Epicenter - Learn about Crypto, Blockchain, Ethereum, Bitcoin and Distributed Technologies - Igor Barinov: POA Network – Enabling Scaling Through Trust in Public Notaries

Episode Date: August 6, 2019

We're joined by Igor Barinov, the tech lead of the POA Network. The POA Network achieves a reduction in transaction costs by many orders of magnitude by having a set of trusted validators. All validators must be US public notaries, so their identities are known and legal recourse against them can be taken in the offchain world. We also talk about the xDai network, which enables Dai transactions on a POA chain: Dai are transferred into a smart contract on the mainnet and then become available to be transferred at much lower cost on the POA Network. Similarly, they can be transferred out of the xDai network and become available again on the mainchain. We talk about use cases, governance, and limitations. Topics covered in this episode: Igor's background and how he got into blockchain The POA network setup and how to become a validator What informs design decisions and how to find suitable notaries The consensus mechanism on the POA Network The role of the POA token The xDai network and its purpose The role of the DPOS token Use cases of POA and xDai Networks Episode links: POA Network website POA products overview xDAI Chain POSDAO White paper Ocean Protocol releases own POA network Beginners guide: What's the POA network Sponsors: StarkWare Sessions: September 16th in Tel Aviv – 20% off with the code EPICENTER - https://epicenter.rocks/starkware Trail of Bits: Trust the team at the forefront of blockchain security research - https://trailofbits.com Vaultoro: Trade gold to Bitcoin instantly and securely starting at just 1mg - http://vaultoro.com This episode is hosted by Meher Roy & Friederike Ernst. Show notes and listening options: epicenter.tv/299

This episode of Epicenter is brought to by Starkware sessions happening in Tel Aviv on September 16. Broaden your knowledge of zero knowledge proofs with highly technical sessions from some of the industry's leading experts. Register at epicenter.rocks slash Starkware and receive 20% off the regular ticket price with the code epicenter. And by Trail of Bits, don't leave your project's security audit to just any firm. Trust a team with decades of experience at the forefront of blockchain security research. Go to Trail ofBits.com to learn more. And by Valturo, the gold hedging platform for the crypto community. Trade gold to Bitcoin instantly and securely, starting at just one milligram.

Go to Valtoro.gov. slash epicenter to get early access to their V2 platform and to start trading. Hi, welcome to Epicenter. Episenter is a show that interviews crypto entrepreneurs, academics, and business people in the cryptocurrency space. I am Meher Roy and with me is Friederica Ernst So Frederica and I are going to chat with

Igor Baranov who is the tech lead for the POA network Now the POA network I feel is one of those projects that challenges a lot of conventional wisdom on how crypto networks and proof of stake networks in particular should be paid

what's your opinion of it Federica Oh it's super interesting So we are trained to think of centralization and dependence on individual actors as something bad, but it doesn't necessarily have to be. So it was super interesting to hear ego's points on this. Yeah. So it's essentially this consensus algorithm where all of the validators are US notaries.

Their identity is known. And it has a very unique economic system, very unique governance system. I mean, if you look at it from a game theory, perspective. Probably you are not going to find it very sound. You can have various critiques of it. Certainly I was getting a lot of critiques in my mind. But the reason we felt this made an interesting episode is, at least for me, the lens of game theory is an interesting lens. But sometimes human behavior is actually different from what your mathematical model of the game is going to predict. And then you need experimental

data. You need experimental economies putting people in situations to see how they actually behave. And I feel from that perspective, it's amazing that a project is doing an experiment like that. What if you made like notaries the validators and you wrote the proof of stake game that way? I feel like they can be very valuable data that comes out of an experiment on how humans actually behave when they are given a position of responsibility in that kind of of me. So without much further ado, we give you the interview with Igor. Hi, Igor. Thanks for being here today. So to start things off, can you tell us what your background is and how you got into blockchain? Hi, my name is Igor. I'm tech lead of POA network. My background, I have education and computer

science. I worked all my life in IT. Most of the time I worked in enterprise, in mobile operators. I got interested with blockchain with Bitcoin and cannot stop since then. It's quite easy. Right. And from Bitcoin to Ethereum and from Ethereum, I decided to stay with Ethereum for a while because it's, I think, the most exciting development that we have for the last four years. What was it that caught your interest in Bitcoin? Before Bitcoin, We participated as a team in distributed protocols. If you remember, there's a Boeing where, you know, SET at home and some other distributed data processing algorithms.

And we participated in this. And when I first time heard about, you know, peer-to-peer electronic cash system, I got excited. And actually what we're doing now with our X-Dai development is bringing this electronic peer-to-peer electronic peer-to-peer cash system for like real world application with like required speed and transaction price and program mobility, which is very important for electronic money. So the idea is even then you don't understand that when you understand that money can be

free and yeah, that's very exciting. So that's how I started. It was quite hard to get, let's say, a job with before, before Ethereum. time in blockchain space. So I worked with multiple blockchain companies and participated in all hackathons. Even before, you know, blockchain hackathon were like a term, right, like just you just come to a hackathon and you explain what blockchain is to judges and after you build something and show it to them. And after I worked as a consultant for a Singaporean company called Akronis, built some, let's say, smart contracts and tools around them,

which after and after this consulting, I started POA and so working with POA since then. You started working with POA from the get-go, so you're one of the founders, but you don't, you don't call yourself that. You call yourself the tech lead. Why is that? We don't ourselves founders and co-founders and we don't have C titles. It's a, it's a form of organization. We decided that all of us are like contractors, contractors to the network and to the protocol. We had some initial roadmap and white paper and ideas about like how to make Ethereum protocol scalable this way, right, and how to build tools which are like not available for for this type of protocols. And we decided that we contribute our sources towards this idea.

When you work in open source field, it's easier to be, like, developer and to be a builder, than to be like CEO and so forth, right? So we're not selling our products. We don't have this, you know, commercial department, business development and so forth. Yeah, so that's why we don't have like C titles. Technically, there are some C titles because it's required by regulation for some, for some operations, but we don't use them, not internally, not externally. And we have quite a flat structure within the team.

So do you have like a corporation that's actually building the things, or is it just the DAW? Yeah, it's quite a common legal structure when there is a foundation which holds some crypto assets and the operational company, which is building stuff and not even one operational company, multiple operation companies and external partners. And technically, it works in legal field. That's right. There are some companies. We're not operating only as DAO because it's quite hard to pay AWS bills, right? As a DAO, you have to have a company to pay from the same with, you know, with people and other expenses. I get why you're, why you don't have a CEO, a CEO, or other C-T titles towards the outside.

But how does the governance inside POA actually work? Because someone actually has to decide which direction to march in, right? Yeah, we have two fund managers who decide how to spend funds. And it depends on the situation. Let's say four to five team leads. And we have consensus team. We have Black Scout, which is our open source, blockchain explorer team. I have a token bridge team and DAPS and all other tooling.

So we have four team leads and the team leads are working with like external and internal resources. But we all contractors for this project, right? So we don't have like, okay, these people are an employee and these are contractors and these like external contractors. We all decided that we're all contractors, but with like some different decisions. making abilities in different fields. This is very interesting, but before we go deep down into how the governance of the network works, could you just tell us what does the POA network do? And you have another network, which is the X-I network, what does that network do?

And what's the vision for each of that? Yeah, it's changing from time when we originally thought about this idea. If you remember the April Fool's Day post by Ethereum Foundation, written by Vitalik, when he said that Ethereum Mainnet is migrating to POA consensus, he said this back in days. And it was a joke, right? And the first time when we discussed a public permission network with real value based on Ethereum protocol, people thought like it's stupid to make, right? First of all, it's centralized, which is a strong argument.

It's hard to argue about it. Second is that it's not secure, because proof of work secure and other, especially with this type of consensus with exclusive group of validators are not secure at all. So that's what people think, right? It's not that I think. So it started as an experiment. Like, why do we need this experiment to make the protocol, which is like Ethereum protocol,

more accessible to people and let people experiment with this protocol for their own needs. PEOA network basically had different consensus algorithm, different set of validators, different reward structure, but the same basic client,

without any hardcore changes for developers. So it's quite easy to jump from network to network. And the X-Di is a, if we think about this, it's very new experiment, because this network is actually the first stable chain, the chain with a stable native token without any initial emission, without any supply of this stable token.

So all tokens on all native tokens on this network are breached from main net to X-Dai. So it's a new concept. The theoretical concept of hard spoons are known for, for a couple of years, but X-Dai is the first hard spoon with a stable token, which is not competing with Ethereum, right, because the stable token home is on Ethereum. So that's kind of experiment that we made within this X-Dai network, and people love to use it.

I think that it was not possible within the Ethereum main network and setting. So we had to make it somewhere, outside. So that's the main reason to exist. And besides building networks, it's important to build tooling. If we think about Ethereum ecosystem, some tools are not available for site chains and forks. Before it was like in Fura, which was not available. It's still supporting all the main net and test nets and ether scan, which is not supporting competing networks and metamask. right, like three very important tools for developers,

not supporting competing projects, let's say, right? So that's why we decided to spend our resources to build at least one tool, and our blockchain explorer, we hosted for 12 networks, and there are around 20 networks using this blockchain explorer. So that's our contribution to the ecosystem, and we like Ethereum ecosystem and we basically staying here and building tools, which can be used on mainnet, can be used on side chains.

That's idea of POE as a protocol, right? Allow people to experiment, build tools and make this protocol, you know, scalable. Scalability nowadays is not that important like it was two years ago, right? Because two years ago, like everyone thought, okay, what are we going to do with Cryptokitis and so forth? Now people are asking class about scalability and more about use cases. And this, you know, the questions are changing from year to year, right? Starkware is organizing the Starkware Sessions conference during the Tel Aviv blockchain week in September, and you should definitely consider going.

In case you don't know about Starkware, they're productizing zero-knowledge proofs to solve two of the blockchain ecosystem's most pressing issues. Scalability and privacy. And Starkware is co-founded by Eli Ben Sassouin, who was previously on the show. previously on the show. The conference will cover some of the most cutting ads research and applications in the field of zero knowledge proofs. And you can expect only the brightest minds in the space to discuss things like self-custodial trading, Starks for Ler 1, Stark-friendly hash functions, and other cool things you can do with Stark proofs. Many of the speakers are Epicenter alumni, including Ethereum researchers, Vitalik Buteran, Alexei Akunov, and Justin Drake. Martin Copelman

of Gnosis will also be speaking, as well as Arthur Brightman of Tezos. So if you're interested in broadening your understanding of these cutting-edge technologies, there's no better place to do it than Starkware sessions. Join the conference in Tel Aviv on September 16th or come a day early for the Stark 101 Workshop where you'll build a Stark Prover from scratch. Tickets are on sale now and you can find the registration page at epicenter.org slash Starkware. That's S T-A-R-K-W-A-R-E. The first 50 people to use the code Epicenter will get 20% off the regular ticket price. We'd like to thank Starkware Sessions for their support of Epicenter. So let's talk about the POA network first.

It's a proof of authority network. So there's a number of validators in this network. How does one become validator in this network? Yeah, we slightly renamed it. We called proof of autonomy because we think that it's a network based on DAO. It's exclusive DAO by validators and exclusive means that there's a limited group of people who can make decisions within this DAO and they can. let new validators in or they can let out validators on the network.

Right? So the each POA network starts from a trusted ceremony. So there is a bootstraping of this network and we have a special role called Master of Ceremonies. So this role bootstraping new network and onboard minimum required number of validators who will onboard new validators through the governance process, right? So POA network started with one master of ceremonies and three first validators. After that, validators received a governance instrument, which we call voting tab, where they can propose ballots,

and each validator can propose ballot to add or remove a validator from the network. And there is a quorum decision by other validators with equal votes to onboard new validators. To be a new validator, usually validators apply on the forum, or they can apply on some, probably apply somewhere else, but how it started, they apply with their application, and there are some requirements, and there are some requirements which limit the number of subset of potential validators from the subset of all people. But, you know, the subset which can be validators is quite big.

And yeah, it's also interesting that validators of POA are individuals. And as far as I know, it's the only network which is managed only by individuals, not companies. If you look into any other proof of stake or POA consensus, it's usually it's a form of consortiums of companies or financial organizations or some. In proof of stake, it's more likely professional professional. companies, which are providing this staking services, basically, right? And POE is the network managed only by individuals.

Like, there are no companies out there. So what happens if a validator misbehaves? What happens if they act maliciously? Can, do they have to stake some funds that can then be slashed? Or how do you make sure that they behave in a way that benefits the network as a whole? Yeah, we think that, well, there is a, way for other validators to remove validator from the consensus. So any validator can propose a ballot to remove a validator from the consensus. Usually misbehaving in POE is, when validators,

for some reason, is not running his or her node, right? So that's like, that's a common misbehaving that we saw during these two years. Then, for example, they forgot to pay for for the bill or because of the misconfiguration, let's say, parameters of their node is too low and so forth, right? So validators can vote out a validator and we had multiple examples when validators voted out other validators. But the good thing here is that the consensus can tolerate a number of faulty nodes. and even if one validator is, well, if one validator is misbehaving, it's not a problem, right, because the consensus is fault-tolerant. Yeah, validators have this governance application, and this application is a part of the consensus.

So it runs within smart contract, and smart contract is connected to the consensus later. And I read that validators, in order to be a validator, you have to be a U.S. notary. Is that right? Yeah, that's right. That's right. And some people ask like why, right? And do you want to ask why? Or like any other questions regarding notaries?

Yeah. Yeah. Why U.S. notaries? Why specifically that? Okay. So why the U.S. notaries? Because first of all, it's a great way of making K.YC on validators.

Right. When we think about individuals, what is the best way to get KYC? When people apply to be US not only government, it's a US government on state level, is making KYC checks on applicants, and results of these checks are publicly available, right? So also validators are taking some public responsibilities, provide services, and they cannot decline providing their identity to any third party who is interested in cross-validating identity. right if you ask i don't know a validator of any proof of stake network to confirm their identity they can do this or not right because it's on like for you as a like a concerned uh let's say

token holder with um with pea validators you can ask to to make a like not really related document from this validator and this way you can cross validate their identity like you can actually ask them to notarize something for you it's outside of the protocol but you can this way you can cross validate their identity. Also, because validators are individuals, it's interesting to understand, you know, what's their background, especially as a validator. I don't want to have other validators from like some, like social groups. For example, I don't want to see criminals as validators, right?

And when notaries cannot be, you cannot have previous criminals. record. So that's also a good way to prove. And also, notary law is regulated like state by state. So there is no organization in the U.S. which can say, okay, guys, you are notaries, you cannot do this. Right. So it's regulated state by state. There is a diverse group of file data from different state, which is good for, actually, it's good for decentralization, right, when we think about this. Well, when people are notaries, they already agree to share information about themselves, including their residential address in public. So you can validate residential addresses of all validators.

Why the US notaries? Because we thought that can be interesting use case for US-based network. And till now, the B.O.A network is, as far as I know, it's the only US-based network, which is like all validators are US-based and it's public network with like public token. It's a good experiment. We thought that this type of network can be used with some local specific use cases. For example, if some local or federal authorities will be interested to put some information on chain, we thought that the POA-based network can be an advantage.

It didn't happen, but yeah, it can be. So how did you find a number of computer-savvy technology forward notaries who are interested in being a validator and how do you incentivize them to be one? Yeah, with first fall data, it was quite hard, especially for the first one. So I presented on Ethereum meetups. I'm co-organizer of Silicon Silicon Valley Ethereum Meetup. So I presented over there. It's actually when I announced the project. And a few validators applied to be a validator after my presentation on Silicon Valley

Ethereum meetup. And after I visited conferences and meetups, and usually from one presentation, we had like one candidate. Yeah, that's how it started. And after it was easier because when the project received some hype and also when the token started to be traded publicly, validators had clear incentive. So we designed the protocol that the way that each block, one coin is created, and this coin is going to a validator who created this block.

So all validators are incentivized by the protocol to make their, how we call it, public duty by private actors, right? So they're private actors, but they're doing public duty by participating in this type of DAO. It's exclusive DAO because it's a limited group of validators, but it's still decentralized autonomous organization and they're not. One more thing with individuals, it's quite interesting that individuals, they cannot buy each other, right, like legally. Companies can form some, you know, can buy shares of each other companies. They can exchange tokens and so forth, right? So individuals are atomic units.

Validators agreed that they will not form relations. And we had one validator and who invited his girlfriend to be a validator. But after they decided to be a family, he stepped out and now only she is a validator, which is quite cool, right? but she's keeping the rules, like social rules. So they agreed not to form families between validators, or if a family is formed, like one of validators should quit or both. Let's talk about security. You know, DAPs are pretty unique because unlike other types of software,

they can hold astronomical amounts of value. That's why getting systems audited, creating robust security processes, and fostering a culture of security in your organization is so important. And to do this, you should only trust experts with real security expertise. There are a lot of security firms in the blockchain space, but few have the experience and track record of Trail of Bits. And they've been in business since 2012, long before things like the Dow Hack were even imaginable. Trail of Bits works with your team to audit every aspect of your project.

And smart contract code is just the beginning. They'll help you implement best practices around things like DevOps, key storage, and user-facing applications. And once your software has been rigorously tested and reviewed by Trail of Bits, they'll provide the tools you need to make sure that your code remains safe over every new commit. They can even put a software security expert at your team's disposal who'll give you advice and answer your questions when you need them. It's like having your own security engineer on staff, but don't take my word for it.

Go to their publications repo on GitHub to read their papers, presentations, and security reviews. It's no wonder teams like parity, status, new cipher, and or, Organizations like Facebook and DARPA trust Trail ofBits for their security audits. To learn more, go to trailfbits.com, and if you decide to reach out, make sure you let them know you heard about them on Epicenter. We'd like to thank Trail of Bits for their support. In the US legal language, there is this concept of an unincorporated association or an unincorporated partnership. So this is like a collection of individuals that is doing some activity, but they have not incorporated in any of these legal structures. like LLC, S-Corp, C-Corp, etc.

The thing is like if you're unincorporated partnership or association, and if there's ever a legal lawsuit against that association, then the members of that association are individually infinitely liable for if something goes wrong in that lawsuit. So don't your validators worry that this kind of organization where they are identifiable individuals with their records accessible, if let's say there was a hack of the POA network and people lost a bunch of POA tokens or X-Di tokens or whatever,

then all of these individuals are going to be legally liable for that event. It's a good question. We don't know about lawsuit against this type of formations. The first question I asked myself when I started to design the protocol was like, is it even legal, right, for individuals to create their own, basically, tokens and protect someone else resources. And that was my first question,

and I asked for us. And I remember when I asked this question, my perception was like, I'm doing something wrong. Because I thought it was not possible to do something like this. But they explained that it's totally feasible and there are no, legal risks. The same question was with our Bridges product, which can, you know, allow individuals to transfer tokens between networks. Like, is it legal to transfer tokens or not? Validators can be legally binded to lawsuits if, like, whatever they're doing is against, let's say, token holders, right? If they, like, collude technically it's possible. And from my perspective as a protocol designer, I think it's great, right? So it's what protects

this DAO from colluding because they understand that they're publicly exposed persons and they're taking this responsibility. I tried to call this skin on the game, right? They didn't like it, that I call the skin on the game. So we call it identity at stake. So like most of the validators, they already, you know, they're people with, you know, their own, like, families and, you know, their own lives. They don't want to risk this, like, status quo. And they, actually, they understand what they're doing and they agree on this. It's interesting that we have, like, we have different, different professions. Some people think, like, oh, most likely all validators are, U.S. notary in the U.S. it's like a separate way of, it's not a job, right, for most people.

It's just like additional source of revenue for most people. So there are different backgrounds. There are software engineers, business owners, and also there are lawyers, right? So lawyers, they can, they understand what they're doing. And I think that they understand what they're doing. But theoretically, they can face a lawsuit and that's like anyone can face a lawsuit. Of course, of course, this is a question that is not specific for just POA. So like a few months back, I was talking to one of the one of the large investors of MakerDA,

huge MKR holder. And they were talking about these designs of these committees, you know, like the stability fee committee and this fee committee and that fee committee. And committee members there are individuals and they have similar risks. Now, if you're in a committee and you jack up the stability fee and somebody like loses money because you jacked up the fee too high or too low, what protects you from a lawsuit there? Yeah, yeah, I understand totally. We worked with multiple lawyers on legal opinions on our products.

actually it was good to see Stephen Polly here in your podcast. We were a client of him. But I understand that's possible. I'm not a lawyer. I cannot guarantee anything to people participating. And it's possible that I made a mistake when I designed the protocol. And we had validators had many questions. For example, how to make, you know, they're all Americans, right?

they all, well, not all Americans, all US residents, because you can be a notary in the US without being American. So we have a few Canadians and one Russian citizen, but they're US notaries and they participate in the consensus. So the question is like how to tax revenue from the consensus. In some networks, some people don't care because they have some, you know, Cayman entity and they don't care. but here all validators are exposed

and it was a question and they had long discussions about this and well tax season ended so I hope validators made it right and would you want to extend so right now only US notaries can be

validators but in principle you could allow German notaries to be validators as well and would you would you want to extend this geographically or keep it limited to the US? Right. So the idea of side chains is that we can, if we think about scalability, right, and scalability not only about transactions per second, but also about like use cases and roadmap of the network, we can, we can extend it vertically, right? So we can think, okay, so US validators is like not enough, let's add, you know, some other jurisdictions with

their own problems, you know. Let's add some Europeans validators with GDPR regulation. Or we can, we can make horizontal scaling and we can say, yeah, this model is interesting in the US, but let's have a network with, you know, European validators and so forth. So that was our, like, first idea about validator set. And I think that POA validator set is unique and it's good to keep it as it is, right, like US-based network. with not only as validators. And if you want to experiment, we can start a new network

with a new set of validators. With Ex-Dai, we started with network with basically with MakerDAO, right, like with two validators, P-O-N Maker-Dou. Now we have 10 validators, different companies,

organizations from different countries, no individuals at all. Like all members are organizations. Some of them are DAO, like Metacartel or Giveth, Right. Some of them are companies like POA, some, you know, and so forth. So that's the way, right? So if we want to have the same consensus, but with different sets of validators, we can launch one more network.

And there are multiple networks launched or planned to launch on the same consensus and governance model, like Ocean Protocol, Loxo, Loxo, artist network from Austria. Yeah, several networks launched on the same ideas. and the validator set is different. And also they experiment. For example, RTC's network, they have validators with the same governance structure, but validator should put a bond, right? It's not staking.

They should put a bond. If they put a bond, they can be a validator, can be onboarded by the validators, which is quite cool, right, because it creates some locking mechanism within the token economy, which is good for token holders. So what all of these networks have in common is the consensus mechanism.

Can you explain how that works? The consensus of POE? Yes. Well, the main idea is that there is a set of validators, which is managed by smart contracts, by DAO, right? And this is known as a Zeeble protection mechanism, right? Because if we understand who validators, are and why they can create block, we can understand if the network is operating under some

assumptions of, like, zibble protection. For example, in proof of work, right, if you have more power, you're a validator for this block. And, well, more power and you're lucky to create a block, right? In proof of stake, there's a set of stakers who have more coins than any other people from the subset. In delegated proof of stake, there is this set of people, more tokens than anyone else staked. And also there's a delegation people who delegated, right? In POA, there is a set of people who are selected by subset of this set. And the whole set can manage this set by excluding validators or including them, based on DAO decision. So the governance is based on DAO, and the main question validators is deciding on is who are validators.

So this is a part of Zibl control. So second important part is reward distribution. The protocol distributes reward to validators. It's embedded within the protocol. And validators also have what we call self-sustainability emission. So some emissions they can spend on the protocol to support R&D, and so forth. But the underlying BFT consensus can be swapped.

For now, for now we are using ORA consensus, which is developed by parity technologies. And we plan to migrate to a new censorship-resistant consensus in 2020. So that's our plan. Well, it's quite easy thing because when we think, what is more important, what is the most important for pure consensus is the answer to the question who validators are and why validators are these keys, because they were voted in by other validators.

And when you upgrade the networks, say when you switch to the Honeybadger protocol next year, will all validators actually have to agree with this upgrade or how do you follow through on this? Once a year we have a security incident, which can be like incident type of consensus fault. so we have to upgrade protocols. It's because some problems within implementation of, let's say, parity client or some Barks and Ethereum protocol which can cause these problems.

So at least once a year, we have to propose updates to validate us and they have to agree on this. So this type of upgrades are usually proposed in emergency forms. So for validators, they don't have much time to think about it, right?

Most of the time, the upgrades are are proposed for the test network and after the test network they propose to the POE main network in the form of hard force or protocol upgrades. So validate us. Some of them have software engineering background. We have validators working, you know, Amazon, Facebook, some other companies. They can review some changes and agree on these changes or not, right?

If we propose upgrade of the buck and the consensus, which is known, they will upgrade it. But let's say if we propose, I don't know, switching from POA to DPS, it will be very different discussions. So most likely they will ask many questions and I don't know outcomes of this decision by validators. That's not my network, right? That's their network. Sure. Can anyone propose changes or who has the power to propose changes to the protocol? Technically, anyone can propose changes. So anyone can make APR or validators themselves can make their own changes.

This Zeeble control and reward distribution part is supports upgradeability through smart contracts. So implementation can be uploaded to anyone in a form of, of a smart contract, right? And validators can vote to replace this new implementation, like old implementation to new implementation, using on-chain voting without any, without asking anyone, right? Any validator can propose a change to the whole consensus

with the upgradeability of the consensus and also with upgradability of the client. So technically they can replace the, everything. They can do it for good or for bad. If you're holding a significant portion of your net worth in crypto, you're probably waiting for your portfolio to moon at any time. But holding crypto doesn't mean you should be irresponsible in the face of volatility risk. That's where Voltauro comes in. Voltauro is the leading gold hedging solution for the crypto community. And as a stable asset,

trusted for millennia, gold is the perfect long-term hedging solution. And at Epicenter, we've been using Voltauro since 2014 to protect a portion of our company's assets against volatility. Now, you might ask, why not use a stablecoin, Seb? Which is a great. And don't get me wrong, stablecoins are great and a real benefit for crypto adoption. But algorithmic stable coins are still very new and experimental asset type. And some asset-backed stable coins have been scrutinized for being under-reserved. With Voltoro, your gold is 100% insured and secure it in vaults deep in the Swiss mountains protected by Brinks. Every single gram of gold is audited and holdings are made transparently available on their

website for anyone to verify. And most importantly, it's quite literally your gold. You can choose to have it delivered to you at any time. To learn more and to get access to Voltoro's brand new V2 platform, which includes an interface overhaul and trading in dash, like coin, ether, and silver, go to Volturo.gold slash epicenter. That's V-A-U-L-T-O-R-O-D-Gold slash epicenter. We'd like to thank Vulturo for the support of the podcast. If I'm a validator, and you're a validator, the current design is both of us will make the same money

in a year if our infrastructure has the same performance. That's right. What we saw that when token price is going down, for some validators, it's not interesting to be validators anymore. So they step out

and the rewards per year, as you said, for the rest validators, are increasing, right? Because they have smaller rounds and they basically create more points. That's something that we observed at validators, some validators stepping out because the reward is not interesting

for them anymore. But in a system like this, if I am a validator, I would naturally want the validator set to be small. Because the bigger the set grows, the more diluted I get. There are some points, right? So like minimum number.

of validators for the governance is three validators. So without, if you have less than three validators, then you cannot make governance decisions, right? You need at least two versus one, right? Like two or three to vote for a governance decision. So maximum number of validators, let's say 3,000, right? It's a limitation of state storage within the current setting that we have with like the same in Ethereum, right?

The optimal, what is the optimal number of validators? Like when I asked this question myself, like what is the number? First I looked into other chains, EOS 21 validator. Is it enough or not? For example. And also I looked into academic research. There is a paper by Emin Goun and some other researchers from Cornell, both decentralization in Bitcoin in Ethereum and Ethereum.

And Hick had a claim within this paper than a number of validators, like more than 20 in a quorum type of system. It's a... I cannot repeat it as it is. But basic idea is that if you have 20 validators or more, then this is enough to be like more decentralized in the current state of Bitcoin and Ethereum. So that was my like base point. Okay, guys, we need to have like 20 or more validators. validators. So I told them like an ideal number of validators, when I bootstrapped the network, I thought it would be great if we can have like 25 validators. And they never reached 25. And now it's 18 plus Master of Ceremonies, like the validator without voting rights. Yeah, that's 19 validators. When X-Di started, it was only one validator. Because I didn't find a group of people

who wanted to be validated so next day. Because no incentive, no token, no reward, nothing. You just have to run a node, pay money for it, and people were not interested in running this network. And it was quite hard to add, you know, first validators. Yeah, I think that based on my, like, based on our experience with updates, the validator set of, let's say, 20 to 25,

people in basically in different time zones can upgrade the protocol within 24 hours which is not bad so within 24 hours they can solve coordination problem to like to install like new

like to agree that they have to install new software and install it on their servers deploy it and confirm and discuss again and wait you know all that stuff so it's a it's quite good. But with, like, just imagine we have thousands of validators.

This coordination problem will be much harder to solve, right, without automatic updates and, you know, all this centralization points. There's also a token in this network, the power token. So what does that token do? Well, the protocols was designed in 2017. And we looked into how to make it. to make Ethereum protocol more scalable. So basically copy the idea of the Ethereum utility of the token.

And basically it's a way to pay for gas. And also unit of account and medium of exchange on the network. That was the basic promise of this token. Later on, we exchanged use case. We added additional use cases. So the interesting use case was with the first breach which we launched on POE. So this native token, we wrapped it into ERC20 token on main net. And this token, it was like the first ERC20 token connected to coin payments.

So it was possible to use it as a medium of exchange between, let's say, vendors and clients. also it can be landed on borrow it on if land swapped to other tokens on uniswap traded on in a wrapped form or in native tokens on exchanges so there is like arbitration speculation features which is also a part of any token economy right and and possibly will be some staking features if validators of POE decide to to migrate to our new dedicated proof of state consensus Tell us about this new delegated POS. Yeah, this delegated for proof of stake is designed for X-Di first.

So one of the main questions to validators, people think that this structure is centralized. I don't call it centralized, I call it exclusive. Validators can decide who will be new validators. And we can see on the forum that validators don't like some validators. validators, right? And reasons, for example, one obvious reason because a new validator don't know any other validators, right, so people don't trust him. That's one of the problem, like exclusiveness of the protocol, is something that can be not good in some situations.

For example, in potential censorship problems, when a validator set is small and exclusive, and there are some orders, for example, to stop and not do whatever you do, this exclusive validator set can break promises of flivness of the network, right, of the consensus. That's one of the main reason to make the public permissioned network with inclusive access to seat of validators. So that's why we need staking, because anyone with staking token can participate. in a staking as a candidate who can be elected as a validator. And to bring reputational part to this election

and also to select a subset of validators not only based on their own stake, but also by public trust, we added delegation to this consensus, and that's why it's delegated proof of stake. Some people don't like delegated proof-of-stake ideas, especially after EOS and they think it's still centralized. So we looked into problems with EOS and like what people think that it's centralized in EOS and decided to design it differently. For example, delegators in DPS consensus by POE are incentivized by the protocol. so they receive part of the emission reward, they can get some fees from bridges and so forth.

And EOS, they don't get any reward from the protocol, and the reward they get is under the table, basically. And also how reward is distributed between validators and delegators is a bit different. All validators within the validator set, who are elected to be validators, for the staking epoch, get equal reward. Right?

So no matter how many coins you have, you will get equally reward for the, for this slot, right? Let's say if we have 19 validators, each validator will get one 19th. Within each of this one-ninth validator has distribution between validator and all delegators who staked on this delegator. And this distribution is actually also fair. So delegators, when they delegate more on validator, they decrease rewards of the validator,

and they compete between each other. But we can imagine that delegators will have so many coins that reward of the validator will be zero, right, because the pressure will be pressure of coins delegated on this validators would be too big. So that's why we added what we call saturation point. and below the saturation point, reward of file data cannot get below, right, of the saturation point. So, like, worst case in a scenario, each file data can be sure that they will get 30% out of 1.19th of the reward per staking epoch. So that's something that we added for fairness. And also I'd like to mention that the staking consensus, reward distribution,

Zeeble control, all this consensus-related parts of this protocol, but not BFT part. They're all implemented in solidity, which is good for developers because they can, let's say, if someone is interested in taking this protocol and adapting it for their own sidechains, they can take it and tinker with it. And, you know, it's a high-level language. It runs with an EVM, like all advantages of solidity. access to developers, access to smart contract auditors, formal verification, which is very hard to get on, like,

protocol built on low level, let's say. What's really fascinating about POA is that you're like, challenging conventional wisdom on many dimensions. and you have a protocol that does many things that other protocol designers just wouldn't touch. Now, there are reasons why protocol designers wouldn't touch things like this. So, for example, this idea that you can have a minimum reward rate for the validator. It cannot go below 30% of 1 by 19. This is an issue that's being debated in the cosmos ecosystem.

Should they be like a UBI for validators? Should there be like a basic income for all of these validators? The problem that prevents something like Cosmos from doing it is that the argument is if you give a UBI to all the validators, well, the validators will just promise the delegators, hey, delegate with me and I will give you part of my UBI as extra rewards outside the protocol. And so the UBI will not function as a UBI because the validators and the delegates can break the UBI outside the chain. Now, it feels like you haven't worried about a problem like that at all and you will implement a UBI and we'll get to see the result of that.

Right, and this reward is only from stakes tokens. there is no reward that they distribute between them as a reward per block, right? They only distribute reward that they get from the tokens they stake, right? So there is an emission rate, and they get reward based on this emission rate from the token they staked, and people staked on top of them, right? So there is no reward which is created out of nothing, each block, and after based on some distribution you have to distribute it between different parties. Also these DPS consensus supporting multi-chain staking. So the staking token lives on main net, on Ethereum mainnet,

and can be bridged to X-Dai or can be bridged to another network with the same emission curve, right, and the same supply. So if I stake on X-Di-1, I'll get my reward on X-Di-1, and I cannot stake on X-Di2, right? Because bridge knows that I breached my token to X-Di-1. But if I see that there are too many delegators on X-Di-1, or there is only one, or let's say, only three validators on X-Di-2, I can bridge my tokens to the next network and delegate on validators on the second network.

I took this concept from MMO RPG. So if you ever played World of Warcraft, there are servers which are overpopulated. And you have the same kind of game setting, but it's more people. It means that you have to wait longer in lines and some bosses. You have to wait for the respawn.

So you can move your character to another service. using a bridge, right? And on this server, you can play with all your items and get whatever you want from the game, but with different, let's say, economic terms, right? But it's kind of the same game. So that's what we call multi-chain staking, when people can, and also it solves our problem

with, like, what if we want to launch one more network? Like, okay, today we enjoy X-Dai, but the same concept can work with, different with different stable tokens. I'm very interested in Ampleforth, which is a very interesting concept, and some other stable tokens. Can we launch one more network like Ex-Di, but with different native token and the same staking token.

So it's possible. So we've talked about X-Dai for a while now. Maybe we should give a brief introduction of what that actually is. Would you do that? Yeah, sure. So X-Di is what we call the first ever stable chain. So it's a new side chain based on Ethereum protocol with Dai as a native token. How to make die as a native token?

We have an interoperability solution token bridge. And this token bridge can bridge die from Mainnet to X-Dai. So on Mainnet, this bridge allows to lock-unlock die. right and on the ex-di side this bridge allows to mint and burn the x-day so minting and burning of native tokens is something that is unique for for blockchains because usually the emission curve is defined by the protocol and external events cannot just you know send to consensus messages okay please print this amount of coins for this user right That's something that we don't have usually in consensus algorithms for native tokens.

So with X-Di, each block is created with empty reward, like no reward at all. But when Di is locked on the Ethereum site, the bridge can relay this event, and this event is a command to the consensus to mint new coins for a user who locked these coins on the Ethereum side. The market cap of X-Dai is like 20, 32,000 dye. So this amount of dye is locked on Ethereum Mainnet side and minted on X-Dye side. That's something that gives very interesting perception when you use the network. Because if you have an account balance in DAI, it looks like you have an account balance in dollars. Right.

And when time passes, you have the same account. balance us if you don't spend it. And it's something that is very convenient for us in real life, right? And it's very hard to get in cryptocurrencies. That's why this protocol, we started to think how to use it in peer-to-peer payments. And there are some very exciting developments around the protocol by the community. The most well-known is a burner wallet by Austin Griffith. and the Pocketto, which is a burner wallet implementation on iOS and Android by the Pocateto team, who, like, community created this wallet. The second big use case for F.XTI is a platform for smart contracts.

It's interesting to use it for smart contract developers, because they can understand how much will they pay for, for platform usage, right? And also they can give stable balances to their users. And they can calculate unit economy of their smart contracts, which is also important for many use cases. Like if you discuss the economy of a game with a game developer, they have calculation of lifetime value of a user, right?

And how much they spend to get the user, how much they spent on, basically, hosting the user. and what is the revenue of this game developer. And with volatile tokens, like on Ethereum Mainnet, it's hard to predict whether the price of Ethereum will be in a year or in a day. So the POA and XDI networks are not trustless, but they offer fast finality and low transaction costs. In some way, this limits the use cases for the POA mechanism.

So what are things that you think definitely should absolutely not run on POA? For me, it's interesting to see any use case which can be run on the chain is interesting to see. And the question that we don't know the answer for is how the network will operate under conditions of pressure, both on validators and denial of service attacks, which didn't happen much on side chains. So for us, it's quite interesting to see the old type of applications deployed on the network, although we cannot facilitate some of these activities. We are working on bringing private transactions using ZK SNARCs.

We're not developing SNARCs ourselves, but we're helping other teams to bring mixers to PO and X-Di and to integrate these mixers into some wallets like Burner wallet and Pocateto. We have for this ZK privacy fund, and we already distributed 40,000 die during the last two months, and we plan to distribute 60,000 die more. And likely this projects will bring use cases, which can be interesting, right, for the censorship resistance and so forth.

I mean that public networks should not afraid of specific use cases. If public network can operate with any use case, then it's truly public network. Cool. And what kind of applications are getting the most traction on the POA network and on X-Dai? With POA, now we are focusing on use cases with games, blockchain games. And at the moment, if you open state of the DAPs, which is a ranking site for platforms with smart contracts, the game number one is called GON. It's a geocassion game.

It's based on PEOA network. So they decided to use it because it provides tempers. temporary scalability solution and they don't need to rewrite their application to run it on a side chain. It's actually quite interesting game because it's not showing wallets and tokens and balances in tokens to users and they're getting correction and after they will get enough traction with users, they will open this token economy to people. The use case for X-Di is what we plan to focus on and facilitate is peer-to-peer payments, so integration with more wallets.

Very exciting development with X-Dai is integration with Discord. So X-Di is integrated with Discord tip button. It allows, you know, tipping and the airdrobes and some form of gambling, like dice and so forth within the chat, which is quite cool to make in stable coins. And also, we plan to support X-Di as many wallets as possible and facilitate real-world usage of this stable coin. So basically, POA for games and X-Di for payments. What do you hope to achieve in the next year or so with POA and X-Dai?

The main achievement for us will be migration to delegated proof of stake. And the second part of this migration is migration to new BFT consensus protocol, which we are working on for the last year. It's called HoneyBadger BFT. It's quite exciting because it's asynchronous consensus and censorship resistant with on-chain randomness. and it's already implemented. We made the security audits for it

and now integrating with several partners with Luxo and Artis, we're integrating it into parity Ethereum client and when it's integrated, we'll send upstream to parity and let other people to use this exciting consensus algorithm. There are not many censorship-resistant

consensus out there. And HoneyBedger was quite hard to implement. We also had educational articles how the consensus is working and what are our extension to this consensus, like Dynamic HoneyBadger. Because HoneyBadger itself is not a designer to support some features which are important in the POA or DPOA settings.

For example, it doesn't support dynamic validators, set when we can, you know, add or remove our data from the consensus. And that's something that we had to implement on top of HoneyBedger. So we have educational materials on our forum. It's a very hard consensus as explained and it's easy to understand language. Thank you for being on the show, Igor. That was super interesting. I'm excited to see where this project goes. Yeah. Thank you very much for hope. posting me here. Thank you for joining us on this week's episode. We release new episodes every

week. You can find and subscribe to the show on iTunes, Spotify, YouTube, SoundCloud, or wherever you listen to podcasts. And if you have a Google Home or Alexa device, you can tell it to listen to the latest episode of the Epicenter podcast. Go to epicenter.tv slash subscribe for a full list of places where you can watch and listen. And while you're there, be sure to sign up for the newsletter, so you get new episodes in your inbox as they're released. If you want to interact with us, guests or other podcast listeners, you can follow us on Twitter. And please leave us a review on iTunes. It helps people find the show, and we're always happy to read them.

So thanks so much, and we look forward to being back next week.