them. So that's how much they need to kind of get formed in the first place. And I'll get to how they get formed. They require certain energy and network in combination of those computation that cannot be forged in the way. We kind of do the opposite of burning energy. We kind of like save energy in order to create value here. So you have that atomic structure. And now that's the unique number for, or that's coming, that cannot be forged and you have it and you need to send it to them. And you take it and send it to them. Now what happens to the actual file inside the file, you can have the proofs, which we call them like the actual Merkle proof that they can be created by up to 32 nodes using that method that I described earlier which is like the actual distributed computing. Each and every one of them, they're racing to get then the Merkle route, because then it's receiving and then also make sure can I receive it from a mirror or not. And if he's able to receive that proof, then now he has a proof that he owns this file and that proof he can take it and give it to Tufi. And then Tufi's going to look at that proof and he knows that then owns it. Now there's, I'm skipping a lot of details here, but I'm just giving you like at a high level, one file can actually grow linearly while technology grows exponentially. So we address that by the way at some point. Now when you think of that, you might start thinking, okay, what if for example, if Adam is back to watching right now, he'd be like, okay, what about fungibility? Well, we can address fungibility. There's some modules for that. Right now I'm describing at a protocol level, we introduce something similar to the IPv6, but rather than just giving it to the devices, we're giving it to every single object in that system. It's gonna have a unique number. Imagine you have your IP number. Long time ago, people that used to make that comparison, it's like, IP, what's IP? Well, it's kind of like a phone number. It's my IP, it's yours. That's how computers can communicate to each other. So think of every single object in your house has a number, okay? If you own them, they have that number. And that number is unique in the universe. No one has that number utilized. So if you take them and give them someone else, they belong to someone else's house. Yeah, I don't know if that metaphor makes sense. It's just trying to simplify it here. Dan, did you want to add anything to that? Yeah. Did you have any questions at this point, Amir? Or should I just jump in? I'll just jump in, let's keep on going. Yeah, yeah, okay, great. So I want to take us through a slightly different route to get to the same answer, just to kind of give a fuller perspective. So one of our original driving use cases and motivations for building this was and still is supplementing cash and cash primary economies. So cash has a lot of great properties. It has basically no transaction fee at the transaction site. It has a pretty good story about anonymity and privacy in the sense that when you and I perform a cash transaction, someone halfway across the world doesn't have to know about that. And it has a very good story about scalability in the sense that billions more people could be using cash and that won't slow you down at the site of transaction. So it doesn't actually impose any transaction overhead. Now, why is that? It's because with respect to that transaction itself, so kind of ignoring the management of cash and general and the printing and destruction of it, but at the site of transaction, the amount of work that you were doing as part of that transaction is independent of the size of the network. So it's a constant factor with respect to how many people are using cash in total. There could be billions of other planets out there if people are using cash or entities using cash and it wouldn't make any difference to us. So the cost is completely independent and the total amount of work done to perform that transaction. Cash also has some not so good properties like if you have a lot of it and there's no bank nearby, what do you do with it? How do you get it to somebody across the world, right? How do you buy things from other places that aren't nearby you? So there are lots of issues with cash, especially if you don't have banks nearby. There are about 6 billion people on the planet who are unbanked or underbanked and of those about 2 billion roughly have smartphones right now. So if we can find a way to supplement cash in cash primary economies, that A, doesn't deplete the value of that regional economy either through transaction fees that get withdrawn and then sent somewhere else in the world or through other depletionary effects like currency being created in response to those transactions somewhere else in the world but not within that regional economy which eventually has an impact on the amount of value stored within that regional economy. So that's one thing that we need. If we don't deplete value within the regional economy we need a solution that scales very well, clearly. And we also would like to have a decent story on anonymity, privacy, governance. If we can do that though then and this one's on smartphones. So there are lots of conditions and requirements here but if we can meet all of those requirements then we have an opportunity to start to bridge financial products onto this network. So we can put on this platform things like savings accounts, low interest loans, insurance and that becomes a very exciting opportunity. So this was one of the driving use cases that we had in mind for Toyota. Tufi mentioned another one which is that we have applications that some of which are making very good use of ledger based technology for doing things like PKI, public key infrastructure so that we can share our public keys for example. But others would like to be using it but they require higher scalability than is currently available. So with this use case in mind this how do we supplement cash and cash primary economies? We have an interesting problem because ledgers offer like a global public ledger in particular offers some very nice guarantees given some particular thing I can then say who owns this thing. And I have a guarantee that I can find out which account actually owns that thing, right? And that's wonderful. Conversely, if I have an account I can say which things does this account own? And that's also a very nice question to be able to answer. So in cases where I need to be able to answer those questions I absolutely need a global public ledger which I need that guarantee. Which we can talk at some point when we're making announcements happening we're next the day or two about the partnership that we're doing with one phenomenal efficient ledger with the touring award winner Samuel McCalley, it's called Algorand. I'll save that for when the actual announcement happens. If I don't need to ask or we could ask is answering those questions necessary to have decentralized value exchange? And the answer as we know is no it isn't necessary and we know this because again we have cash, right? I don't know how much cash you have in your pocket right now. And given some $10 bill with its serial number I don't know who in the world owns that, right? If I have some serial number I can't figure out whose pocket that is in. So we know that we don't need to be able to answer those kinds of questions in order to have a value exchange mechanism. And then the question is can we get any leverage out of this? If we do away with the ability to answer those questions what is the minimum amount of work that we can do on the network in order to have a secure value exchange mechanism and is that amount of work a constant size with respect to the number of users of this network? And the answer to that second question is no it's never going to be a constant size because as soon as you say that you have a network of users and they're communicating then you have at least a log factor, right? Just because of those communication pathways but log factor is pretty close to constant. So if we say there's going to be a cap of a trillion devices using this then we can say that that constant factor is 40 if it's log base two and then we're good, right? Like we've basically got our constant factor there as long as we can keep it to a log. So maybe we can do this maybe we can actually build something that has scaling properties such that if the average number of transactions per user is constant let's say per day. So let's say 10 per day or 100 per day, whatever it is then we can essentially scale arbitrarily the number of transactions per second that the platform is capable of by scaling up the number of users. Does that make sense? Okay, so how do we do that? Well, let's give ourselves some leverage what we're going to need as much leverage as we can. So one thing that we can do is to say well, we're hoping to supplement cash. So maybe we can give ourselves the ability to have atomic units of currency. And if we do that, that gives us a lot of leverage over a situation where we have units of currency that can be arbitrarily broken down or recombined, right? We have to solve a much simpler problem now. So we have atomic units of currency that can be transacted so I can transfer the ownership of this. And so the semantics become very simple, right? I can prove to you that I own something and independently of that, I can also transfer that thing to you. Now that proof doesn't necessarily require that we hit the network, which is a nice property. I can prove to you in the middle of a desert that at the time that we both went offline, I owned this thing. And if that time was fairly recent, then you know that I actually own this thing. But then we need, it would be wonderful if I could just go ahead and transact and transfer that thing to you in the middle of the desert with no network. But short of having trusted hardware that we both trust, there's really no way to do that because I could just keep a copy of it, of course. And so I could say that I've transferred it to you, the transaction actually works, you get it. But then I have a copy of it still that I own and I just keep transferring that. So if we have trusted hardware that we both trust, we're fine, we can actually solve that in a purely network-free way. And then we get our constant factor and it's wonderful. But short of trusted hardware that we both trust, we can't do that. So we need the network to be able to do that transaction. And now we're in an interesting place because we, let me step back one step further. How do I prove to you that I own this? If we look at this model where we have trusted hardware because it's a little bit simpler, then all I actually need to do is have a series of transaction blocks effectively that show the signature of the previous recipient transferring it to me, the signature of the previous, previous recipient transferring it to them. And so on, back to the origin of this particular file. And if we have that in the trusted hardware world, then we're effectively done. So what does that look like? It looks like the kernel of the file plus a series of transaction blocks that show this transaction happening. And what does that look like? Well, that's really just a blockchain. So what have we done at this point? We've taken the parts of the global public ledger that refer to this particular asset, hold them out of that ledger and push them down into this file itself as a part of this file. So we're at a pretty good place now, but we still have to solve this double spend problem where I could give it to you and give it to someone else. And to do that, we're going to need the network. And we have an interesting opportunity here, which is that in a lecture based system, you absolutely have to prevent double spend. So the protocol itself has to give you a hundred percent guarantee within the protocol that you cannot double spend. And it has to do this because if you could send both me and Toofy a million Bitcoins, for example, and it was the same million Bitcoin and that went through and that double spend happened, you got lucky and the bad thing happened, then that fundamentally breaks the Bitcoin ecosystem, right? Now there's 22 million at the end of the world or something and everything is broken. If you were to send me a Bitcoin, I can assure you nothing bad would happen. That's okay. That's okay. That's okay. Sorry, I just had to go to it. Yeah, so we have to have this guarantee in that world in ledgers that you absolutely cannot double spend because there's no limit on the amount that you can spend. And so being able to double spend can fundamentally break your economy. However, the counterpoint to that is if you put a hundred percent guarantee that you can never double spend, then you're left in a situation where I only have a probabilistic guarantee that the assets that I think I own are actually things that I own. So the money that I have in my wallet is only probabilistically in my wallet. And there is a bad thing that could happen. And if that bad thing happens, then I no longer have the money that I think is in my wallet. And- Because again, they're not really in your wallet, right? Right. So that's right. So the probability that someone is going to take it away from me sort of decreases over time asymptotically towards zero and that's a nice property. But we're in a situation now that we have these atomic units of currency and we're transacting at this level of atomic units of currency. We have a situation where we could flip that around and potentially say when you receive one of these, you have 100% guarantee once the proof goes through that you own this. So you absolutely own everything that you own. However, there is a very small probability that if someone based on the amount of the network that somebody owns, so if they own 10% of it for example, then there is some small probability that if 10% of the network is colluding that they would be able to make a second copy of that thing. So there is a small probability of counterfeiting. And as long as we can keep that probability small enough that it is economically infeasible for anyone to do this counterfeiting, then we're in the same situation that essentially every government of the world is in with respect to their fiat currency. It's kind of like, think about it. You know, if you really need to spend $150 going for $100, you really want to do it. And chances are, you know, but if you do it, if you do it, it's possible it would like 0.0000, I don't know what, maybe two. Can we rewind back a little bit and kind of do a deeper dive on this atomic units of currency? Yeah, definitely. And we can do that. And before we get into that, we need to give some overview of what we are building. Because sometimes we get mistaken that we are building currency. We're actually building a technology. I understood what Dan said. Going back with the ledgerless system over here for the Tota Chain protocol is, you know, for example, with the blockchain, everything is hashed from previously. So you see it's all linear, it's all in order, that you are taking data files. I'm not too sure if it's from the header somewhere within the block. You're taking those data files and subtracting it to the actual file itself. That was a little bit metaphorical. So I was saying you could think of this as a decomposition of a global public ledger into the set of assets that are being tracked through that ledger and having each of them track their own ledger independently. But I think that's an illuminating way of thinking of it, but we shouldn't think of that as what's actually happening. That's really kind of metaphorical. So what's actually happening is that you have some file. Yeah. That file has a kernel to it. So that kernel has a little bit of structure. Yeah. Part of that structure is a payload, which combined with the rest of the structure gives it its unique identifier. So essentially we take the hash of the kernel and that's the ID of the file and it's guaranteed to be unique up to collisions because of the structure of the kernel itself. Okay. So once we have the kernel, then we essentially the very first thing that happens upon creation of one of these files is that it is assigned to an account and that happens through creating the first of these transaction blocks. So then we have a kernel and we have a transaction block and if I own that file and I transfer it to Toofy, then that is going to end up with a second transaction block connected to that file. So now we have two things. We have a file ID, which never changes. It's always the hash of the kernel of that file and we have a file value, which is a hash of the entire thing with all of its transaction blocks. And that hash value increases with more transactions, correct? The hash itself doesn't grow, but it changes as you add more transaction blocks. The value itself changes. Yes, the value of the hash changes, exactly, as you add more transaction blocks. Got it. This is where you can start thinking about it that everything that is happening is on chain, but that's ledgerless, okay? So this is different than when folks say it's off chain, but it's like ledger based or whatnot. This is on chain. Everything that is actually happening is part of the blockchain and that can be utilized. And this is a point that I wanted to make earlier. A lot of folks when they see what we're doing, they think we're okay, we're building our own cryptocurrency whatnot because it seems like the trend nowadays, people are going to go and build their own cryptocurrency. At this stage, we don't have any implementation that we're building our own cryptocurrency. So we have certain implementations that people would want to use the Toyota to increase certain ledger performance or whatnot. For example, if you think of Bitcoin, if you take each and every Satoshi and you put in unique number for that Satoshi for the life of it, there's a way to actually, I'm not going to get into the details now, but there's because as soon as you get those kind of details and a lot of people start thinking, whoa, why don't you do that instead of, so there's a way to actually do that without even having to fork while in the three transactions per second that happens on Bitcoin as a throughput, you can think of each and every one of those transactions can be multiplied by three million if you were to utilize the Torah. So the Torah's throughput can be three million transactions per second if it needs to be three million. So if you have two billion users, if you have like two million users, it can be 3000 transactions per second. So it grows as the user base grows. And that's another thing that is super cool about what we're doing, that it's not just the throughput that it's available as if you're running a database and you can, hey, look, Amir, I can show you, I can run three million transactions per second. Well, in order to show you, I need to get two billion users first. So if you have two billion users, I can show you. If you have two million users, I can show you 3000 and so on and so forth. So it grows as the size of the users grow. And that's very important thing to keep in mind because it's ledgerless, we're able to get to this scalability because as soon as you save three million transactions per second, any database expert, they say, oh, you know, BS, this can't be because if you grow three million transactions per second, that's a growth of petabytes per day on a database. Network cannot handle that. There's no server can handle that and the storage and all kind of stuff. Well, that's why we don't have a database. That's why it's ledgerless. This is a hugely important point to achieve the kind of scalability that we're looking for. Even if we discovered ways of scaling global public ledgers to have the kinds of efficiencies that we would need computationally to achieve the scalability, the data increase per day is simply too much to sustain it. So the answer to that is don't put all of those transactions in one place. Just like cash transactions, keep them local. And as you build up the consensus mechanism that the network actually supplies, then limit the amount of information that's being shared so that at each stage of that consensus process you have half as much information that's being shared until you get to the end. And the only thing that's shared globally is a single hash value that represents the root of this entire Merkle tree. So now we have consensus on the state of the network but no one can actually do any queries at all because all that is shared is this global root. So I can prove to you that I own something by showing you the Merkle proof that traces from that root to the thing that I own and goes through my wallet along the way but you can't actually query where something is right now or what does this account own? There's no way to do those kinds of queries. If you need those then you absolutely need a global public ledger but if all you're concerned with is exchanging value then you can do that in a ledgerless way. It doesn't give you the qualities that a ledger gives you and you can't answer the same questions but it does give you the ability to do this type of transaction. By the way, what Dan just mentioned that's a breakthrough and that's unprecedented. We take pride of that. We don't want to take credit for building this protocol all on our own. We have over 20 people that helped us get to where we are today and at some point would like to credit all of them but from now until then we have one thing that is working to our advantage is that what we just told you most people do not believe it and if they do then we won't have any business because then they're just gonna go and do it. So we do have the advantage of actually running a lot of businesses and so far we have a lot of joint ventures that I'm sure I'm very aware of some of those. We have a lot of partners. We have a lot of clients lined up and what we do we build that technology and if you want to run a global decentralized bank for $12 million to $100 million we can provide you with the entire package that has over 30 patents. It includes a lot of other cryptographers in space and whatnot which will also include the total. So that includes some ledger includes like a lot of stuff like that. So that is our business. So the business so when people do things that's okay while you're really in cryptocurrency and you just want to take over a dash and you want to take over Zcash the entire cryptocurrency penetration in the world today is 0.008% okay and that's the $100 billion and people are like whoa you know like you don't want to come and take us over. We're not after any single cryptocurrency we're after replacing the 99.992% that exists in the world that is not cryptocurrency and before we even get there we need to increase the size of that pie. Like Dan mentioned the 2.5 billion people that they are unbanked at all they're not using dash, they're not using Zcash they're not using Bitcoin, they're not using anything they're just using cash okay. Those are the first set of people that we think they can be powered and many of the businesses that you're partnering with they're going after the emerging market and you have three countries in Africa we have number four we could be contemplating over the next while which will be an announcement of how the Torah is actually being created because that's a partnership with an energy company. We'll save that for later because we basically do the opposite of the burning energy where like the saving of the energy is actually what builds up that value. Well that's why I want to talk about what you mentioned is atomic units of currency I just want a little more clarity on that. Yeah so think of we mentioned the actual unique number of each and every file and if you think of that file that it's growing linearly inside the file you're actually having data that it's the Merkle proof of each and every transaction. There's something that is common between all the nodes in the system it's the Merkle root of every single block. So you have the hashes of all the blocks and this is why then when he mentioned earlier is that if you are in a desert with someone else you're able to prove to them that you own something because when you show them and when we talk about you prove to somebody and what not this is all cryptographic proofs okay so it's like your machine is moving to the other machine. Your machine already knows of that Merkle root that happened let's say two weeks ago and the person who's with you is trying to show you that they have million dollars worth of TOTA they're able to actually show you that when they received it two weeks ago the Merkle proof that they've received those TOTAs they got to the Merkle root that you already have and that's a cryptographic proof that can be provided anywhere without any access to the network to people that can just like get in via you know just wifi or any connection they can get with each other and can have that proof. So that's what's being built you know inside that file and that file keeps growing linearly the other thing that it's shared across the network each and every node is aware of in every implementation with the last wallet or machine whatever you want to call it and with the last note and we call it note some people they call it coin or whatever you want to call it okay. There's a reason by there why we call it note because there's a lot of scams out there that they put anything put coin next to it. It's like they're trying to bend from the name coin we don't want to bend it from anything we're just ready to it's a note okay. So this is basically how it works at very high level who work to dig in deeper into algorithms and whatnot it's probably not a good idea because we're keeping there's a lot of stuff that we were building in and we're planning on launching over the next while which then we'll talk about the actual open source and what's going to be covered and you know who we're going after and whatnot but there's certain elements that they would stay in core until they actually released and once they release they're all open source okay. So it's not like hey we're keeping that in core we're not sure we're going to show it to anyone but there's a reason if you think about the evolution of hacking for example if all the hacking that is available to most people that you know of today just the knowledge of how to you take it and give it to people eight years ago you won't be able to run Bitcoin for more than a day for example. If you give it the same for Ethereum if you give it to two years ago or three years ago you won't be able to run it. So it's very important to know that and not to just be super open and this is a message to almost everyone who are building out there in the open. It's good to keep that to your core team because there's certain elements that you want your core team to get there first before you kind of divulge it to a lot of the world because sometimes the hackers they're sitting and listening and catching up to everything all the work that you're doing and they try to get one step ahead of you. You want to keep that for the moment to release so then it's the knowledge of hacking is a bit behind you. There's a way to hack Bitcoin if you're interested you can find there's like in DEF CON last August in Vegas there's a lot of folks who can actually show you proofs of how to do certain hacking on Bitcoin. If you use those same methods a few years ago you would have been anyway so I won't get you to all. Yeah that's a complicated and nuanced topic. I want to jump back just a little bit to what Tufu was saying about proving things to each other in the desert. The use cases for doing that with currency you can find some but they're not too common but when we start thinking about these files that you were talking about in there these atomic units as being things beyond currency then suddenly we can see a wealth of use cases for being able to prove that I own something without transferring it. So the thing that I own for example might be some credentials. It could be a visa, right? It could be a work permit or something that allows me access to a particular thing. It could be a credential that says I'm over 21 so I can go into a bar in the US for example. And so I think what we're really interested in with the Toyota protocol ultimately is building a long-lived level playing field platform for the management of ownership and provenance in general not just to do with currency or with cash. So that's really sort of the broad goal and within that broad goal this idea of atomic units makes a lot of sense within the smaller goal of dealing with currency itself and dealing with ownership and provenance of currency. Of course there are problems that need to be solved there because in our daily use we have a range of maybe five orders of magnitude in terms of the currency that we use and then maybe another two orders of magnitude above that in terms of what you could reasonably expect to put on your credit card over the course of a given day. So seven orders of magnitude is a lot to cover with atomic units of currency and it does raise some complexities that we have solutions for but they're not perfect solutions. So this is something that we're kind of working on for the next phase. It doesn't actually impact the core protocol. It's more things that are being added on like value add systems around that and on top of it part of this bigger ecosystem that Tufi has alluded to. I don't know if that answers the questions that you had in mind when you asked about atomic files or atomic units. No, that clears up some things. I still do have some uncertainty but it's becoming more clearer to me. Okay. Did you have particular questions that you wanted to ask? We've got, I think 15 minutes left so we can dig a little deeper. Yeah, I think a lot of people are also gonna have the question of the laws of entropy. So for example, if there's like no transaction fees how does that act mathematically, how does that work out? Yeah, so this is important. It's not the transaction fees. It's that the transaction fees don't deplete the value of a regional economy, taking value out of that region and bringing it to another geographic region instead. And this is where we're sort of trying to jump in here for a second. A lot of people, they confuse costs and fees. So the cost of running the total network so far has been if you really want to get the three million transactions per second for two billion people and you want it to run the entire time three million transactions per second for every second, okay? The net net cost of humanity is equivalent to running a video chat for 19 minutes per day for each and every individual. Okay. So when you compare that to almost any other cost, this is a much lower cost because it's kind of divided between everyone. Now when it comes to what part of that cost you want to take it and split it into fees versus creation of a new value. For example, Bitcoin has most of that cost and creation of new currency, okay? So every 10 minutes, there's 12.5 Bitcoins, right? Now that's a cost. The fees, they can add on top of that that gets me put into the users or the user that can just like add the fee and then so they can get their transaction. But that cost is hitting the users indirectly whether they admit it or not because if you were to get all the sum of the users, they are paying for that indirectly. It's not like those miners are just like some altruistic folks like, oh, you know, it's okay. We're not gonna charge you $4 billion a year for all the electricity we're gonna consume and all of this stuff. You want to charge it and you want to make profit and you want to build a business and you want to see how they're going to grow over the next two or three years to be 10 times bigger than what they are today and having all of those plans. So the net net cost for Torah is equivalent to video but the fees are split between tiny little fees. You don't even see them, like the users won't necessarily see them, but they're more around like the actual crypto economics of the system and combine some of the game theory things to prevent certain attacks because if you remove the fees completely without having crypto economics included into the entire equation, then if Dan and I want to get the system super tired you can start sending him like millions of dollars and he sends millions of transactions every second. If that happens, we end up causing a lot of the costs to both of us and the fees will be divided. 30% of the fee to me when I'm actually sending him 30% is his and that's 30% of the actual overall cost and 40% of that cost is actually being created in new value or new currency. If you were to use it for currency, okay? So I keep repeating that many times because it's like a lot of people they think that we're building currency, we're building the value exchange protocol, we're building the framework and we have several implementations of the platforms and some people are even building apps on top of that. So when it comes to Torah, we're building the protocol and the framework and what people want to use it for, they can use it for, I don't know, cars, ownership of cars, ownership of houses or money. We have certain governments that are actually looking at it for the first time, we see governments that are actually looking into decentralized way to deploy it so then the next government cannot undo what they're doing today. Because as much as they fear the corruption in the future, there isn't anything they can do about it but putting something decentralized and that's a part of decentralization. It's one of the part of decentralization is that once you actually have something running, it cannot be undone. And there isn't anything out there today, just all listeners, I know you have like a lot of viewers here, there isn't anything that is fully decentralized today that can actually get to the throughput of 1,000 transactions per second. There isn't anything, okay? We're talking about three million transactions per second and we stand behind it and we can prove it and that's how we're getting a lot of inbound. But again, I think it's always important to point out the trade-offs in that, right? So we get that by giving up certain things that we don't need for our use cases. If you need those things for your use cases, then you need, for example, a global public ledger and it goes on to straddle. We do have that package in one of the businesses that we're an up to in the next class. So again, I'm just talking from business perspective, I do have a fiduciary duty when it comes to business so we're not really doing anything altruistic here. My personal goal, by the way, just so you know, I mean my personal goal which I've told you before, my personal goal is to get to global prosperity, okay? But when I say that, this is not just an altruistic goal. When we get to global prosperity, that means everyone on the planet is a buyer of whatever goods that you're selling. That's good for you if you're selling stuff and it's good for folks that they're actually starving today. They're 12.5 million people dying per year from poverty. I wanna see different problem in the world. I wanna see problem where people, they can't buy enough stuff. All those stuff they don't necessarily need, that's a much better problem. And when they do that, that's good for my business, for your business, whoever's running business, and it's good for them too, okay? So that's, I just wanted to put that. Anyway, sorry then. Yeah, I was just going to go back to the transaction fees. I don't think we've talked about the role of validators within the system yet. So, we mentioned the prevention of double spend, so how does that happen? There are a set of validators who are chosen for a given file ID within a given Merkle route, so a given cycle or block. And those validators need to sign off on that transaction. So, all of those validators are colluding with the sender and the receiver. If there is any double spend happening within that cycle, one of those validators will spot it. So, we can pick, for example, 32 validators. I ran them from across the users of the network. Receiver both give them a small amount of incentive to perform this validation. And if any one of those validators receives a message from the sender that they are sending this file to both Alice and Bob, then that validator has cryptographic proof that a bad thing is happening. They alert Alice and Bob and everybody else in the network, et cetera. So, that's where the transaction fees come from. They come from the sender and the receiver paying those validators to do their job. Does that make sense? And because those validators are chosen from among the users of the network uniformly at random, it means that we avoid this concentration issue. And also, as long as we do this correctly, avoid the bifurcation of the class of people using the system into validators or miners and users, right? And that setup of concentrating mining power, of course, can make a lot of sense for a ledger. But in our case, the whole goal is to avoid centralized bottlenecks or these places where all of the information is going. Because as soon as you have one place as soon as you have a transaction pool, for example, you're already stuck. You've already lost the kinds of scalability that we're looking for. So, we have to keep the work for the transaction as small as possible with regard to the size of the number of people using the system. And we do that by picking individual validators from the system in a way that Tufi mentioned earlier. So, it's a deterministic pseudo-random function, if you will. So, I can prove later that I've chosen the correct validators and not just picked ones who were colluding with me. And now, they give their validation and I probabilistically give them a file in exchange, if that makes sense. Yeah, it doesn't. And the validators are picked by the protocol itself. Now, let's say this node becomes a validator. Are the validators then interchanged in the future? No, so. The transaction. Oh, it's per-transactions validators. The transaction has its own set of validators for that particular block or rather, each file, let's say, has its own set of validators for that block. And then that's that. And that's to the randomly chosen every block so you can't know who you're gonna validate for the next block. Okay, I get it. So, in any given block, you may be validating two or three or four, for example, right? Yeah. Yeah, right. Transaction here is the smallest kind of transaction. It's sending a single file from one user to another user. If you're sending 100 files, that counts as 100 different transactions. But again, we address that by denomination. So we have, so with the total tree, there's like each and every branch represents certain deployment. And within that branch, there's like denominations from like the actual inception of each and every note or the creation. And the denomination could be something, 10, 100, 1,000 or 0.1, 0.1, 0.1, and so on and so forth. And that makes it actually super powerful from the perspective of when you're sending transaction from A to B, often the average amount of notes that you're sending is about like 23. And that becomes a little bit more constant. Now you take that and you multiply it by 32. That's how many nodes are actually contributing into this transaction. It might sound that that's, oh wow, this is like a lot of work multiplied by 32 or whatnot. Compare that to 9,880 miners that they're running. Well, and that size is fixed regardless of how many people are using this. That's right, that's right, and then. Amir, when I said 9,800, I was just actually talking about one of the counts of how many miners that exist in a certain blockchain that is out there. I think it was the count of Bitcoin, and it could be today, I don't know how many miners today, but I didn't really look at that, something like that. But that's basically replicated. So whoever is listening when you actually run, you know, transaction on Bitcoin, this is actually being replicated 9,880 times across the globe and all of that, which is fascinating. I mean, Bitcoin is like the best thing that ever happened to humanity in the last 10 years, and I'm actually so excited about the AI integration with that, which is something that, it's a shout out, if you don't mind, I can talk about the AI decentralized. AI decentralized is another initiative that I'm running on the side of things with the help of the two gentlemen that are here as well. And it's basically an initiative to collide AI experts with the blockchainers. The intent here is to get the expertise that is happening in the evolution of the science of AI to be collaborated with the evolution and the science of blockchain. The way we see blockchain is actually a branch of AI, in a way, you know, it's like, it's a multi-agent systems, anyways. I mean, then in our background is, you know, AI and Cryptography anyway, so it's like we kind of see them as one and we would like to combine those two sets of experts in one, they're 870,000 global AI experts. Our goal isn't over the next 18 months to combine them with 280,000 blockchainers and that's an ACM initiative. And by the way, I kind of, you know, I chair the ACM conference committee globally. So I'm doing that and we have ACM. ACM is the number one computer science globally, 110,000 members. So just in case people don't know what ACM is, that's also shoutouttoacm.org. So yeah, that's all, if you have any questions. I think that's a good place to end off on the show. It's been an hour, so we'll cut it off over there. I just want to thank you guys for kind of educating us. I have some more clarity now. Obviously I'm not 100% yet knowledgeable on how. We're going to have to do another five or 10 minutes. So you know, I mean, cryptographers spend over 10 hours just to wrap their hand around it. So you did phenomenally well in one hour. So thank you for having us. Well, thanks for coming on the show. If people want to get more information about the TOTA protocol, what's the best resource? You know, tofeeatprivacyshall.com or danatprivacyshall.com, that's email. Or they can go to totachain.com and send us something there. So yeah, we're super excited about this space and we're hoping that the entire world could get to be decentralized for the sake of future of humanity. So we're building the tools for it and we're hoping Homo sapiens will use them. Yeah, we will. Awesome guys, well thank you so much and I'll talk to you guys soon. All right, bye-bye. Good night.