Space Summary
The Twitter Space AMA with PraSaga CTO David Beberman hosted by PrasagaOfficial. Delve into the future of blockchain innovation with PraSaga's CTO, David Beberman, as he discusses the development of the world's first layer 1 blockchain centered around smart assets. The AMA session highlights the significance of smart assets, scalability challenges, regulatory considerations, and community involvement in shaping the next generation of blockchain technology. PraSaga's distinctive approach to layer 1 blockchain development and its impact on traditional industries showcase a forward-looking vision for blockchain ecosystems. Explore the potential of smart assets in revolutionizing decentralized applications and asset tokenization.
For more spaces, visit the Unique Projects page.
Space Statistics
For more stats visit the full Live report
Questions
Q: What defines smart assets in the context of blockchain technology?
A: Smart assets are programmable digital assets that can represent real-world assets or unique properties on the blockchain.
Q: How does PraSaga differentiate its layer 1 blockchain from existing solutions?
A: PraSaga aims to prioritize smart assets, efficiency, and scalability, offering a distinct protocol for next-gen blockchain development.
Q: What role do smart assets play in transforming traditional industries?
A: Smart assets enable fractional ownership, automation, and secure transactions, revolutionizing industries like real estate, finance, and supply chain.
Q: What are the key factors influencing the success of smart asset blockchain projects?
A: Factors like governance, security, utility, and ecosystem support are vital for the success and adoption of smart asset blockchain initiatives.
Q: How does PraSaga address scalability challenges in layer 1 blockchain development?
A: PraSaga utilizes innovative consensus mechanisms and network design to enhance scalability and accommodate growing demands in smart asset blockchain applications.
Highlights
Time: 00:10:45
Smart Assets Revolutionizing Blockchain Discussing the transformative impact of smart assets on decentralization and asset tokenization.
Time: 00:25:19
PraSaga's Vision for Layer 1 Blockchain Exploring how PraSaga's blockchain protocol reshapes the landscape of smart assets and decentralized applications.
Time: 00:40:55
Regulatory Compliance in Smart Asset Development Insights on navigating regulatory frameworks and compliance requirements for smart asset projects.
Time: 00:55:30
Community Engagement in Blockchain Innovation Emphasizing the importance of community collaboration and feedback in shaping successful smart asset blockchain solutions.
Time: 01:10:15
Scalability Solutions for Smart Asset Networks Strategies to address scalability challenges and enhance network efficiency in smart asset blockchain infrastructures.
Key Takeaways
- Importance of smart assets in blockchain technology.
- Building the world's first layer 1 blockchain.
- Innovations in blockchain for smart asset development.
- Challenges and opportunities in blockchain for smart assets.
- Implementing smart asset technology for diverse industries.
- PraSaga's unique approach to layer 1 blockchain development.
- Future implications of smart assets on blockchain ecosystems.
- Interoperability and scalability considerations in smart asset blockchain.
- Community engagement and collaboration in smart asset blockchain projects.
- Regulatory landscape for developing smart assets on layer 1 blockchain.
Behind the Mic
Morning Greetings
Good morning. Hello, David. Hello. Good morning. Good morning. Good morning, Michelle. Good morning, David. How are you? I'm very good. I've been well. Morning, Michelle. Awesome. Good morning, Michael. I'm doing good. I'm actually in California. Better than being on the west coast of Florida.
Discussion on Current Events
I flew through Florida and Bahamas. Yeah, my. I think my mother's house just got destroyed yesterday. I have to find out. Yeah, definitely. Prayers to everyone who lives around the area. She did evacuate, but I haven't had a chance to get back and find out. But I think there's a lot of people digging out. Man. Two storms back to back. But, you know, there's no such thing as, you know, climate change or anything that doesn't exist. Right. Just ask the governor DeSantis down there. I'm not sure, though, because there was one year where there was snowing in LA in February.
Weather Challenges
So I remember driving from Vegas, like, I have to drive to the Bay area, like, do the snow. So I think I drove from LA to highway one. And it was pretty awful, you know, luckily I didn't die, but I would have. I could have. I guess it was making. Getting worse. Yeah. Yeah. So hopefully. I don't know. I heard that there are things happening in Europe as well. That is a little bit different. Oh, God. Yeah. Okay. So it is so hot in Europe, and it has been now for probably a decade that, you know how there are all these wonderful water fountains that are, you know, for looks and everything, which probably originally water fountains are really get clean water, but, you know, they're all looks.
Cultural Shifts Due to Climate
You're not allowed in the water. Right. in Europe all summer long, everybody's in the water fountains. That's, that's, Yeah, that's, you know, I was in Norway and I'm looking around going, I mean, it was, you know, brutally hot and, you know that everybody's, in the water fountains. So, kind of that, whole cultural thing change really quickly. Yeah. To the sense that it's dangerous and kills people. Like, we should be a little bit more aware of climate changes. Anyways, it was just history. But hopefully my mother's okay and my sister and anyone else who happens to be down and down there right now.
Concerns for Family
I was talking to a friend who's, who lives in Tampa and it was definitely emergency situation. So, yeah, hopefully. Yeah, Tampa is right in the. Right down Sue's Venice, myself, so same. Absolutely same area and got hit. Yeah, I've been. There's great YouTube videos these days. It's amazing. All the cameras to. You can see absolutely everything. Yeah. I'm pretty excited about this space with you, David, as CTO pro saga, and I've known you for many years, so I basically told everyone that you were, like a genius and, you know, and influenced a lot of the Internet.
Introduction of Professionals
So I think we're gonna begin. We'll wait a little more, like a few more minutes to go into more of the deeper updates. But let's begin by with a little introduction. So I'm Michelle. I'm one of the advisor of Prosaga from a long time ago, and I'm helping with marketing and just really sharing about Persaga and what Persaga has done. So I'm pretty excited to chat with you here. Knowing you for so long and really supportive of your work. Do you wanted to provide us with a little bit of intro about you, how you have influenced and help in the Internet ecosystem that you know of what we know today and then, and give us a little bit history of how you became CTO?
David's Background
Oh, that's a lot. Anyways, yes. Thank you. And thank you for supporting us for so long. And my line, your line about geniuses. I've worked in my life with real geniuses, people who are like, guy who invented the electrophoresis, the western blood electrophoresis gel. That's the basis for DNA sequencing, that type of stuff. Nobel Prize winner. And I was actually taught by the Nobel Prize winner for the cats gain and that type of stuff. So I can tell you with absolute guarantee that I'm not a genius after having met them. Just want to be clear on that one. There's a their way out there, and I do know how to tie my shoes, which is also a telltale that I can't be a genius.
Career Skills and Accomplishments
So what do you ask me real quickly, my background? I'm an electrical engineer, computer engineer, software engineer. Started writing software about 14 years old. Built my first, you know, ham radio out of Heath kid when I was, I don't know, ten, 1112, something like that, and was playing around with radio shack experiment boards like when I was six or seven. So, you know, it's been my whole life. Turns out I'm not a particularly great electrical engineer anymore, but I have written a lot of software. I think you asked a little bit about the Internet. I've had some personal impact on it, you know, when it was a lot smaller and a few people could do a lot more, you know.
Involvement in Internet Development
So in the late 1980s, I was the network management special interest group chair. Not sure. Secretary over at NIST, Natchez just stands in technology, science and technology, which is the precursor pretty much to the way we manage the telecom systems. It was, this is when simple network management protocol SNMP was just coming out and their approach was what's called ASN one, abstract syntax notation and managing objects and systems viewing the network as objects, managing that way you know, from there you know the. You'll find my contributions in the Internet protocols in various places. The, one of my significant ones is the block acknowledgement protocol for Wi Fi 802.11.
Technological Contributions
You won't find my name on it. It was at the 800 and 211 meeting and went over to the people working on that particular piece of the spec and said please put this in because it was designing something with it told them how to do it. And then you'll find the port address translation of the NAT box. When I was over at a company, routerware, were pretty much the first ones putting that out. In fact I believe were the first ones putting that became a wind river networking stack. And you know, you use it, you all use today various versions of that, of course. And now it's in the Linux network ip tables and that type of thing.
Startup Challenges
I also did a couple of things I never saw the later day, which was a web farm load balancer, but unfortunately it didn't quite. My approach was with a company you might remember called Gateway. They bought a company called Advanced Logic Research. I was doing it with advanced logic research. Gateway bought them for about quarter billion dollars and then six months later decided they made a mistake and shut the whole thing down and threw me out with it. So you can't always have those. You also probably know what HDSL is in the first. When we used to run the Internet over copper, you know, between, when you had dial up and we had fiber.
DSL Market Contributions
I'm one of the pioneers in the DSL market and did probably the first ethernet over HDSO with the baby Bell companies. Can't remember the names of them anymore. Anyways, that's the background, you know, wrote a lot of stuff. We did all that. I think you want to know. Do you want me to talk about real quickly about how we founded this? Yeah, like I'm interested about how like who came out with the idea, what was your. So like I thought about this.
Founding of Prosaga
First of all, to be clear, Mike Holdman, who's also on the call with us, and myself are the co founders of Prosaga. We ran into each other, which is the evolution to where we got here we ran into each other at various trade shows when I was working with a embedded real time Java programming language company of all things. And I was focused on the Internet of things with the. In one of the early adopters and early members of the Internet engineering.
Internet of Things Initiatives
Sorry, the Internet. The Internet of things. The IIC Internet Institute. The hells the IIC stand for? IoT, the IoT consortium originally started by the OMG object management group. And I was. I was a member of it for. Well, technically, I'm probably still am a member of it personally and for my company then. And Mike Oldman ran into me, he was doing a messaging bus, an Internet messaging bus.
Technical Collaboration
So distributed system that's perfect for hooking together IoT devices. And we put some solutions together. And that brought us to a blockchain solution. And what happened is went to design a solution with the existing blockchains at the time. Throw it a couple names. Of course, bitcoin was around. Ethereum was around, stellar was around. A whole bunch of them were around, and iota was around. So give you a timeframe, 2018 or so.
Challenges with Blockchain Technology
And when were looking at it, we ran into two problems. One of them is the performance of blockchains for what we wanted to do was not stable enough. You know, you need to be guaranteed that the, when you put a transaction through that, it will go through in relative amount of time and at a relatively stable cost. And were able to see that wasn't going to happen. We're also able to see that as we wanted it to grow capability, it was going to slow down.
Designing for Scalability
And that brought us to. Okay, we need to come up with a scalable design that supports relatively stable cost and a relatively stable transaction throughput. And that's really where this comes from. Everything else kind of evolves from it. But that was the original. Aha. We can't build what we wanted to build because technology wasn't ready for it yet. That's why we're doing this, actually. That's the fundamental motivation.
Identifying Market Differences
Let's talk about differences and let's just go over some of the other chains on the market right now. Of course, the first one is Sui. Sui is really hot right now. How is prosaga different from Sui? So, two things. I have to be clear that I don't study everybody else's chain all the time. Okay. There was a time when I was reading every white paper on every consensus protocol, plus going through the history back to, you know, the raft protocol and those types of things.
Approach to Data Management
So if I say thing if I misrepresent another chain. My apologies. Essentially, though, at the very high level, our approach. So where I view us as the third approach to the data model for how to manage blockchain. Okay, the first model is bitcoin model with Utxo. Right. Very well understood. It's a non Turing complete script approach. It works really well for just moving ownership of, in this case, arbitrary coins around, but really could have been anything.
Innovations in Smart Contracts
Then along comes a smart contract, which is kind of like doing multiple versions of bitcoin on the same chain where you're attaching code to it and instead of. I'm sorry, and your accounts are represented as entries in the smart contract. And then the job of the blockchain is to maintain the states, the smart contracts. Now, I believe the third one is that. I believe. Sorry. I believe that SUi has also come up with the idea that there is a concept of essentially global objects which you would think of as global smart contracts, which other smart contracts can refer to.
Revolutionizing Contract Models
Right. I believe that they have that capability. We kind of turned that all on its head. I'm sorry. One other piece, remember, is that the big advantage of the smart contract over the bitcoin is that those are Turing complete programs where you use the cost of execution gas normally called to avoid the regular terminating, halting, well known Turing halting problem that Turing himself defined and wrote up. And that's the famous, his famous solution.
Core Technology and Scalability
The, what we did was turned the model on its head instead of looking at, and there's a reason for this. Everything we did, first of all, everything we did is to promote scalability, dynamic scalability, stability, capacity, transaction capacity, all those things. And what we did is we turned the model on its head and said instead of having a smart contract that has data inside it, which represents all of the accounts that are currently part of that smart contract, meaning usually what, how much coin they own, but actually it's an array of anything you want it to be.
Enabling Innovative Execution
Each account owns its own data, and that allows us to do something very interesting when it comes to parallel execution on multiple shards. But that's the core difference in the data model. And that enables a scheduling algorithm. It eliminates, it allows us to do scaling and sharding without doing cross shard communication, without doing cross shard transactions. Rather, we do cross communication. We don't do any cross shard transactions.
Efficiency in Transactions
And we do that without full time partitioning like you would in an l one l two solution also. So by flipping the reimagining the data model of what we're doing. We can enable that. And that's kind of the core. The ability to schedule. That's the core piece of our technology when it comes to scaling and performance. There's other pieces, but that's that piece.
Discussion on Metaverse X
Awesome. And how about metaverse x? Which one's metaverse x? They do. Mike, give me, give me memory so I can answer which one's which. Are you there, Mike? Can you mute for a second? I. It rebranded not that long ago, Dave. It used to be called Elrond, and. Then I think it was called egld. Or maybe that was just the ticker. But yeah, I just got shards.
Sharding Technology Insights
Oh, yeah. Okay, yeah, good. Let's talk sovereign shards. Okay, that's close enough. Yeah, so, yeah, with the. Right, right, so we're doing, we don't.
Sharding and Data Management
We are doing a sharding where the data is. Think of it as lazy, synchronized, temporarily sharded. Eventually, all nodes know, all shards, all block, all shard chains. Right. And the data at. If you wait long enough, all nodes will have all their state up to date. We don't do a full time partitioning. We don't do that because it enables all the scale we want to be able to do. We do something else though, and I want to talk specifically about the sovereign chain versus what we're doing because we can control with a scheduler and the scheduling is still a decentralized system where transactions execute and account state goes.
The Concept of Enclaves
It is possible to create instead of a sovereign chain, but we kind of came up with the word enclave. What that allows you to do is create, have a shard or set of shards that are managed by designated nodes that run designated accounts while still being able to send out summary or opaque information to the rest of the chain and take advantage of the hash power of the rest of the public chain. This is a, it's not an l one, l two. It's not a. It's not a sovereign chain. It's a different concept. And it's targeting specifically situations where, say, multiple businesses want to take advantage of the security of a full public chain, but also don't want to expose their own data. And so they, but they still want to exchange between each other.
Focus on Accounts and Smart Contracts
And so we have ways to pull that together because again, of focusing on the account as the container of data in the objects within the account, we have to talk about that a little bit versus the smart contract, which is this serialized entity that contains all the data of all the accounts that are related to it. So, David, thank you. Nir. And Solana are next, and they're a pretty big chain. So among all the l one's, how is prasaga different from Nir as well as Solana? Okay, Solana does the shredding and Nir is the. No, that's Casper. I've read near. Does the microtransactions cross shard, micro mic? Yeah. Okay. Yeah, sure. So let's. You know, you're asking questions that take a lot of architecture design to explain the differences, which I'm sure you were.
Solana's Approach to Sharding
Solana does, breaks up their transactions into what they call shreds and sends them out of, in parallel to multiple nodes. Then those nodes exchange with each other, and in the end, everybody executes all the transactions. Right. What it does is it takes the network traffic and creates a more less of a bursty network traffic. Because if you have makes it, let's say I have a megabyte block and I'm going to send it out to all the nodes. I got to send it out to all the nodes. That creates a burst both on the sender and across the network to each one. If I say break that up into 100 megabyte blocks and send some of those blocks to one node, some of those blocks to, I mean, some of one of those blocks to some of the nodes, a different of the blocks to different nodes, and then have them exchange at any given time, you're only sending 100 megabyte blocks instead of gigabyte blocks.
Optimizing Network Traffic
It's a. Yeah, kind of like a pipeline type architecture. It works very well for optimizing the network. It doesn't actually change the throughput of the transactions, okay. Because you still have to execute all the transactions. So for us, the key difference in sharding is that then when a node is on a shard validating a block, it has to execute the transactions. If a node is not on the shard, it receives the previously consented blocks and is then able to, instead of executing the transactions, just update its state. And if you can't do that, if you have to execute every transaction on every node, then in the end your throughput is limited to essentially the average of the speed of all the nodes executing that are running the blockchain or the average of the supermajority of all the nodes running the blockchain, depending on the consensus algorithm you're using.
Sharded Execution and Synchronization
Right. So our key point is we're not doing that. So what we're doing is sharded execution. Okay. We're with a late synchronization of data, we're doing sharded execution. And as a result, the more shards you have, and if you think of. Forget about each shard being a consensus. Think of each shard as its own cpu for the moment. If I can run multiple cpu's in parallel, the more of them I can run, the more capacity I can have for the chain. So that's basically what Solana is doing. What multiverse is doing is a shard chain. No? Absolutely. And they are doing sharded execution. However, their approach to handling a transaction that needs data that from more than one of their smart contracts, their smart contract model is what happens if I have a smart contract with its state one shard, and I have another smart contract with its state on another shard.
Constraints with Smart Contracts
And worse, let's say I have a third one on another shard, and I wrote a smart contract that's calling between them right now. We do very primitive ones. We know them as Defi. But what happens if you really want to get complex? Their approach is to do micro transactions between the nodes using a. Thanks Mike, by the way, is using a microtransaction approach, where they essentially stop the transaction, create a separate transaction, put it on a overlay beacon or main net chain, have that reach essentially consensus, have the targeted shard, then read that, do a local execution. Right. Then send the result back in yet another transaction, you might say, and there's a paper that shows a five part exchange. Be able to pull that off.
Concerns Regarding Congestion
There's nothing wrong with this. That works. Let me cure shard end systems work. The question is, how much congestion do you end up with as the types of things we want to do with the blockchain grow? To put it simply, if I'm only ever managing the ownership of a single token in a single smart contract, then I can put that smart contract on at least a shard. It will run as fast as that one shard. It can't run faster than that. We actually can. But normally, you, in a smart contract world, you're limited to the serialization of a single shard, but that then becomes much slower, as you have to essentially message between the shards. And now you're tied to the throughput of one overlay chain, main net beacon chain, wherever you want to call it.
Avoiding Overhead Issues
And so we spend a whole bunch of time trying to avoid all of those issues, which are what a near does, what a. I can't remember all the names of the shards. I mean, of all the chains, but most of the sharded chains are following that type of approach there is, to be clear, I believe there's one other sharding chain out there that's done something slightly different, where it looks like they're viewing the state as reading it from an IPFS type solution. And then the nodes I gather are pulling from that continually to assign themselves to individual shards. I didn't look at it too deeply, but I, it looks, it's, it's an interesting idea. Your overhead though, is now, your throughput is now limited to being able to pull the data that you need out of your, out of the IPFs system, which is yet another networking overhead.
Explaining Scalability
No engineering. You push so, you know, the light with the balloon, you know, you push in one area, bulges out the other. Anyways, hope that was somewhat clear. Awesome. For everyone else who doesn't understand what charting means is just basically breaking things down. For example, if you have one big basket of apples, sometimes they're too heavy to carry, so you might want to break it down and put it in different bags so that you can carry them and transact with them. Thanks. I didn't even know I started too far down the line today. Sorry about that.
Easier Transaction Management
No, it's okay. I think that's easy. Breaking down and just carrying it in different baskets and different bags are easier to transact and to transport. I used to work at Cisco systems, so we always use the truck as an example to carry data. And I know that apples are, you know, baskets and apples are really simplified, but easier to understand. Michelle, one of the analogies that we do, and Dave, if you want to go through it because you came up with it, is the grocery store and the lanes in the grocery store and how smart contracts are.
Visualizing Transaction Processing
Yeah, do we have a second? I'll do that real quickly. If you'd like, visualize a bitcoin or a single chain as a single checkout counter, and you're sitting behind everybody with your basket, regardless of what's in your basket, right. In a sharded chain, a basic sharded chain is you would have a checkout counter for each like, say, type of thing you can buy. Like, you know, you're the, you're only able to buy groceries at this checkout counter and so you're still lined up between everybody wants to buy groceries. That would be like smart contract with a token and you're doing smart contract execution.
The Role of Parallel Processing
And other people are doing totally unrelated transactions, but they're using the same token. Right. So in a sharded parallel in a grocery store checkout line, each token would be a separate checkout counter, which is still good. That's still better than having them all lined up between one checkout counter. What we want is something slightly different. If I have account a doing a transaction with account B and account C doing a transaction with account D, I don't care about what tokens they're using. I want those to happen in parallel because they're not touching each other.
Improving the Checkout Experience
So that means that I go out to this grocery counter, I mean, the checkout counter with my basket, and I can walk through any one of those trying to find the shortest line, which is what you really want to be able to do in a grocery store. So we're trying to be the grocery store checkout counter of the blockchain community. Awesome. And if anyone can animate that, ping me and let me know. Sure. It'll be kind of cool to have that analogy animated and also get show long lines.
Constant Stream through Checkout Counters
Yeah. What you would see, by the way, is that in, and of course in a single chain, you'd see just a long line of shopping carts. In a tokenized smart contract world, you'd see some of these checkout counters with really long lines and others empty because nobody happens to be using that token at that point in time. And with us, what you should see is a relatively average constant stream through every one of the checkout counters. Like going through a toll booth. Also, same idea. Awesome.
Addressing Scalability Concerns
So another thing is scalability. And I do not want to go to the grocery store, buy some tomatoes, and pay $50 in gas fees when the tomato is like $5. So how does Prasaga blockchain address scalability? And what specific challenges are you aiming to solve with it? Well, first of all, what we're trying to do on the financial side is use a currency monetary policy, which isn't. You don't see that in the blockchain world today.
Monetary Policy Insights
Real quickly, history. Bitcoin, of course, is the having model fixed number 18 months. Right? The second type of model is I'm going to just arbitrarily, every block, mint a certain amount of coin. And what's interesting is if you look at those two numbers in the limit, they're essentially the same. If I keep adding as I use a bucket of water, if I had a drop to a bucket of water, an empty bucket, I had two drops. That's twice as many. But when the buckets reasonably full, adding another drop of the same amount has virtually no effect on it, right?
Stable Currency Concepts
It does keep growing, but the percentage growth becomes in the limit. You know, if for those of us remember calculus. In the limit that would be essentially zero, but you never quite reach zero. Of course, the third way is to just start with one fixed amount of coin. So I'm just going to do a tge and hey, that's it. And whatever happens. None of those make a stable currency, as we've all found out. I. What we know about stable currencies are that the amount of currency has to respond to demand supply pricing.
Understanding Constant Supply
This is very well understood stuff. However, for the blockchain community, that's never actually been done. And what we can do when it comes to a blockchain, the only thing you really can control is the rate of increase and decrease of coin supply. Meaning you can. The rate you generate new coin at which you can call minting or mining or any word you want for it in the rate that it gets taken out, which is some form of burning or locking up or something else, anything similar to it.
Integration of Economic Expertise
What we did is I'm not. My financial modeling knowledge starts from ground zero, but we happen to have on board as our advisor, a recently Nobel Prize nominated economics expert, Doctor Lawrence White. He's written five, six books and he's the chair over at George Mason University. I think I forget where he's actually chair at. And he also, one of the books that I would recommend reading, which also has a chapter on bitcoin, is called better money.
Working with Economic Models
And he walks through the history of gold and goes on to the general history of money and then into where bitcoin is and points out the fundamental flaw. And bitcoin is the monetary policy which we've worked with him to actually implement on our chain. And what that means is that it's responsive to various signals to not just increase the money supply, but be able to increase, decrease, and also pull, temporarily pull money out so you don't necessarily have to burn it and do that under a mathematical model.
Stability in Monetary Policy
Not arbitrary like, hey, a group just got together said, hey, let's go burn some coin. So you end up with both a stable and intentionally stabilized system and something that's not subject to the whims of an, you know, an arbitrary group of people. That's a. I would strongly recommend to start for a background, go read that book first, because a lot of what we'll do, what we're doing will make much more sense.
Architecture and Infrastructure Design
Awesome. David, I want you to talk a little bit about the architecture of prosaga in just a high level. And also regarding the infrastructure. How do you design it so that it could scale and it's secure, because I think if you're in the blockchain space for a long time, you'll notice that a lot of the chains are really hype when it comes to it. And then, however, it doesn't scale and it's hard to perform. Some of the chains are congested.
Challenges of Current Blockchain Solutions
I myself personally have faced a lot of challenges interacting with a lot of the chains right now. So can you just really talk a little bit about the, architecture in a more simpler way? and how that's different? Yeah. Problem is, I am talking in a simpler way. okay, so, let me. Let me start at the basic point of view of the consensus.
Understanding Consensus Architecture
When you say architecture, I don't think you want to know about the program process architecture. I can tell you what it looks like, but I don't know if that helps you. first off, the general blockchain architecture is that there's individual shards, which are running a consensus protocol, which is a hybrid solution based on both signatures and proof of work. And then it's using a braiding between the shards to recruit the hash power of the nodes of the other shards.
Transaction Completion on Individual Shards
And then there's a overlay chain that is included in the individual shard blocks as well, which is used strictly for managing what we call account ownership and account migration between shards, so that there's never a transaction from one shard to another. All transactions are, by definition, running to completion on a single shard. Non blocking. That means that you don't go part way through transactions. Oops, I need something from another chain.
Ensuring Security in Architecture
I'm going to go send a transaction over to it. What did you ask me for? Architecture. One quick dig down about the proof of work. Our personal viewpoint is that depending strictly on signatures, which is proof of stake in the end is signing, where they. You can use an aggregate signature approach. We will be using an aggregate signature approach for each one. For that part of it is not sufficient in the long run for security.
Balancing Power and Security
And the reason why I say that is the speed of the chain is artificially slower than if I had all the keys and I wanted to make a fake chain on a single high power computer. See, in a bitcoin world, if I have a single computer, there's no way I'm ever going to keep up with the hash power. Literally. That's why you need 50% or more totally take over. There's other games you can play around 25% in a signature based world.
Potential Risks in Signature Based Systems
If I can get into all the keys, or at least the super majority of the keys, at any point in the history of the chain, I can fork it and create a fake chain, and there's virtually no way to tell the difference between them. And that includes also maintaining all the coins, all those types of things, because I will have access to all of the keys for it. You know, is it going to happen? No, but it's theoretically possible.
Implementing Proof of Work
And I didn't really want to have a chain where we're trying to say, come use this for global security, for currency, and know that was sitting there. So what we did is we threw a proof of work on top of it. So it goes through and does the short term validation of the blocks on each shard individually. And then each shard essentially has a mining pool around it that does a proof of work on it.
Integration of Hash Power
There's another, there's a couple more details to it to make that work. But now you've got one level of hash power, but I still want the rest of the hash power of the rest of the nodes on the system. So basically you send the, you gossip the blocks to everybody else because they need the state anyway. And then I. Any given shard can simply incorporate the hash of a block that they received from another shard in a current block.
Braiding Hashing for Security
And you essentially get braided hashing. And you can do that of course, in parallel, across multiple and make them intermix in any way that they feel like you don't really need to be very particular about. You just want to make sure that you get lots of overlay, because that means if I wanted to now hack that chain, I have to, even if I have all of the keys in the world, I still have to overpower all that hash power.
Combining Security Concepts
And so I'm essentially using kind of the tangle concept coupled with the BFT proof of state concept coupled with the general proof of work as a level of difficulty. So I kind of threw those all together, works pretty well. There's nothing, that's nothing particularly difficult about it. It does mean that your proof of work. So you do need a subcommittee of nodes on any given shard to do the proof of work.
Ensuring Performance with Proof of Stake
Sorry, using our word to do the proof of stake so that the thing will actually work with performance, because that's still a full mesh shared architecture. So you can't that, you know, I know that some nodes have some system, have found 21 nodes to their maximum. I think we go a little higher than that.
N Squared Problem and Protocol Overview
But you still have the n squared problem. And then the overlay, which manages strictly account ownership, not transactions, runs a straight proof of stake hot stuff style protocol. If somebody wants to look up that paper. I made a couple of minor tweaks to it, but basically that approach and that runs, the intent is for that to run faster than the individual shard blocks. So essentially the account negotiation of where the account should be happens faster than the blocks. The blocks then manage the actual account ownership transfers. That's the overlay of the protocol. So think of it as way thing is a bunch of vertical shards or vertical checkout lanes with this background scheduler that kind of points where your transactions go to which checkout counter like something like that. And then that uses a scheduling algorithm. And that's our secret sauce, that's our, that's the core piece that solves the, or tries to solve the congestion problem that you'll see in every sharded chain out there. I can tell you about that in a minute if you want it.
Technology Development and Key Milestones
Awesome, David. I know it's really easy to fork a chain. And then however, I think this project has created the technology from scratch, which is awesome. I want you to tell us a little bit about some of the key technical milestones that the company has achieved so far. And I know that there's some alpha coming out as well, so would you address some of that? So first of all, everything's built from scratch. There's no, I was not able to use any. Well of course I've used libraries and things like that. We're currently using the lib p two p library for gossip, using the gossip sub module on top of lib p two p. You can find that in GitHub. We're using the go programming language in Python. There's that respect. There's a block explorer which is written of course in JavaScript as well. We're using GRPC, which is for Google protobufs, a kind of a sister system to that manages the client server type relationship or client to client relationship milestones.
System Implementation Phases
So the very first thing we did a few phases. The very first phase we actually did was wanted to implement once our first class object model, metaclass model, which is the core piece that enables the account model, that enables the input to a scaling scheduling algorithm. And we implemented that actually initially as just a trial proof of concept on Hyperledger because it turned out it was relatively easy to work with that. I couldn't scale it, but I could do that one piece. We actually implemented that not with Python, but with a very manual go approach to it, put that together, verified, use that to then come out and start asking basically the industries, various places. Hey, does this look interesting? Should we go do this? We got a very positive response. The next milestone was reimplementing that one piece in Python so that you now can execute a transaction and start supporting an object state database that is account based.
Object-Based System and Scalability
So an account is an object. So when I mean by an object I mean that the classes, the code for the classes is actually in the database itself. You're not compiling code into a program or into a smart contract, you're just instantiating objects from classes that exist in the database and the ability to add new classes to the database. So it's an ever increasing scalable system supporting both regular inheritance and multiple inheritance. So you can do what's called mixin classes as a set of foundation class to make this run. So we verified that piece, prove that all works, you know, can it be more optimized? Sure, but absolutely is running. System went off, took the bypass to go build the consensus state machines and go using, you know, protobufs, GRPC, the Lib P, two p, all those types of things.
Database and Financial Model Design
And did the object state database using something that supports our sharding model with multiple trees. And it's a relatively complex system to make happen, but the end result from the views user point is you send transactions in, you get, you know, your object statement for you. Then the next core piece, two more pieces, designing out the financial model, working with doctor White for the last year or so. More than that, I think on what we want to be able to do there. And there's a lot of details of how that makes it possible to have a relatively stable currency. How to, when I say relatively stable, programmatically stable, not tethered, not what you call a stable coin or whatever the other terms are. It's still its own self managed currency.
Scheduling Algorithm and Environment Setup
And then the last piece is the scheduling algorithm, which is solving which is coming up with a partial solution for what's known as an np complete scheduling problem, which is the equivalent, whether you're talking about sharding multigor cpu's or traveling, is actually known as the generalized traveling salesperson problem. And our solution to that is how we do account assignment, which makes it possible to make the scaling happen. Everything else you need is the infrastructure to be able to make that one thing happen in the milestone is we've, we did a, an initial proof of concept standalone, hey, let's just run this algorithm. Does it actually work? Then built a second round, you know, that's built up for this and then integrate into the system and we're in the middle of putting it all together so that we have a development test environment for people to run.
Release Plans and Future Integration with AI
Currently we're running both a bunch of simulated nodes on single machine, and then we're running multiple nodes using docker containers in the docker overlay network so that you can scale out on multiple machines for test environment. Our first release for this, by the way, will be that as a developer, you essentially run your own chain, not connect to anybody else, start it up, you can run, change access to it, try it out, run its own databases, do all those types of things. And then obviously eventually we'll put up a full test chain that everybody can connect to as soon as we take it public. That's good. Awesome. I won my own chain, David, but I wanted to ask you, because I'm an AI person, so I want to ask you, how do you envision the role of artificial intelligence and blockchain evolving together? And does Poseida have plans to integrate AI?
AI Integration and Predictions
First of all, you know, predictions are difficult, especially about the future. What I can tell you is what we can do, what we're doing not, I don't know what the rest of the world's doing with it. We're doing two things. There's a saga GPT under development, and what that'll help you do is both developers and users of the blockchain of saga chain take advantage of it, because we're a think of as a global class based system in every class, unique, and it's put up on the blockchain. Wouldn't it be nice if there's some way to find out what classes are available, what accounts available, what designs are available, all those types of things. If I have a regular class browser type approach for a computer science person, sure, they can scroll through that and just imagine how much code is going to be and you're going to start seeing something that's virtually impossible to deal with.
AI Enhancements in Transactions and Learning
If I have an AI program that can start, you know, providing you with answers to queries to it, that's a hell of a lot more accessible. Second part is we just in general writing transactions. So in the simple case of a wallet type thing, your transactions are written for you behind the scenes will work pretty much the same way when you get into any sort of complex type behavior, because we can support essentially an unlimited amount of complexity of a message passing system, you start running into, hey, how do I write these transactions? And we don't want to end up with, you need a computer science degree to be able to send transactions. So hey, we've got the computer science person in your AI program right here for you. So that's a second way to look for it. And just as a general learning tool, be very helpful. Obviously there are tools for internals for developers who want to work on our code base itself.
System Architecture and AI Scheduling Potential
I separate the classes and transactions that run on the chain versus the chain itself. Those are two different things. There'll be some application of that there. I'm not sure how useful that is for the general public, but it might be useful. Then there's a place that we have a possibility for that's unique, which is in the internals of the system. Again, back to how you schedule where the transactions will execute to maximize the capacity of the chain. There are opportunities for using an AI algorithm as a scheduler to anticipate the best distribution of those accounts across the available shards, and also to anticipate dynamically increasing or decreasing the number of shards. That's not an area that we're currently working on, but it's definitely, we're using a different approach with a more traditional nonlinear search technique.
Future Goals and Potential Use Cases
But it's definitely a future direction for that piece. And it'll be a plugin. You'll be able to drop it in. It won't impact anything that's already out there. I have one more question, and then I'm going to open it up to the community for questions here. I think I invited a number of technical people either in the blockchain space for a long time, ogs or people who code. So I wanted to open up to questions, but my last question is, what is your long term vision for prosaga technology? What do you see to platform in the next five to ten years? Well, once it's up and running, it's kind of like a. You're asking me like, what am I going to do with the cpu? You can make it virtually do anything. It's not limited to just dealing with tokens.
Community Engagement and Collaborative Opportunities
So we've been in, we're a member of the digital DTC, the DCC community, the IEC community, the various other communities. A lot of. Yeah, sorry. Yeah. Digital twin consortium and industrial space. We have some acronyms from. Yeah, the IATF, ieee, DTC. OMG, the world. Yeah, I know, I'm sorry. Just like blanking on the words. What you see in those types of places quite often is in one of the places we've looked very heavily at is general supply chain applications. It. You have a many relationship. Just real quick, I'm going to. Real fast. Spent a little time I was out of the top ten automotive manufacturers, about five or six of them used to be my customers. Chrysler, Toyota, Ford, GM, etcetera.
Challenges in Supply Chain Management
And what you have, were two or three tiers back in the supply chain to all of them. So I got a real good look at the many to many relationship between the suppliers and the customers, and it's a mess. And you have all sorts of issues of how do you track what's going where and how do you know where it's going? And you have a general problem also in the military space, which I've also spent a lot of time in, where you get gray market parts from sneaking into chains from all over the world, even into our ships, like, into our military equipment with gray market ships that were bullshit. Literally coming out of, like, you know, China or wherever else, just taking crap, putting a new cap on it, making it look like, hey, this is, you know, an intel cpu.
Blockchain as a Solution in Supply Chain Issues
And of course it's not. You put it on the board, doesn't run, and all sorts of stuff like that. You have an immutable chain that's shared among these companies in this enclave concept, where they essentially have a better way to prove, provided you can show how it got on the chain. Garbage. You have to have a garbage and garbage out problem, but you reduce the opportunity for the interchange between hands, of mixing in parts, doing all sorts of funky things. And I think that'll turn out to be one of the places it'll be very interesting. I am sure there's lots of other places, but that's one of the ones we're looking at closely at right now.
Q&A and Community Interaction
Awesome. Does anyone have any questions that you wanted to ask, please raise your hand. I'm opening up to questions. I'm sure the community has a lot of them, especially technical questions, so please do. All right, I think Sheldon has a question, and. Go ahead, Sheldon. Hey, guys. Thanks for having me up. Long time fan. Been keeping an eye on y'all's work for quite a while. There's. I can't speak too much, but I'll try to keep my question short. There's a series of charts and research I can show you from the past, like, say, four to eight years that show that enterprises are largely unhappy with blockchain solutions they've adopted. It's kind of the Long island blockchain disease.
Challenges in Development
And at the moment, in the current state of development is if you do the wrong thing, you'll kill it. So we have to work on a better, I don't want the right words, make it more resilient so that if you actually do something wrong, it won't bring down your class and your object and never be able to use it again. But I don't know. Oops. Deprecated sounds pretty cool. There's, there's something permanent about blockchains that is, you know, part of the business's responsibility to get right the first time. I see that in public chains all the time. I appreciate your point, because things need to be able to be changed, but. Well, there is that. It's also, we will have a versioning capability which allows you to reversion a class, but you then get into mutation issues of, do you want to allow to mutate an existing object that's based on that class? And the answer to that is, it depends.
Mechanism Based Development
So we're mechanism based. We'll build you all the stuff in the world out there. I don't always recommend using all of it, though. I'd love to see a framework for automatic deprecation with auto genuid or some generic hash or something like that. It sounds like an interesting way of versioning. Like, hey, oops, you broke production. You know, this thing now needs to be fixed. Then you can reference the broken object and, you know, bring back a class that's actually useful. That's pretty cool. Mutation, all that. And that's. Those are tough problems. Let me say that those are problems can all be solved. They're tough problems to solve where the. You don't take a huge hit on performance.
Discussion on Security and Education
Awesome. I think Nathan also has a question. Nathan, you wanted to ask your question? Yeah. Hey, how you doing? Great. We met. We were hanging out at this brunch when we did the. The venue walkthrough for the breathe conference a long time ago. Cool. I don't know what you got. Yeah, I have no idea what happened with that. I was just in town when we did the walkthrough, so, anyway, I. We fell out of touch. I'm the guy that organizes the blockchain stage over at DefcON, and I know that I was talking to you, and I think. Boss, you remember this conversation? Yeah, it was myself, actually. Nathan. Yeah. Mike. Hey, cool. Okay. Yeah, no, I saw you guys were hosting a space. I wanted to get back on here and get linked back up with you guys and just talk a little bit about security, research and education so it doesn't have to happen here.
Further Talks on the Future of Technology
Now, this probably isn't the best venue for it, but just want to get back in touch with you guys. Hey, I remember I've been to def Con, Blackhat, and Defcon. Not in the last few years, but 20,000 plus people. That's it. Yeah. It's wonderful stuff. Yeah. You know, there's different questions about security. Right. There's. The security of. Is the chain immutable? There's also security of is your code on the chain going to, you know, be a rug, pull that type of thing. One of our positions, which is it's a soft position. It's qualitative, not quantitative, is that if you're running a inheritance model of code, then most of the code, most of the use on the blockchain is actually using the same code.
Observations on Code and Blockchain
So it's kind of like the same idea as the open source. Everybody's looking at the same code. If there's a bug in there, everybody's going to see it. Right, which is it's going to happen a lot faster than if somebody threw out a new smart contract with its own code. And you now have to walk through that code and hope that there's not something you missed in it. So what you get out of it is think of, and we will probably do this, by the way, create essentially a set of certified classes that say, hey, these things have been heavily verified. You can use them. And to use them, all you do is say, give me an instance of that class. Gives you an object pressed, or you're using it, so you have a much, hopefully a smaller attack surface for that type of environment.
The Future of Energy and Technology
That said, of course this is software. As everybody find out, there is no such thing. If you. The mathematical proof of incomplete. Of the mathematical proof, the incompleteness theorem guarantees that there is some way to hack everything in the world. You just try to make it really hard. Awesome. If you guys cannot hear us, please leave and then come back. I think. I think there are some rugs here. I just want to make sure that people get to ask questions and listen because I'm enjoying the conversation. I know that is super technical, but again, it's fun and it's about really building something. Oh, sorry. Let me. Nathan, please make sure to connect with Mike, by the way. Thank you.
A Brief on Internet Connectivity Innovations
Nathaniel. If you. If you cannot hear anything, please leave and. Oh, is that why a bunch of people just dropped and came back? They keep coming and going. They didn't realize. Yeah, because Twitter is sometimes rugged, so people can't listen. So I think it's just the better solution is just to leave and then come back, because I have people who pay me. Yeah, well, I'm on an old fashioned cable which has 200 megabits down and, like, you know, 15 megabits up. So that's. Welcome to a pre fiber charter spectrum. Spectrum. If you're listening, you suck. Sorry. I'm actually looking forward to. I'm looking forward to. That's Leo's low Earth orbits.
Melting Pot of Internet Solutions
Yeah, except if they, the only problem with them is if they ever collide, you're going to have debris like everywhere. But they're powering a lot of the airlines, too. And like, you know, during hurricane season, like what happened to Tampa right now, I think Starlink is going to be perfect for emergency situations like that. There's another system going up by the other satellite vendors. He wasn't advised at that type of stuff. Low Earth orbits, they're really cool because you're like 100 millisecond up and down. I was using Hughesnet and it's a three quarter of a second up and down. And so be interesting to try to run nodes across the satellite system like that.
Discussion on Renewable Energy
Hey, David, on that topic, what do you think about computation and, you know, an energy and power and, you know, I think we're at mining disrupt the other year and I think the world's moving to clean energy and I know that fusion is still a moonshot and we're still not there yet. But what is your opinions about? Do you think that there will be a bigger convergence? Because I think I see AI growing bigger. I think we're moving, yeah, we're moving from, you know, gen machine learning to machine reasoning now. And, you know, and single AI used to bot communication and compute and, you know, and then at the same time, AR and VR are, you know, not supporting a bigger community right now.
Insights on Battery Technology
Right. So how do you see that playing a role in. Well, actually I was going to do something a little more mundane. There's an interesting little piece when it comes to power consumption for the blockchain. Right. The problem, of course, with the bitcoin world is no matter how many machines you put on, it doesn't run any faster, but it does consume more power. If you shard a chain, the more machines you put on, the more capacity you get. So essentially you get a improvement in the power to capacity. Actually think of that way that ratio stays constant, which is better than having the, you know, capacity stay constant, the power increase.
Nuclear Power Perspectives
So that is an advantage of the sharding model over even probably like a Solana model. I think you're asking slightly different questions of which I'm, you know, my knowledge base in some of those island more than others. The, my personal opinion, this three relative things, interesting way back when a battery company called a 123 came into existence, that was basically the DoE funded them. And that was actually over a cup of coffee. I was trying to set a conference with the DoE and sat down and said, look, we need better batteries. And a 123 came out of that. That was supposed to be spin off out of MIT for some nanotechnology.
Future Concerns and Hopes for Technology
I don't think it quite worked out right. And they eventually went under. We need, obviously better batteries, and we need batteries that are based on elements that aren't going to run out anytime soon. What we have right now, even though the great and the cars are great, is probably not that sustainable. And I don't know what the answer to that is, by the way. I'm hoping that somebody in a lab comes up with, hey, I figured out how to make it run with, you know, cmos type technology. You know, there is a technology out there claiming that they have a solar panel with, you know, near order of magnitude better efficiency.
Innovation in Energy Technology
So their efficiencies are now what I believe I heard Max 22%. Maybe I'm high, maybe it's 18. And somebody's saying they can get way, way up there again. You get those things, and then those claims either happen or not. We need better solar panels. We need better, we need cheaper ways to manufacture them. We need those things. But I'm a big nuclear power guy, I believe strong. First of all, I think we should be doing fission reaction right now. There's better ways to do it than before, especially because, hey, we made all these really great control computer systems with wonderful message bus architecture and an Internet that works in all these embedded systems.
The State of Fusion Technology
And we ought to be able to control a nuclear power plant a shitload better than a bunch of guys staring at a bunch of dials all day long, which is what they were doing. And we have, oh, wait for it, AI, that'll actually pick up problems early. And then, you know, the fusion, they keep promising fusion. I don't know. I mean, when I was in college, the big emag test, both in engineering and physics, was around a toroid, a doughnut shaped thing. I was wondering, why are they always giving this on every test? It's because they were trying to make a toroid for a fusion reactor.
The Solaris of Energy Production
And so that was the approach they've been trying to do with a containment around the, using magnetic containment so that the plasma doesn't touch the walls. You know, there was recently the proof of ignition that they can do for a microsecond or something. I don't know. And there was a company that claims that they were doing some very strange looking design where that looks kind of like two bottles pointed at each other and colliding in the middle. And it was probably bullshit. Because if they actually did that and they stood in the room, they were going to die because they had no containment around it at all. I was watching a YouTube video.
Avoiding Hasty Judgments
They're going, yeah, better not turn that thing on. So I don't know. But if we can get to Fusion, hey, that's home free. But that's where. I don't believe that long term solar and wind is enough by any stretch imagination. But I think it's a good stopgap. Hey, in the way, DSL over copper was a good stopgap to get to fiber, right? Yeah, I've met some other fusion companies. I'll share that with the community. But I'm hopeful. I mean, it's still moonshot, but I'm confident that a lot of PhDs are working on that.
Access to Technology and Internet
So hoping that would. I know that a lot of people still don't have access to the Internet, but I'm hoping that more people will have access to the chains because it is web three and is co ownership and that it is a little bit more revolutionary. Outside of all the hype and the graphs and the coins and all the prices and stuff like that, there is still technology and it's still a ledger and that is still a change from the past. So I'm really excited about that. David, I wanted to thank you for being here and, you know, for the chat.
Concluding Thoughts
Thank you for your time and thank you for all the prosaga team and everyone who joined and stayed. Hopefully you enjoyed this really nerdy chat with us and see you again next time. Hey, Michelle, I'd like to bring one thing up to Sheldon, if you're still here. So one of the things that we're in the process of doing over at the digital twin consortium, Sheldon, and why we're working with them, is to start to develop process flows within different industries. That is what is the general process flow of a blockchain.
Standardization in Blockchain Processes
Is there, is there like an overall generic and in each. But everybody customizes it down below? We do believe so. Like, for instance, there's an example of our auto, the supply chain class tree. Just an example, but it's got like class fungible asset class, non fungible asset class, part class assembly. What we're doing is defining these process flows and then finding out which objects in there that is a part, have generalized definitions globally. So we can now standardize, make it an international open standard under, you know, IETF, ISO I, Tripoli and other bodies.
Creating Smart Contracts
So this is how you represent a smart, or this is how you represent an asset or an object on a blockchain as these programmable smart assets. You know, one of the questions we always have is why we know what a fungible asset is. It means the same thing everywhere. Why do we keep writing what a fungible asset is? That is a smart contract where you clone that smart contract, basically be duplicating it and put it right back on the thing instead of make it instantiating from a single global instance of code so that anywhere in the world you are, you see a part, you know exactly what that is, and it's going to be defined with every other part.
Inviting Collaboration and Participation
You just customize what your part happens to be. We are now starting to put together two things. First, with the DTC in order for them to do this, and it will create these design docs and then they will go to the working groups and they'll run testbeds where the enterprises that are part of DTC well then apply or implement the code to what the, you know, to see if it does solve and work within our enterprise. The second thing we're doing with that is from those docs. We're doing our own separate proposals for submission to standards bodies of those assets and such.
Building Relationships in the Industry
So if you are interested or anybody else on the phone in participating, we would love to have. We'd love to be in contact because we are. We believe that this is a. One of the things you were asking about, Sheldon. This is how we believe that we can now, because we have this single class tree, because we can make single instances of code to represent always used objects, we have a good ability or capability to expand this beyond. And if any of that made sense.
Closing the Discussion
Awesome. I'm in contact with Sheldon. So yeah, I will basically connect you guys on telegram or so. Awesome. I think we are going to end this chat. Thank you for everyone for coming. I love technical discussion. I think they're fun. And I think there are the builders who are making this happen. So appreciate all the work that you've done, David, and thank you for making this happen. Thank everyone for coming and all your awesome questions. Alrighty. See ya next time. Have a good afternoon.
