Zero Knowledge Podcast: Interview with Igor Barinov and Austin Griffith


#1

Zero Knowledge Podcast has just released the latest episode which feature interviews with both Igor Barinov and Austin Griffith where they both discuss all things POA related, from POA Core network, Bridges, xDai Chain and Burner Wallet. :bridge_at_night: :fire: :computer:

Check out the latest episode below!:point_down::point_down::point_down:

If you would prefer to read instead of listen, then check out the transcript here !:point_down::point_down::point_down:

Zero Knowledge Podcast: Bridges, xDai and Burner Wallets with Igor & Austin (transcribed by Sonix)

Anna Rose: Welcome to Zero Knowledge podcast where we explore the latest in block chain technology and the decentralized web. The show is hosted by me Anna. And me Frederic

Fredrik Harrysson: In this episode we sit down with Austin and eager to talk about bridges Eckstein and Austin's latest project that ties all these things together called the burner.

Anna Rose: So before we jump into this week's episode I want to again let you know about the zero knowledge summit that's coming up on March 22nd in Berlin. It'll be a day full of workshops talks and breakout sessions all about snark Starks and zero knowledge topics. If you are a dev or researcher working on these topics or looking to dive in please submit an application to attend. The link is in the description and just FII it is a small technical event with limited spots so be sure to get your application in as soon as possible. So now here's our episode on bridges xDai in the burner wallet.

Today we have Igor from POA and Austin from Gitcoin labs as well as our usual co-host Anna. Welcome everyone. So why don't you guys kick off with introducing yourself your works or what your background is and how you got into this space. Austin you can go first.

Austin Griffith: Yeah. So I got started just as a lowly DAPP developer. I built a couple of little games and as I played with the games I realized that UX was really clunky I had to hit yes and met a mask every time I wanted to drop my you know my fishing bait into my fishing game. And so then I started playing around with how we make you act better and how we kind of abstract away gas and that led me to many transactions and then eventually to doing our candy for a coin and then building out the burner wallet which is kind of this this web wallet that lives in your browser and it's kind of helping with onboarding and it makes the onboarding process a lot smoother and there are some tradeoffs there with security because your private keys live in local storage but we can kind of get into that later but that's that's basically my story.

And what about you Igor what's your story.

Igor Barinov: I have a long story. I started in bitcoin time. I worked as a consultant for several blockchain companies and I transitioned to a consulting company based in Singapore and helped them to build consortium types of networks.

Igor Barinov: And after that the company where I worked for they closed their blockchain department and we decided to start working on something that we were familiar with and to solve some problems which were interesting. And two years ago we thought about how can we make Ethereum network with some form of permissioned consensus. But with a token which can be tradable on exchanges basically how to make public permissioned network. And our main idea was can we make something like DAO and put this DAO in as a consensus path. And that's how we created this idea of the first POA Network which is what we call POA Core DAO with some independent and non affiliated validators can decide who are validators in this network. And as far as I know its the first implementation of this type of what we call metaconsensus where we have a layer of consensus of underlying BFT consensus and on top of it we'll have this DAO which manage this validators set. This network was launched in December 2017 and that in production since then and gave us tools and ideas for further experiments.

Anna Rose: I think it's great to hear where you guys are both coming from. I know that one of the big topics we wanted to start with today was the concept of bridges. I think maybe as a start you can explain how can you explain sort of the POA Network idea. But why why did you need to build bridges.

The idea of POA Network is to have sidechains. Sidechains means new chains with their own state and they operate in public internet and they're not connected to main Ethereum network and when you need for some reason to move assets from network to network you have to use some centralized solutions as we know as exchanges. So exchange is a is a form of bridge where state of one network is moved into a centralized state of the of the exchange and after moved after some transformation of the decentralized state on another network. So why do we need to have bridge for POA Network? It's very simple. If you have if you ever start a sidechain or a new blockchain and you're successful and launch it, one of the main question is how to get listed this sidechain on exchanges. And this is what usually people don't talk about because it's a recommendation from lawyers to not discuss exchange listing and so forth. But in reality you have to do this. Otherwise community which we call contributors but many people call them investors will put pressure on projects and this project usually starts with a choir with their word when and after like when "when token sale, when mainnet" And after "when exchanges".

Igor Barinov: So "when exchanges" is crucial question after you have your mainnet launched and we had an idea "Okay guys it's kinda hard to make a native token listed on major exchanges. So let's have it listed that at least on one big exchange. But also let's make our token tradable on DEXs decentralized exchanges" and to make it tradable on DEXs as you have to have this token on Ethereum So some projects start ERC20 token and after they migrate from Ethereum to their own sidechain like EOS, right or Tron. But some projects like POA it's not a very popular way but they start mainnet first and that that they bring their native token into ERC20 representation on Ethereum mainnet. That's that's that's why we've built our first bridge and this Bridges is exciting for many reasons. So first of all the liquidity of this sidechain two networks at the moment. Right. So ninety five percent is on sidechain itself and five percent on mainnet. This token is tradable on exchanges on both sides. And these bridge is in operations since May 2018 and then production and since then without outages without security incidents.

Anna Rose: So before you continue because one thing I think we haven't really done yet is define exactly what a bridge is. And Frederik I don't know if you want to kind of start that off. Like do you want to start off that definition because I know that parody also has done some bridge work.

Fredrik Harrysson: Sure sure. First I wanted to add a comment as well to what you are saying about the exchanges because when an exchange goes to list something if it's a native token and they want to list that they need to run a full node of that token. They need to integrate that into their whole framework. And this is a ton of work for an exchange. But then ERC20 tokens they already have all of the infrastructure to basically just flip a switch to it. So that's why it's a lot easier for an exchange to list in the ERC20 token than a native token I like they don't have to set up all this infrastructure through run extra stuff. But yeah. So bridges it's a fascinating topic it's something that's been going around for a long time. There's obviously now projects devoted entirely to like build bridge infrastructure like Cosmos and Polkadot it out to some extent. And what POA is doing. And yeah like we at Parity started this says we had to be a proof of authority networks running and companies and various things and these companies wanted to either interact with data in the mainnet that's in some way or like we didn't really necessarily come to it from the talking perspective.

That was the first thing we built. That was sort of the easy thing. But like a lot of them they want to run a completely private network with you know their own stakeholders and authorities. So what what POA Network has been doing like this DAO on top I mean essentially kind of drives POA towards POS where there is like another layer of governance and organization around who these authorities are and it's not just like you know these three big banks are the authorities. So yeah I think it makes total sense and that more model to then want to like bring a lot of this stuff back to main. But yeah bridge as a concept you know Ethereum to Ethereum bridge is one thing. Bridge to like bitcoin is something completely different so it's hard to talk about it as a generic concept but a bridge in Ethereum to Ethereum is one smart contract on either end and a set of bridge validators and these can be the authorities of the POA network or they can be chosen in some crypto economic game. There's various strategies here.

Anna Rose: In that example like would, so say it's like ERC20 a theory and may net bridge to something else. There could be validators or an a POS system on one on the sidechain. But what is the who are the validators exactly on the Ethereum mainnet. mainnet. Because right now it's still proof of work.

Igor Barinov: So there are validators between networks and these validators between networks can be validators on the sidechain or can they can be independent party. For example on an xDai Chain which we're going to talk about today the validators of the consensus are the same validators as validators of the bridge and also what we added to the bridge is governance and also they're part of governance. So on the bridge there is a three or four signatures required to relay an event between two networks and the same on multisig which manage the bridge, three or four signatures are required to change list of validators and three or four signatures are required to modify elements of the bridge and also three or four signatures are required to shut down the bridge or upgrade the bridge. So it's the same model.

Igor Barinov: The the the thing that I like about bridges that set up so can be very different for different use cases. Let me a tell bit about the history of the bridge and decisions we had. So we start as first to explore a concept of bridge. Basically there are two families of bridges; in one family you relay part of the state of one network to another network. It's what BTC relay or piece really or plasma like relay are doing and another family oracles based bridge where a set of oracles observe event on both networks and after a number of event or one transactions are what we call aggregated they relay a message in the form of transaction from, basically from network to network. Within smart contract they can make lock unlock of tokens or mint and burn and these different modes of bridge operation can give different scenarios. For example POA20 bridge where we move native tokens from POA Network to ERC20 representation which we call POA20, native tokens are locked and unlocked. And validators relaying this event create new tokens on the Ethereum mainnet. So they mint and when tokens relayed back, they burn mint tokens. So it's like lock/unlock and mint and burn. If we set up ERC20 to ERC20 bridge, it can be lock/unlock and lock/unlock. It can be lock/unlock and mint/burn. It can be mint/burn and mint/burn.

Anna Rose: So you've just described two modes of bridges. You say that you'd have tokens on one chain and you want to bring them to another and you can use this thing called a bridge where you either lock them on one chain and you have them usable on this other chain. And then when you want to bring them back you lock them on this other chain and they become usable on the main chain again. The other one is this mint and burn and you mentioned Frederick that it's often two sets of smart contracts on either one of these chains. Is that is that the full definition of what a bridge is?

Igor Barinov: There are some additional pieces of this thing. Example if you have two smart contracts and you have validators between them you want to know. "Do you have like enough balances on this validator. Do you have the same balance of minted tokens as locked tokens. So that's why we created monitoring tool.

Anna Rose: Ah.

Igor Barinov: So that that's simply third component components right. So one is a set of smart contracts. Second is set of validators, bridge monitoring tool. Also we provide deployment tools in the form of ansible playbooks. So it's good to have the unified distribution for validators. That's that's for many complaints. And also the UI. Right. Which is important.

Fredrik Harrysson: Mm hmm. So I think conceptually you're right and I mean it's it's these sort of two smart contracts and authorities that Oracles or whatever you want to call them observe events on one network and make them happen on the other whether it's locking or minting or whatever. But of course as an actual product as a whole it needs a lot more built out than just that smart contracts.

Anna Rose: We used to talk about and I don't know how this is exactly related or if this is actually the same thing but is this it an atomic swap? Like what's an atomic swap and what's the bridge. Are these different things now.

Fredrik Harrysson: I would say so an atomic swap typically means that one thing is moving in one direction and another thing is moving in the other direction and you're guaranteed that both will happen or none will happen whereas here you're really just moving something in one direction.

Igor Barinov: But you have a claim like a bond.

Fredrik Harrysson: Yeah you can be guarantee that you can move it back later but not both are happening at the same time.

Anna Rose: So they're kind of. Are they in the same family or are they completely unrelated.

Fredrik Harrysson: I would say that's an atomic swap. It's a type of bridge because like if you want to bridge Bitcoin to Litecoin you use an atomic swap where you move Bitcoin on one side and you move the Litecoin on the other side and you're guaranteed that either both will happen or none will happen. And you do this through hashtime lock/unlock contracts contracts. But yeah I would say it's a type of bridge.

Anna Rose: What do you think Igor?

Igor Barinov: I agree. So it's better to think that atomics world is a part of bridge and not otherwise.

Anna Rose: I don't exactly get how do you move. How do you move litecoin onto that chain though. How do you move it in that even like you in that example without the freezing without the mint/burn.

Fredrik Harrysson: You don't move anything on the so you just move Bitcoin to another Bitcoin address and move Litecoin onto another Litecoin address. You're never actually like moving anything in between. I mean you're not doing that in the other bridge either but conceptually maybe you are. But like in a in an atomic swap it's even less so conceptually.

Anna Rose: Okay cool. That actually cleared up something. You kind of mentioned this before so the decentralized exchange kind of takes the role of moving tokens from one chain to another in a way. But these bridges like do they compete with exchanges? Do they replace the decentralized exchange? Do they work in tandem decentralized exchanges use bridges?

Igor Barinov: Umm, many questions.

Anna Rose: I don't know where to start.

Igor Barinov: Decentralized exchanges on Ethereum, it's it's a way for a token of a sidehcain to give liquidity on another network right. Also this bridge token can be listed on centralized exchange. And interestingly this bridge token can be listed in one pair with native token. So for POA there is a pair on one exchange for HitBTC where POA can be traded for a wrapped form of POA, called POA20. Yeah that's that's interesting. And then theoretically so do they need to they need this type of bridge just to operate. No. But can they get advantages? For sure yes, especially if DEX's will move out of Ethereum mainnet to let's sidechains or shards in the future then they can bridge existing tokens so they sidechains and get advantages of fast speeds and the scalable platform which is POA right now. So that for sure can be an advantage for our decentralized exchanges and for centralized too. I personally think that decentralized exchanges are more open to this type of innovation.

Fredrik Harrysson: You know I'd say the holy grail for a decentralized exchange is to be able to trade between major networks. And so right now on a decentralized exchange you you can trade wrapped bitcoin and then be guaranteed to kind of get your bitcoin out on the other side and that is actually then bridged. But this is pretty new. Like a year ago when everyone was working on bridges like the decentralized exchanges were just trading ERC20 tokens. And like only the native Ethereum one. Like that they didn't really bring on bringing other assets whereas the ideal goal for a decentralized exchange is to trade ZCash, Bitcoin everything on the same place. And for that you need bridges because these networks will never like abandon their own consensus model and like opt into someone else's. So you need to bridge these networks to be able to trade them properly decentralized.

Anna Rose: Austin do you have any thoughts on any of this?

Austin Griffith: So I just think that my way of explaining a bridge is is it's just an account that you're sending it to. So so we built Igor's Bridge right into the burner wallet. So it's a nice little native thing so someone can just hit like an exchange or, I think that button is exchange actually. But they basically just hit the exchange button and they can go. I want to go Dai to xDai and they just kind of hit it hit go and it does it for them right? So we're able to use his bridge without having to use his UI which is UI is great right. But I want it to be native within mine and and all that's happening is basically I'm sending Dai from my wallet to his his validator right. Or the other way around. So for me it's just like sending a transaction and a couple minutes later it lands on the other side. So that's that's how I think of him. Little Oracle magic in the middle.

Fredrik Harrysson: Yeah. So when you say exchange there. I mean you are exchanging Dai for xDai. So in that sense it is an exchange but it's a one to one exchange. You're not like speculatively trading in in that sense of an exchange. So it depends on how you define exchange.

Austin Griffith: Exactly and there's not a lot of speculation when it's pegged to the dollar.

Igor Barinov: When we launched our first bridge in production it was one of the first bridges on the market right and that there was a very interesting to understand is it legal right? Do we need to make KYC and AML on both sides can we make it this way that it will not constitute my transmission and will not be regulated? And those questions are still open but we we modified some things about this bridge that we think that it doesn't constitute money transmission and we have a legal opinion about it and to make it not my transmission, you have to think about are both networks centralized? right let's say they are not. And is this party which relay events centralized or not. So for all our bridges we have two or three or three or four signatures required to relay an event and these parties are independent from each other, non affiliated with let's say business activities. And also we don't call this exchange because the exchange is as you mentioned can imply these things. We call them self transfers. And that's interesting when you send funds over the bridge you can receive it another set onto your own wallet. So you can send it to the next destination only by having private key so it's like moving assets from from pocket to pocket because if Alice can send funds to Bob. More likely the bridge will be money transmitting service and we'll need to have KYC and AML on both sites like regular exchanges are doing. They have to have accounts, confirmations verifications, limit, suspicious activity reporting, compliant with bank secrecy local regulation and so forth. So it's very important aspect for interoperability and I think that that every project should think about it as like are you money transmitter or not when you transmit tokens between networks.

Anna Rose: I guess it helps to also define the difference between a decentralized exchange and a bridge which is that as you just described in a bridge you are actually locking a token in your wallet on one side and moving it to your wallet on another.

Fredrik Harrysson: Decentralized exchange in itself is only a smart contract on whatever network we're talking about like Ethereum and it doesn't ever leave Ethereum so you can't trade like anything that is on another network in a decentralized exchange without a bridge.

Anna Rose: What I just mentioned was more like who where like who owns the accounts that hold the things and what I understand is like in the bridging you own the accounts on both sides and so you're not really exchanging anything you're just moving your own funds around. Whereas in a decentralized exchange are you are you also moving it between two accounts that you own or not?

Fredrik Harrysson: You would move it to like you would move your Ether or your token X into the exchange into the contract you leave it in the hands of the contract Yeah and then you know someone else puts their stuff in and then you know you'll get theirs and they get yours. So I guess that's a money transfer but it's managed by the chain and by the contract so there's no entity to do the money transfer.

Igor Barinov: That's the problem here is when you promote this type of solution usually it comes from Twitter account or you know a website and they have some owners but it's a different topic. What what Austin mentioned that he sent funds to validators but actually he sends funds to the pool. And when he sent this amount to this pool which is like my contract generates an event and validators so listen to this event so they're not custodians of this funds so they can make some iterations with this funds within boundaries. Right. We added limits and quotas for validators which is important for a black swan type of event. And I think this is advantage of a pure implementation of interoperability and that is management functionality when one multisig which controls the bridge can control these smart contracts can define quotas and can define limits. What what may happen? Validators agree the oracles on different computers they don't have access to each other computer right but by social engineering or fishing they can be hacked. Right. And in this situation when we deploy a bridge we can specify what does a maximum damage breach can handle per 24 hours it's a daily quarter and daily limit per day. Also this limit can be set to zero. So validators will not be able to relay any events. This type of multisig can be operated by DAO and be operated by some token holders. You know the economic mechanism and these can be embedded within like Multisig which control the separations and there is no need to to modify smarts contracts of the bridge itself.

Fredrik Harrysson: All right. So we've talked and mentioned xDai a little bit and the Burner Wallet and a couple of these things let's dig into what those things are and I think useful just like let's start out with the xDai what is that is that a network of its own and how are those. You know I assume it's bridged in some way because you were talking about it. So let's let's dig into how that works.

Igor Barinov: xDai is a sidechain for Ethereum. It's a new chain based on Ethereum1.0 protocol with native currency in Dai. Dai is a stable token pegged to U.S. dollar and the market cap of xDai Chin consists only from bridged Dai. Bridge between Ethereum to xDai locks and unlocks Dai on on the Ethereum side and that mint and burn xDai one to one on xDai side. The biggest advantage of this type of network is that we have a stable native token first time in history and good think of having stable token is that balance of stable platforms execution is stable. Also if you think about Ethereum token for platform execution, you have two dimensional volatility. So gas prices are volatile and token price volatile right. In our first network in POA we set the gas price fixed so we remove one volatility platform but the token price is determined by open market so we cannot predict it. That's why this type of non-volatile platform usage we can have only in the network with stable native location.

Anna Rose: What you just described. Let's go. I want to understand the workings Exactly. Sure you have Dai which is ERC20, that lives on the mainnet. And you've now connected it to a sidechain or like a new network with this thing called xDai and xDai is not an ERC20.

Igor Barinov: That's right. It's it's a native token. Within Parity implementation of Ethereum, there is a special systems smart contract with block reward and you can define reward for each block. We change the implementation of this block reward. And now that this my contract is connected to to the bridge. So when an event is relayed an event is locking of Dai on it on mainnet side, now this block reward mints new xDai to a person who deposits Dai on mainnet side.

Anna Rose: I see and this is what you just described right. So on one side you have on the mainnet you have the locking but on the sidechain you have this minting and burning. So every time you lock on the mainnet you minte on the side chain.

Igor Barinov: You mint Native Native tokens.

Anna Rose: Native tokens and then you burn, meaning you...?

Igor Barinov: You send it to zero address which is like 0x0000 And unlock Dai on mainnet side.

Anna Rose: So who manages this sidechain. Because like so there's block rewards but there's no miners there.

Igor Barinov: There are no there are no miners. We use the same what we call proof of autonomy consensus where at the moment there are four violators and each of them has equal right in governance process and governance for us is very practical. Governance is defined by a validator set. So each validator can propose a change to add validator to the network or remove validator from the network. So in this model there are four validators at the moment it's Giveth, POA, MakerDAO and Protofire. which is a software development company and so four independent validators. The network itself started with one validator and it had what we call trusted ceremony when first validator starts with the network and generated first initial keys and distributed these keys to the first initial validators and after burn all unused keys. So in our setup we had the 12 initial keys as master of ceremonies distributed to three validators and after destroyed himself basicaly.

Anna Rose: Can you create new keys still if you wanted to?

Igor Barinov: No no there's this events are observable now governance DApps so anyone can. You don't need to trust me you can check the governance contracts and you can see events which happen to the network. So I was a master of ceremonies and I don't have an option to generate new keys and those keys which are generated in advance are destroyed. So there are four validators at the moment. Each of them can add or remove validator and three signatures are required to onboard new validator or kick out validator from the network. The same set of validators are validators of the bridge and validators of mutlisig which manage funds of the of the bridge.

Anna Rose: So now we have a sidechain with a native token that stable.

Igor Barinov: That's right. It's very exciting.

Fredrik Harrysson: There's obviously a lot of I don't know if you want to call it problems but obviously it's a lot better than being completely centralized with one entity as in our normal like Coinbase centralized exchange. So you have four parties that are independent but obviously it's not as decentralized as the theorem itself. So like that's a thing that smart people say is decentralization is a scale it's not a switch that's on or off. But yeah just crypto economically obviously this validators have no incentive to add other validators they would be doing so completely altruistically I assume. And if there is I didn't actually know they had fixed gas price because that also means like they're like you have no way to deal with congestion for instance.

Igor Barinov: Yeah it's it's an interesting idea and we're in the production for quite a long time. Right. And we know that the models scalable horizontally for example you can create a new sidechain and like xDai and some projects like Colu already created a sidechain like xDai but with their own token as a native token and the setup a bridge between Ethereum and their network. It's not a stable sidechain but it's application specific sidechain. We know that the model is scalable with a same security assumptions and we know that contacts are audits and the consensus is audited, Ethereum 1.0 is battle tested and peer reviewed. So we know that we have the security assumptions and can scale the network horizontally. Also the network is quite easily scaled vertically you will have professional validators they're incentivized to get as much transactions as possible in the network. I think valid transactions. And if we think about incentive mechanisms for validators to validate blocks, there is no block reward for them or they can collect their transaction fees which are small and basically introduced to protect from denial of service attacks. Their motivation is that in future will be a small percentage of fee when you want to convert xDai back to Dai. So from Dai to xDai is free. But if you want to go back to let's say convert Dai to the U.S. dollar or other tokens then for the exit you have to pay a small fee. And we don't set it up but it would be very small and comparing to other services. So that's the economic model. And as we know everything is temporal right. And right now say so temporal solutions for scalability. It started as a temporary solution for scalability. But now we see that sidechains plus bridges plus apps as Austen built are enablers I call them enablers because they enable new use cases which we didn't have before and relatively inexpensive. So if you combine all expenses were held on xDai and it's it's much less than average ICO and time frame wise it started on ETHBerlin and then like less than half a year and so many people around excited about it out Burner Wallet, xDai Stable Chain, the idea itself that specific sidechain. So that's amazing that we can make it this enabling technologies by combining existing tools that we have already. And that's why I personally in Ethereum ecosystem and that's why I'm staying here and not moving or to other ecosystems.

Anna Rose: You're not going to Tron and anytime soon?

Fredrik Harrysson: I think it's a cool use case especially the xDai thing. It's obviously trading off scalability for security but for a lot of ups that's totally fine. And that's sort of the tradeoff that you have to make in the Triangle of security, scalability, and decentralization.

Igor Barinov: As you can see on our forum to MakerDAO engineers asked if we can make 1500 transactions per second on an xDai chain clone. And there are multiple parties interested in experimenting with this type of bigger xDai chain type of network.

Anna Rose: I think this is definitely given us a good look at what xDai is. Now let's move on to this Burnet wallet. Austin what's the burner wallet?

Austin Griffith: So the Burner Wallet it is basically a public private keeper that lives in your browser, so kind of like the tradeoffs like Frederick said how there is kind of a gradient of decentralization we're willing to make these tradeoffs to go into xDai to move stuff around quickly. Same thing with the burner wallet it kind of takes it another step where first of all if you want to get into crypto you've got to like download a wallet and you're the first thing you're gonna get hit with is they're going to ask you for its 12 word seed phrase and they're gonna make you tape it back in. And that's a really good idea if you're going to have a lot of money right. But let's let's make some tradeoffs to get a UX smoother and the burner while basically just creates you an account as you hit it and it's just a web page and your phone then you can use that key pair to transact on xDai or even back on the mainchain its just little slower. So so with the xDai I know we've talked about it but let's just a TLDR of xDai basically you've got five second block times right. So your transaction is going to go super fast. The native token is Dai so low cognitive overhead you don't have to think about what is .001, like 0.001 of a penny.

Anna Rose: Though it's not Dai it's related. The native token is xDai.

Austin Griffith: But you could quickly bridge in and out real time you can move in and out. Right. So if I if I have 10 xDai I can have 10Dai in two minutes right. So so we've got a quick block times we've got low cognitive overhead and then we just super cheap right. So the gas price to deploy contract is like 0.00004 or something like that. Right. So it's a super cheap to deploy the contracts. So then on top of all of those benefits that you have we've got a burner wallet which is just a web wallet that comes up in your browser and immediately lets you start moving money around in five seconds. So that's basically the concept of the Burner.

Anna Rose: I know that like I think it was that Igor demoed the Burner Wallet to me and it was one of the funnest demos I've ever seen in the space because it worked really fast and they didn't have to download anything right.

Austin Griffith: I think we're ready for these trade offs right. Like the UX is is finally like we're finally ready for this kind of UX. I mean like a junkie little HTML version basically on the plane to Prague for DevCon. And when I got there I showed a few people it's like hey look I can send I can send you this new this hard spoon, xDai came out and I can send you some of the currency and then I send it to a few people and they're all like "Whoa that's super cool" and immediately I was like oh wait a minute. Yeah this is great.

Anna Rose: It's like "Oh my God it works like all of the apps that we've known for all of these years. And there's not huge latency and weird things we have to write down on pieces of paper and hide them away".

Austin Griffith: But you need be afraid of that. Right. Basically that that seed phrase or that private key that protects this money is stored in local storage on your phone. And so we had to come up with a name that was very scary and someone on Twitter was like call it a Burner and you know you kind of think about the old Baltimore cop show but it's a really good example of like what like the name really define what's going on that right. Don't put a bunch of money into a burner.

Fredrik Harrysson: That's a good point. I mean I think that's like MyEtherWallet kind of started at the same place where is this like we're going to have everything in a browser is going to be super fast and you like you can just send tokens and then all of a sudden someone is sending like five million dollars on MyEtherWallet with all of their stuff in their browser. And are you Are you still scared that like people will start using this with significant amounts and suddenly you know xDai is worth 40 billion dollars and it's very you know plausible that someone will try to attack it.

Austin Griffith: Yeah yeah it's I mean it's scary. We've got to we've got to provide the education around that right. So first of all I think I own I think I own like maybe a fourth of the entire market cap of xDai and we've been joking around since it will be used at Denver for buying hotdogs and stuff that I'm a self-proclaimed hot dog whale but but what what we what we need to do basically is people are buying into xDai. We need to be careful to do good on the education side. Right. So. So as someone uses the wallet basically well we'll get you in. We'll let you use it. We're not going to hit you with a seed phrase but we can tell when you have more than 30 or 40 dollars behind that and at that point we need to provide that education. We need to slide down the saint. The thing that says hey let's let's learn about seat phrases now you've got enough money in here. And then when it gets to three hundred or four hundred dollars and it's even even scarier. Right. So the education is going to be very important on our part to kind of ease people into that right. We're kind of at the top of the funnel when we talk about just like we've got a web page that you're going to move funds and then it's it's going to be up to us to provide the education to move people through that funnel to get them onto a better wallet with better security and and probably even a better network. But I wouldn't say that in front of Igor, xDai is great.

Fredrik Harrysson: Yeah I mean that that's what I think individual user education is super important obviously. But even if you do that well and let's say no one has more than 10 bucks in it if there's you know a million people with 10 bucks in it that's the network is worth 10 million dollars it might not be a problem if there's 100 million people who have 10 bucks in it you know suddenly bribing those four validators might not actually be that hard. Like at what point will Giveth, will a collude with the others for a Billion dollars? maybe.

Austin Griffith: Igor do you want to answer that?

Igor Barinov: Yes so we had, we will see more change like xDai and each chain will have some different set of validators, different entities different countries and they will be aggregated in some things like you, let's ask Austin, so let's say if we have two networks like xDai is it possible to connect both of them in Burner Wallet or query the process of two RPCs.

Austin Griffith: Oh absolutely yeah. So the burner is basically made for like plug and play RPCs. So so yeah we can bring in 721s from mainnet, I'm trying to build it like that so. So there's like a config file that gets plugged into it where you can pick the RPC, you pick the tokens you can pick the balances and just kind of have them all there.

Igor Barinov: Yeah I mean if you have two networks like xDai that can be connected to one Burner Wallet and let's say your query balances on two networks and the one network is the default network work of more funds it's all much easier to make on like on wallet right then on the network side how to integrate charts and how to ultimate answer.

Anna Rose: Are you saying that like as soon as xDai side chain one has like maybe 10000 xDai tokens in it you switch to a new one which has different validators because that was something you could do? I mean that's a very high number of very small volume side yet?

Igor Barinov: That that's that's what we discussed when we thought about enabling. Recently we thought about a way where like locked Dai can be used as as a loan on a platform like Compound or you know MakerDAO the platform itself and get some revenue from borrowing this locked Dai. But let's say we don't want to to test it on the on production because that first it's launched one more xDai Chain and make incentives for merchants to sell something on xDai because when you sell something for xDai you can you can set a lower price because this locked day will be loaned to someone and revenue from from this from this loan will go back to the network and will incentivize merchants to decrease price. Right. So you can basically you can say yeah here is one product for one U.S. dollar or four 0.97 xDai and this type of experiment most likely will be on that on a new sidechain. I have a.. I know that some people want to have xDai but for a different table tokens because you know everyone is talking about xDai but no one is talking about xTether right. I'm not I'm not saying that Tether is interested in it but you know other stable tokens might be interested in this type of smart contract thats powered by their native token and by Burner Wallet. So that's that's that's also field for having multiple payment systems connected by one UI like Burner Wallet. And as Austin often mentioned that's quite easy to integrate multiple payment networks. So I think this risk will spread and decentralized across multiple chains. But that's also beauty of sidechains, becuase you can have multiple chains with the same security assumptions.

Fredrik Harrysson: I was curious on the Burner Wallet what's the backend for it. So you mean like for instance integrating multiple RPCs. Is it Infura backed or are you running your own servers with nodes or what are you using.

Austin Griffith: Yeah. Depends on the RPC end point. So for mainnet we do talk to Infura we we can connect to our own nodes or we can connect to https://dai.poa.network. It's all about you know who who has the best uptime for us right. So the burner is just a static web site you can download it from the github repo and run it. So the backend is basically all these RPCs and chains. So to talk a little bit more like this is this is a giant experiment right. Like xDai is just a hard spoon of Ethereum it's an exact copy of Ethereum so to speak to that a little bit more you can do Constantinople type stuff on xDai now. Right. So you can do create2. Igor and team deployed Constantinople out ahead of everybody else. And we could play around with it on xDai where it where it's a little less risky. Right. So you can do a create2 counterfactual deployment on xDai right now and that's cool. But what else. Right. So when we start talking about ETH2.0 there's there's gonna it's gonna be really awesome to have these little playgrounds to to dive into a sidechain where the whole market cap is 5000 dollars and we can play around with a lot of these different techniques right. And along with that the the the frontend is a giant experiment right. We can play around with let's let's make sure that people will understand the UX and let's get it translated to different languages and let's also maybe create a sidechain that's pegged to a currency in a solid South American country maybe not a hyper inflated South American country but this is all just an experiment to see how we get people to use it and how well it works and how we kind of drive adoption and usage and all of that.

Fredrik Harrysson: This kind of goes back a little bit to an episode we've talked about where we're talking about testnets and we talk quite a lot about value bearing testnets and that being a thing that that's kind of nice and something that you you you want because you can't really test something on a testnet unless serves value like no there's no incentive to break anything you know if you're building a DAO on a testnet that you want to try that out you kind of want a value in that DAO to ensure that it works as like you can test script economic incentives or anything like that without value. So it's almost a little bit like a value bearing testnet. And I think it allows for a lot of experimentation like you guys said which is good.

Anna Rose: I think this idea too like the Burner Wallet as I mentioned like when I saw it I was just like wow this is so easy this is so great. You've sort of like fixed a few a few problems in that small build. I know it's I mean not new alone but a few problems have been fixed by this which is a stable coin which you can actually buy kind of real world things from potentially like real world venders who will accept it because it's gonna be consistent. And like you mentioned you've gotten rid of that problem of meeting when you're using a stable coin like Dai needing Eth to send it around. And I have to say like I had that I had that experience firsthand when I was in the blockchain week in San Francisco and I went to a MakerDAO drink thing and they were giving away Dai and I was like cool Dai but I didn't have anything set upside to setup the far you know set up the wallet put the Dai on the thing and then I realized I got there was like okay cool I got it and I went to pay for something with this Dai and then they're like Oh no no you also need Eth. So then I had to go to somebody else and then they had to send some Eth so that I could do this thing and I felt that clunkyness. And so obviously when I saw the xDai in a browser I completely felt like Wow that that is smooth. And you've also added in at least the version I saw the ability not only to work in xDai but also move it into Dai and even generate gas or something like you've you've been really thinking how to, its like awesome playground small amounts and then bring it back to the mainnet using bridges.

Austin Griffith: So yeah exactly. Well it kind of a lot of this started I think it was I think I was in San Francisco. I don't remember where I was exactly but I was at a dinner with plenty of Ethereum wise people. Right. Well people probably have plenty of Ethereum I'm all sitting around at the table and it was time to split the check and and someone was like "Oh we'll Venmo you" and I was likd "No this is Ethereum people what do you mean your Venmo me." And so that's where a lot of this started. And then talking with Ollie Hondros who is working in a lot of Venezuelan countries like it all kind of came together like oh we need to make this like really really really easy to use. It's all about the user experience right. So once it once we had it in the browser and it was working we just started holding all these events there Cypher Punk Speakeasies. But basically you have just a phone or an iPad or any kind of device and you create this pop up point of sale system where we just walked into this bar in Boulder and we took their iPad that they had sitting there and we pointed at xdai.io and we hit the received button and they had a PoS system. And then at the door we would hand out these paper wallets that were loaded up with some tokens that they just basically you shoot the tokens with your phone. And now you have ten dollars and you walk over to the bar and you scan the beer that you want and it shows up on their screen it shows up on your screen and you get a beer in hand. So we just rounded off all the sharp edges along the way and we did this over and over again where someone would say "oh I wish I had you know a max button here. I wish this you know I wish this form would save the information for me. I typed in the number and then I hit scan and when I came back the number wasn't there." Right. That's that's been like my last month has been just watching people use this thing and writing down every like scribbling all sorts of notes and just watching people use it to try to make it as smooth as possible.

Fredrik Harrysson: That's a great great story in user studies and like actually UX development.

Anna Rose: Frederick I love that link that you just made though to the value bearing testnet because the speed at which something like this can be used adopted like it's amazing. But like you said like if it gets too big then there's a security issue. But if you can keep it actually small then you can just have a shit ton of activity where there's no reason to like break it in a sort of a bit of a safe place where all of this user testing could potentially be done. And then when other solutions come in that can actually protect larger volume sidechain stable things then maybe the UX will be there because I know that's been a huge complaint we've done a few episodes on like the issues of users not new user experience.

Austin Griffith: And being able to bridge in and out of these sandboxes in a minute is this super powerful maybe.

Anna Rose: Do you think these site chains could replace the test net.

Igor Barinov: I think they can be like a staging environment. In most software development scenarios, usually we'll have a stageing network, test network and production network. In crypto space, we didn't have this stageing environment and I don't think this site is will be a replacement so but they can be used to just cryptic comic behavior and insensitive which is hard to testnet. Also I'd like to add about new features in burner wallet with you. Actually you asked Austin about this feature and it's like I'd like to explain it if possible right. When you get your xDai, right usually you want to convert that today. Right. Because Dai is stable and on exchanges and Dai is more stable than xDai so to speak. And when you convert xDai into Dai or then Burner Wallet like a week ago you're locked into a Burner Wallet because you don't have Ether to send Dai back. And now there is a feature where you can sell part off your xDai and get some Ethereum. So theoretically Austin, can we send Ethereum from Burner Waller right now?

Austin Griffith: Sure yeah. So what he's explaining is basically the exchange and it goes from Eth to Dai to xDai. And then if you run a native token on xDai this is a good way to explain. We keep saying that xDai is a token but what do you consider Ethereum a token or is it like a native currency.

Fredrik Harrysson: Ether is a token yeah.

Austin Griffith: Okay. So basically I think of Dai as a token that runs on Eth. And so what we've done is we've deployed a token for each Denver called buffiDai and that runs on xDai. So if you think about the conversion process you've got Eth and then that goes into Dai and then Dai goes into xDai and then xDai goes into buffiDai. Right. And you have all those different bridges and rule sets that can be tweaked depending on how you want to run those.

Anna Rose: And actually to the user they're gonna go the other direction right. They're gonna start with the what do you call it buffered Dai?

Austin Griffith: Like so buffer corn is the is the mascot of Denver. So buffiDai.

Anna Rose: I keep hearing "Buffy Dai".

Austin Griffith: BuffiDai yeah. I don't know how to how to pronounce it

Anna Rose: Vampire slayer.

Austin Griffith: Well yeah.

Anna Rose: But uh I think I'm hearing it wrong. Okay so they're gonna hear they're gonna get it from the other direction. So I still can't pronounce this buffiDai. But you know Yeah. buffDai died to xDai. xDai to Dai and then Dai Dai, No wait. Actually no, but you're going. You're ERC20 to native, native to ERC020, ERC20 to native. That's.

Austin Griffith: Exactly right. Yeah. And anyway and it all happens natively right in the Burner wallet right. So you can go from buffiDai day to Ether by just hitting buttons right in the UI within the Burner. So why you ask. So let's say what we want is to stimulate an economy at ETHDenver right. We want this pop up economy where we can have these food trucks that can be accepting buffiDai a token that we just made up and then that buffiDai to have value. But we don't want participants to just take their 30 bucks or whatever they get. And just off ramp it and walk away. So what we do is we have a special contract that says the token that we've minted to all these users can only be off ramp by the vendors. Anyone can can on ramp some into it and they can offer rent the amount that they on rent or whatever they can get out what they put out right. They can withdraw what they deposited but the main the main balance that we minted them can only be taken out by the vendors. Right. And we can just do that with a simple few simple lines and a smart contract. So what we what we've done is we've minted this token that only really has value to the vendors. They let people buy those things with the tokens. We don't have to have a bunch of money wrapped up in it because we only have to off ramp what the vendors off ramp. And that's what we pay them for. So it allows us a lot of flexibility with the tokens and we did the same thing with the speakeasy. You walk in the speakeasy you get 10 dollars and it shows up as 10 dollars right it says dollars right there in your phone. But really it's just a token I made. And at the end of the night after everybody is paid then I can just go to the bar and look at the balance on the iPad and an I off ramp for them to fire from there is really cool.

Anna Rose: It sounds like there's a lot going on around next day and around the burner wallet. How about you guys just give us a little bit of an update about what's coming up.

Austin Griffith: Okay. So basically eat Denver is going to be the trial by fire for the burner wallet no matter how many little 30 person events I have when three thousand people all show up and start using this thing. I keep saying on Twitter that it's going to go up in fire because I keep using the fire icon for the burner while hoping that like going up in flames is a good thing. But it could be a total disaster. Hopefully not. But we'll see. So after ETH Denver basically there's a lot of really cool things on the roadmap for the burner because what we have is sort of this this easy to get to web 3 provider. And so from there we we've been thinking about ways that we could use this for other onboarding right not just a wallet but for dabs if ADAP can have a wallet that's integrated into their page. So we get going we always have this kind of gas Bitcoin unicorn project that's out on the out on the horizon. Well if we deploy wallet does get coined CO and it's a burner wallet and then we are frame that in and inject it then basically you could have gas less experience where you go to get coin. It detects that you don't have web three a little burner while it slides down. You dropped five cents into it or something like that and you're good to run for a while and you can instead of just having like every single pop up to sign things since it's only a nickel you can kind of just put it on autopilot and let it go. So going back to like those games that I built about a year ago where you were constantly hitting that dialogue if I build a wallet into that or the other way not just an iFrame but if it's like a sign in with burner and we use post messages to turn to talk between the burner and the app basically you kind of create this metaMask that lives in the browser and then you can kind of just send it two dollars and you could go play my game and just put it on autopilot and now you're just clicking buttons and building ships and you're you're sailing ships and you're going fishing and all of that stuff that was kind of impossible last year because it was kind of clunky now becomes possible with this injected web3 that comes from the burner wallet.

Anna Rose: And collectibles.

Austin Griffith: And collectibles! Yeah. Where we're going to do collectibles at ETHDenver and we'll see how that goes. But yeah that since since xDai is kind of a sandbox for Ethereum and and the burner is kind of a sandbox for a wallet it lets us iterate really quickly like we were able to deploy collectibles like I started working on it two days ago and it's live in production now. And to be able to do that on a wallet would be really hard if I'd like ship that to an app store and stuff. Cool.

Anna Rose: What about you Igor?

Igor Barinov: For the for the bridges, we're working on the incentive model how to take fees when funds are moving from from networks and networks. And this is important to keep validators incentivized. The next big feature is arbitrary message bridging where the arbitrary message can be based between networks and that's important for games and applications with metadata there are multiple requests for this type of functionality. And we are working on it right now. And for xDai the most important is to have this seamless experience as long as possible. We make our infrastructure more out of scalable parts that we are we're providing for example RPC endpoints are our scalable and we print the battle test them is Denver. And what's interesting work we're we plan to do with bridge for xDai is we plan to implement some form offer zkERC20 between when you can standard. So you can deposit a node on the Ethereum side and you will get anonymous xDai on our side. So this will provide some plausible deniability to this Burner Wallet transactions and especially if you burn your private key after you use it. So a user can deposit on Ethereum side and get this anonymous xDai, use it, burn it. And from from Burner Wallets, there is literally no change is required. Right. Because it's the same xDai but with different source of origin. So that's that's that's our plan for bringing private transactions. That's cool because you can have private transactions without actually implementing any zero knowledge on xDai side because it will be on their site.

Austin Griffith: Yeah it's really exciting. I was likely to ask tech guys and I was trying to think of how we could get Aztec into the Burner Wallet. And I've been thinking "man we're going to have to deploy Aztec over to xDai" and then Igor comes along and is like "let's just bridge them" and so that there's we can really kind of experiment out on the edge with what we could do with this stuff but then kind of pull it back and look at the other side at the top of the funnel is how we how we educate new users right now that we have lots of users coming in and they can just open up a web browser and have a wallet that's another thing that we need to be really pushing the envelope on is educating them and getting them into a better wallet. Right. Like after you've used the burner for a little bit you know now it's OK to give them a download for a better app and now they're OK. You know it's not it's not you adapt to our just trying to get your beer you're actually trying to protect your funds. And now it's time to slow down and learn seed phrases and download apps and I think that's something that we would really want to push with the burner while it is just that education play and kind of better on ramping for people into you know initiated crypto.

Anna Rose: I really like that you're introducing it that way because I think that the concern and how likely it is that people will just like accidentally delete their cash like I hope that that comes through constantly that this is playground that this is like play a game money what it's like to be taken too seriously. But I do think that the ease of use will make it popular.

Austin Griffith: Thank you.

Anna Rose: Cool. Well listen guys I want to say thank you so much for Yes on the podcast and taking us through this journey from proof of authority to proof of autonomy to bridges to xDai to burner wallets. Its a long journey but it's really really exciting what you guys are working on. Sounds like there's a lot of potential in it.

Austin Griffith: Thank you so much for having us on. This is great.

Anna Rose: And good luck at ETHDenver. That seems to be where it's happening.

Austin Griffith: It's gonna go up in flames.

Fredrik Harrysson: All right. So thanks again. And to our listeners. Thanks for listening. Thanks for listening.