Epicenter - Learn about Crypto, Blockchain, Ethereum, Bitcoin and Distributed Technologies - Jordi Baylina: Hermez – ZK Technology for Scaling Ethereum

Episode Date: September 21, 2021

Hermez is an open-source ZK-Rollup optimised for secure, low-cost and usable token transfers on the wings of Ethereum. The Hermez network team is currently developing the next version of the decentralised ZK-Rollup with full support of Ethereum Virtual Machine (EVM) by making a full set of zkEVM opcodes.Jordi Baylina, the Technical Lead at Hermez, joined us to chat about how the protocol works, the recent merge with Polygon, and the roadmap leading to Eth2.0.Topics covered in this episode:Jordi's involvement in the infamous DAO hackThe Giveth community he created with Griff GreenHow Hermez use EVM opcodes to combine zk and EVMHow the protocol works as a userHow do token transfers work?Currently the advantage is in scaling, could this also translate to privacy?Hermez's merge with Polygon - how the joint model will lookThe Hermez roadmap and how it fits with Eth2.0Jordi's views on the future of blockchainEpisode links:HermezGivethJordi at EThCC 2021Hermez on TwitterJordi on TwitterSponsors:CowSwap: CowSwap is a Meta-Dex Aggregator built by Gnosis. It taps into all on-chain liquidity offering the best prices on all trades and provides some UX perks (no gas costs for failed transactions!) and protects traders against MEV - https://epicenter.rocks/cowswapThis episode is hosted by Friederike Ernst. Show notes and listening options: epicenter.tv/410

This is Epicenter, Episode 410, with guest, Jodi Baileena. Welcome to Epicenter, the podcast where we interview crypto founders, builders, and I'm Frederica Ernst, and I'm speaking with Yordi Baerina today, who is the founder of Hermes, among many other things. We will get into that in a bit. Before we talk to Yordi, I'd like to tell you about our sponsor this week. So, Dexas are great, but they're vulnerable to problems like MEV, transactions and high gas costs. Cow Swap tackles these issues head-on and offers a new kind of trading experience.

Built by Nosis, CowSwap is a Metadex aggregator. That's right, it's a Dex aggregator aggregator. It fights M-EV by matching overlapping orders directly. If no coincidences of once is found, that's where the cow comes from, trades are settled on a variety of underlying on-chain AMMs, depending on which pool offers the best price. And best of all, no gas on failed transactions. Volumes have gone up significantly since its launch in June. So in June, cowswops saw trading volumes of 28 million, and that's gone up to 321 million in August

with over 12.5,000 traders. There's also a former proposal now on the NOSUSA DAO forum regarding a token for NOSO protocol and a revamped user interface with a new wallet. Activity History panel, better USD price estimations and warnings about trade actions has also gone live. So start trading on cowswap.exchange or learn more in the CowSwap Discord. So, Yudi, it's super lovely to have you on. I mean, you've been in the ecosystem really long time. I think the first time I became aware of your existence was in the context of the Dow hack. So if memory serves correctly, you were part of the.

collective that brain the Dow, right? So tell us how this came about. Well, this is a little, this was a little bit my batism in blockchain, at least doing real things. And it happened that, you know, I went absolutely deep in Ethereum. For me, it was like a drug, just, you know, studying all day long and just understanding the, the, the goodness of decentralization and decentralization protocols and all that. I just arrived randomly to the Tao project. The Tao, you know, at that time was like the trending project at that time and was very cool and people was very excited about the Tao. And for me, I learned solidarity from the Tao. I just created some look at democracy, smart contracts for

the Tao and it was really excited for me was just like I already discovered that this was like a new world a new way of understanding the the world in the more decentralized manner and I became like part of the Tao you know just like another another community member there and just very excited like everybody there and learning a lot from many people technical and I would say from one from the community people But for me was a very incredible project. But then the hack happens. And at that time, I knew a lot about the smart contract of the Dow because, you know, I used that smart contract to learn.

Well, I just realized how to what happened at a little bit with Adao. I was able to reproduce that in the TestNet. So I contact the sloppy people at that time. and from there, they just became what we called there, a Robin Hood group just trying to save the Tao at that point. It was like the last guy to arrive there. But for me, but I changed my life, absolutely. You know, since there, I was just first happening there,

and then I just quit my last job in not CryptoWorld. And since there, I've been 100% in this space. and yeah just change absolutely my life my life has two sides one before the Tao and the other is after that before now and after the Tao

it's just who I divide my life in some way and yeah I'm been I'm there since the last five years and very excited and very excited and I'm so proud to be here yeah super interesting what

what a moment to enter the ecosystem system, right? I mean, it made the mainstream media and everything. And basically back then, it just, there were so, I mean, there weren't a lot of people around, right? Well, there were probably different people, maybe not that many. For me, for me, it was a lot of people at that point even. But, you know, of course, no, we changed it. It changed a lot and many, many new people entering the space. But, you know, just, uh, I remember at the beginning, you know, in these DAO hacks, I just, I started because I didn't knew anybody.

I was just in my home, in my laptop, just talking with chance. And I remember the first time that I meet physically with other hackers, just with that out. And it was very, very, very exciting. It was very exciting, exciting moment. I cannot forget. And yes. and it's interesting, because it's something that's very technical,

and you start getting in because of technical curiosity, but then you discover the community. You discover that it's real people, real, very smart people and very committed people and people that really believe in what's doing. And when you discover this world, you just get engaged. you just get here and it's super exciting. You then went on to found Giveth, right, with Griff? Yeah, exactly, yes.

With Griff, it also was helping to recover all that. Actually, it was the guy that was just talking with everybody. I was just behind the scenes, just trying to help there. And with him, we got very friends there. and then we started the project, the Giveth project. Actually, we started together with Gryff,

but I would say that's more the Grif project that minds. It's mine too because we started together, but it's, Giveth is the dream of, the Griff dreams. And, well, I don't, I'm sure you know, Grif, but, you know, Grif is very idealistic and he loves,

he believes, he has a vision of, a kind of a world and Gibbeth is like a piece, a very important piece to build this decentralized this decentralized world and

it was and I would say it is because Griffith still is trying to build in that. He didn't give up and he's also well he's working there and I'm sure he will get it what he's achieving. It's very interesting.

Just to give some background, so Giveth is a donations platform, right? So it's a donation platform, but it's more community. I would say it's a very, a way of a way of solving the communities, you know, how the community is organized and how the community can be funded and how the community can start doing things with things with less friction that, the current system has. And this is about Gibbeth organizing communities,

funding communities. And he's also, for example, just working in this, his Tao Stack project. And it's all about this. It's just about ways of funding, ways of funding people that just want to do good things.

And this is what's Gibbeth about. Cool. Yeah. No, it's definitely super interesting. project to check it out. So let's talk about what you've been up to since. Well, I can say that since there from that time, well, I was doing at the beginning, I was doing a lot of audits and all the ICO boom at that point and was still learning

and continue learning. I didn't stop learning since there. It's just a new world and, you know, you have to study a lot here just to understand what's going on. But there was, at that point, there was a technology that I already realized that was very exciting. It's all these zero knowledge technology. So since there, I was at the beginning just studying, studying a lot of this technology. Then I was building some tools for this technology. We created the project ID3. IDEO3 is a project for a servering identity. We're still working on that project. It's a protocol for Cilson and Iredity. That's where zero knowledge is the

key piece for, you know, for privacy and for recovery and for everything. And with this technology, we just started the roll-ups because it's cool because the zero-nology technology. It's the same technology that can be used for privacy, but can also be used for scalability, for scaling. It's strange, but it's a technology that allows these two things. So then we started the Hermes project. Hermes project is a roll-up, layer 2 ZK roll-up for doing payments mainly. Both launch last March is actually working right now in Mainnet. Now I'm working in the ZKVM.

That's mainly layer 2 ZK roll-up where you can. deploy smart contracts on top of that. And this is what I'm fully committed right now and what I'm working hard right now and what I'm very excited to work on. And yes, we are doing a lot of progress. And yeah, I hope that we can see it very soon. Well, yeah.

So let's talk about Hermes. So as you said, Hamas is a ZK roll-up built on Ethereum. So obviously, Ethereum has a pre-existing EVM, which is not optimized for ZK technology. So basically, starting from that, there are at least two options, how to get the zero knowledge part and the EVM part together. So either you make the zero knowledge part solidity compatible or you build a full set of EVM upcodes. And you have, you guys have decided to go the second path, right? So from where I stand, this comes with very obvious advantages. So basically you inherit the full compatibility with the tooling and the ecosystem.

You keep the Ethereum security model. But I assume it also comes with a lot of challenges, right? Yes, absolutely. You said before, it's the current EBM in Ethereum. It was not designed it. I don't think nobody realized that. that this could be run in a ZK-friendly environment. So, yes, it's not designed for that.

We are doing a kind of interpreter. So what we are doing is, okay, we are just, this is not compatible, but we are trying to make it compatible. So that's a little bit what we are working. This, theoretically, this should make, so should be a little bit, not optimal, but when you check at it, it's more optimal that what you may think at the beginning. So if you think deeper, if you think deeper in all the problems you have and if you're doing in the right way,

you see that you can create a very, you know, this specific micro-up course, you know, specific, you have the freedom to design the ZKBM that's the base EVM that you want. So you can design this EVM in a way that's very optimal for interpreting the Ethereum IBM. So it's like, let me just put an example. I want to imagine that you want to do an interpreter of Python. okay and you want to do it in C. But it's not, it doesn't fit.

It's not the right way. But imagine, but imagine that you don't even, that the C is not even invented. So you can invent your own language there. That's, it's a language that needs to run a single program that it's only be, needs to, that the only thing that needs to, the only thing that the problem needs to do is to interpret Python.

So you can write a very optimal, language and compiler for C to be very optimal to just compile a Python interpreter. That's a little bit the example, but it's not that, so it's not that unoptimal as it may looks like. Of course it is, but it's not as much and it's doable. And of course, this brings a lot of advantages from compatibility. You just go just upload to code. You just compile the program in solility and you get the same output and you just run it there here instead of there. It's like a little bit, for example, really the optimistic roll-ups like Arbitrum or optimistic little bit the way they are doing, you know, they are not, maybe they change

something in the solidity, but it's more because, not because they are changing the ADM is because they are, it's not a layer, it's not a layer one, it's a layer two. So there are maybe some of course that makes no sense in layer two. But it's more technicality is that what it is. So a developer, ideally should not care if he's doing a smart contract for a layer one and for a layer two. So you're basically, you're taking upcode by opcode, and you're kind of transposing them into the corresponding ZK opcode.

So basically, if I wanted to have like a five upcode string, A, B, C, D, C, D, E, I would go ZKA, ZK, B, ZK, C, ZK, C, ZK, E. And I would get the same result. Is that kind of roughly correct? Yeah, it's actually, as you are running, you have a program, sorry, the program. When you compile it, you get a set of bytes, you set of bytes that are upcodes mainly. So you just take those bytes and instead of running, here you just run there. It's like at the end it's from the user perspective is like having a side chain. It's not a side chain, but the same way that you are using a polygon or XDI or any other side chain,

that you are just signing transactions and deploying smart contracts in the side chain, it should work the same way with ZKBM. The only thing is not going to be a side chain. It's going to be a layer two. It's going to be a roll-up. But from the user perspective, it should not be many differences. Okay. And in principle, can you transpose any op code into a ZK op code? Yes. Theoretically, yes. There are some up codes that are harder than others, but we have been analyzing.

And I would say, yes, there are some, you know, as I told you before, there are some exceptions that are more, for example, you know, if there is an op code that's difficult. How much is difficulty is. Okay, in layer two, you don't have difficulty or anything related. So this upcode probably will exist, but maybe we'll return some hard-coded information. Okay, or base code. Who's the minor? Okay, you don't have minors, but maybe you have a coordinator. So maybe this op code returns a coordinator instead of the minor.

Okay, so there are some tricky opcodes that are very related. to the layer one, but most of the old codes will be the same and they should not be many problems to implement it. Here is another thing is a cost, you know, maybe there are some out codes. So the gas cost in the layer one is one. And then the cost in the ZK roll up, the specific cost for running that in the roll up, maybe a little bit different. So here may be some adjustments that would need to be done, but they is to keep exactly the same gas model that layer one. And the worst thing, maybe it's

going to be some transaction that would not be able to run in the ZK overlap. For example, if you are abusing, one of these of course that's very expensive in the ZKROLUL app, well, in the the Zika Blah, maybe it's going to be a limit of, for example, exponitiation. Exponitiation, maybe it's a hard of code. Maybe you can only do like three explanations or four explanations, and if you are doing more,

these transactions would just be impossible to mind, so it would be an invalid transaction. That's the thing that may happen. But if it's executed, it would be executed the same way that in layer one. Okay. So basically,

you have like this library of say 140 or 140 minus the couple that make no sense ZK opcodes and then I mean what we use on layer one lot is these pre-compiles right I mean there aren't as many I think there's

like 10 or something but do you have equivalence for those as well yeah theoretically yes but these are hard and these are I would say this would be expensive you know a no code there are that are expensive but pre-compiles is like another upcode it's not really an

upcode but can be in the other code but they could be very very very expensive for example you have paintings so theoretically you should be able to run a so to each cut a painting inside the rub it's doable so technically it's doable but the cost it would be quite expensive in ZK, EVM terms. So maybe you can only do one painting or two paintings in a single transaction compared to that. But theoretically, it's possible.

Okay. And as of this summer, we have this ZK pre-compire, right? Can you kind of nest these with the AmS network and kind of get something like a recursive um uh zero notage um system if you have paintings if you have paintings actually you can have a roll up as a key roll up inside another as it can roll up yeah you have some sort of recursions if you have paintings yeah as i tell you here we need to see what's the final

cost of these uh paintings you know this is i tell you this is going to be the last the last the last thought code or the last, yeah, the last pre-compiled smart contract that we will implement it. But it's repetitive. You know, at the end, at the end, a painting is to do a painting. You need the Miller loop. You need the final exponition. Both things are very cycling. So it's very loops.

So it's very, it's much doable in the way that, so who we are building all these polynomial systems. So all the way that we're building this proof is absolutely global. But what may happen is that it's maybe a transaction can handle only maybe one pating or two paintings. Okay. So, but let's talk about what's on layer one in this, right? So basically, so when I, as a user, when I send a transaction, I don't send it on layer one. I send it to a coordinator, right?

A coordinator or a setup for coordinators. And the coordinator kind of, the coordinator takes all, basically the mempool of all these transactions and kind of executes them all together and then sends a proof to the chain, right? And basically a new Merkel route. Is that correct? Yeah, mainly is that, yeah. So why is the gas limit so crucial? The gas limit of the coordinator? The gas limit of the upcodes of pre-compiles.

No, this is more for building the proof. Okay, it's just for the proof. You need to build the proof, but not everything. The proof is like at the end is a circuit and the circuit can have a limit that number of operations. So if you exceed these operations, then you just, you just will not be able to build the proof. That's the same to say that you should will not be able to execute that transaction. Okay.

So how many transactions fit in one proof? We need to we need to do the final numbers. What I will give you is just like a first first rough. But the way, so we have like a double system. So we have a growth 16 or block that's validating a set of. starks. In each starc in each stark we don't have this is just it's not the final numbers but we expect to at least in each of start to be able to process half a million gas okay and we can have I would say easily about 1,000 starks in a single proof so this is maybe two orders of magnitude of

of improved terms to order of magnitude of the current Ethereum. But again, this is, you know, very preliminary theoretical calculus there. Okay. So in your white paper and also on your, on your homepage, you say that Hermes is very much focused on token transfers. So why do you say that? Because in principle, if you have all upcodes,

you can run any smart contract, no? Yeah, yeah, but it's like, okay, we have until now, so what we have is a payment, a payment, this is what's running, and this is in the mainnet, okay? When I'm talking now about ZK VM, this is not created yet. This is what we are working technically right now, and it's going to be the next version of Hermes some way. So if you are talking about payments, we are talking about what we have right now, I can give you full detail of what we have right now.

If we are talking about executing smart contracts, if you are talking about ZKBM, this is what we are building now and this is why I'm telling you about the quantity of gas. In the current Hermes, the quantity of payments, we allow 2,000 transfers per proof, and you can fit about 15 between 15 and 20 proof per block so this is more or less a total of 2,000 transactions per second if you were using full Ethereum

just to run you could not do anything else in Ethereum with those numbers but this gives you scalability, these two orders of magnitude, scalability on the roll up. And the main limitation is more in the, well, it's in the proof, but it's also in the data availability. So here are the limits that are, like, well, it's just, it can be, you can, you can do it better.

There is margin to, to improve these things. So this is just like the first version of ZK all up with these numbers. Okay. And so basically, as a ZK mechanism, you use either Gras 16 or Planck. And as far as I know, Gros XVI actually requires a trusted setup, right? Yes. So you'll do a ceremony at some point? We did the ceremony for the Hermes.

It's a specific ceremony for the circuit. If we upgrade the circuits, then we have to run like another trusted setup, I hope. But this is, I would say the Grosz 16 is perfect. it's like has all the advantages and it's the best proving system and everything is perfect except for these trusser setup. So if you need a truce setup, if you cannot handle a trusser setup, then you need to go to other proof systems. Blanc also requires a trusser setup, it's a universal trosser setup.

It's not a specific setup for for for for mess and personally I think that it should be enough. Growth 16 I can I can accept that's you know it's not perfect because at the end you need you need to end up trusting a set of people that run the specific trust setup. But in Plum, this process setup, this is universal process up, it's much bigger with much more people. It was created beforehand. It was created with different companies, with different people around. So I think it's horrible. And, you know, who knows in the future? Maybe there will be other proof systems that will not require process set up here.

For example, that Ziegatch people is working hollow, that does not require trust set up. And there are other starks that's not required, but the proofs, the proof sites is quite big. But there are other, there is a lot of research in this field than maybe it's going to be some proof system that requires. But for, from a pragmatic way, I think that growth 16 or plonk is, is good enough. Cool. So let's talk about how this looks for me as a user, right? So basically, I sent my transaction to the coordinator. So what's the job of the coordinator? The coordinator is just a little bit like a minor, if you want, just has some similarities with there. The idea is just collect as many transactions as it can and build a patch, which means building. a proof and sending that to the layer one. That's what a coordinator does in Hermes. And how do I become a coordinator?

Let me just, okay, I explain you in the Hermes, maybe in ZKVM, we will change this a little bit, but the idea of Hermes is that you just buy or you just beat for a slot. the winning of the who wins the bidding just gets 10 minutes and in that 10 minutes you just can build as many batches as you want so if you have a lot of transactions and if you are fast and if you have a good equipment equipment you will be able to extract all the value of those transactions and get all the fees of those transactions and so by extracting the value you mean the fees that people pay and not I mean in principle you could also extract um meb or maybe coordinate extractable value right yes in the case of in the case of uh it's not very much the case

the thing that the mining extraction value this is all this is more for when you have smart contract you have texas and you have uh other other kinds of things but in the ckbm this will become and so in the ckbm we probably will will will work will do it a little bit different. What are your ideas as to how to do it? Well, the idea is more about using more sequencer. So how would that work? So basically there would be a deterministic way of seeing which transactions go in which order?

Yeah, exactly. A little bit, a little bit will be that. But that's the idea. Okay. So the coordinator currently has to pay for the privilege of kind of mining the block or coordinating the block. What happens to tokens that the coordinator has to pay for this privilege? Yeah, well, this is part of it is burn and needs to be burned for the protocol. And part of it is donated and part of it is research.

recycled to the to the to the users of the to see to the network itself. Currently is 30% is burned 40% is donated and 30% is recycling. Okay. And if the coordinator behaves maliciously, can they be slashed? They can not. So what first is what means behaves maliciously because coordinator cannot do very much. So it can just forge or not forge.

And if it forge, it can forge the transactions that they have. So first thing is what means maliciously. So this is a little bit like saying what happened if a minor behaves maliciously. Well, they cannot insert their transaction, an invalid transaction or steal funds.

Things that may happen here is, for example, that he's censoring, for example, some sort of transactions. Okay. And in the case, in of their roll-up, we have a, we call it a forced transaction, and mainly it's a layer one transaction that's forced to be mined by the, well, forged by the coordinator. So coordinator has the obligation to execute those transactions. So your phones cannot be locked there.

So if you are doing, if you have the funds there and the coordinator don't want to execute your transaction, you can always do a layer one transaction, probable to exit, and the coordinator will either forge that transactions or will not forge anything. Okay. So, but what happens if the coordinator submits a false. proof. No, it's just it's like doing nothing. You know, it's, it's, that's, that's a, that's a difference between optimistic and ZK.

In ZK, you just send a proof and this proof is validated by the smart contract. And if this proof is invalid, then the smart contract, you know, just the transaction fails, transaction throws. So it's like the state doesn't change. Okay. So it's impossible to forge an invalid transaction in the ZKee roll up. even better when it's forged so once the transaction is forged it cannot be rolled back you know it's it's it's final it's it's there and this is the big

difference between ZKROL-up and optimistic roll-up in optimistic roll-up you need to you know a coordinator can can forge so can send an invalid transaction there and then raise this mechanism and this doesn't exist in ZKRWLAP in the secure wall-up you just send a transaction. You send this proof, this is cryptographic proof, which is hard and expensive to compute. But the cool thing is that once you submit there, the smart contract verifies and you're okay. You are, you are, that's it.

Okay. So basically the worst thing that the coordinator can do is they can, they can censor someone. So basically the way that you currently combat that is by having users submit things via call data, right? or how does it currently work? Well, yeah, a coordinator can sensor because at the end, it can choose which transaction currently in the Hermes. It can choose which instructions they include and which transactions they do not include.

But, you know, the user has always the option to create a layer one transaction, forcing the coordinator to execute that transaction. So it's simple. So it cannot be censored. Maybe you will have to pay the gas of the layer one to run your transaction. But you cannot, so the coordinator cannot censor you. Well, it can, the only thing that can do is don't forge anything, just to stop the network in some way.

But in this case, there is a time mode that anybody at this stage, anybody can forge and create a blocks. can forge. So if everybody, if everything is a stop, you can always build, maybe it will take one day to compute the proof in the normal laptop, but in one day, the transaction, so the next, you have for sure that in the next batch that's forged, either by you or there by somebody else, your transaction is going to be a chicure. So at the end, it's impossible to be censored in the worst, you know, in the worst scenario, You cannot be censored.

Cool. So a couple of months ago, we had Alex from Matterlabs on. So at a very high level, how would you characterize the difference between the ZK roll-up that Matterlabs is working on and the ZK roll-up that you guys are working on? Well, first fall, you may just say that Matterlapse is an incredible project and they are doing a very good job in this space. I'm not so in the in the case of the ZK m the CK VM part I'm not 100% sure of what they're doing because they have not not published too much yet okay so I cannot talk very much how they are doing they are following more an approach of you know just compiling solidity to a more

ZK specific set of old codes, which is a good path, you know, it's an interesting path and it's just here. We just choose a different path. We just choose a path to try to, probably it's a little bit harder in things, but it's simple in another, okay? So it's a different path and we'll see.

You know, we think, we believe that, or path is the right one but we may be wrong you know we see and but the interesting thing is that even if at some point

we just see that this is the wrong path you know just we may happen we can always go to the other path you know just so it's like because we're implementing at the end we're implementing the same

we're actually putting a I'm seeing there. So if at the end it's very easy to say, okay, just let's forget, forget about compatibility. Let's just use, let's simplify the IBM, just run this, this, this and this and let's run. We can always do that and this is always an option, but we really believe that we can be very, very compatible to the current EVM. So so far we've talked about scalability a lot. So basically the other thing that zero knowledge in principle can do is it can afford you privacy, right? So I mean, even despite the fact that currently everything is kind of centered around advancing scaling, could this also translate to better privacy down the road?

Yes and no. So I told you before that zero knowledge can be used for privacy and for and for scalability, okay? But they cannot be used at the same time for both. Okay. So if you want to have privacy and scaling, you need to some sort of recursion. So you need, for example, so you do like, you have like a kind of a tornado cache or a kind of a Zcash, some layer of privacy that's running inside maybe a ZKK rollout. Okay. But both things at the same time, you know, I know that for example, autistic people is working in that direction. You may ask. them, how they are doing and so on. They are, you know, great people and they are doing very interesting people. But we, our focus is, at this point, is more

on the scalability. Just having an Ethereum, but running in a ZK, so scale Ethereum. And once there, of course, if you scale Ethereum and the Stereum handles privacy, maybe you can run it inside the scalability. But we are trying

to split the split the problems and trying to solve one specific problem. Sure, sure. And I mean, it's a hard enough problem to solve. So a couple of weeks ago, the news broke that Hermes merged with Polygon, the scaling solution. So basically, if you look at Hermes and Polygon, they are architecturally, they're built

in different ways, right? So basically, Hermes very much inherits the security guarantees from Ethereum, whereas this currently is not true for Polygon. In a way, it's a standalone layer one chain with a periodic Ethereum commit. So I haven't really found a lot about how the joint model will look. Can you talk about this? Between Hermes and Polygon, of course, there are things that are very complementary. That's why I think it's a good merge. But we both share the same vision here.

The vision is mainly that we are not doing here these projects. We are not here just to do some gadgets just to play. We want to bring this technology to real people. We want to scale. We want to go mainstream. We want to grow. And this is the vision of Polygon and this is the vision to Hermes. And said that, we have been working a lot in this Zika relapse, you know, in this technology.

Polygon people, of course, accept that this is the right way to go. This is the, so they believe. and in the while they are doing things just to try to get customers just to get people using that and they are of course maybe it's not a perfect solution side chain

but I would say an enough solution for many projects and probably the best solution that you can have it right now if you want to build smart contracts and you don't want to pay maybe a huge amount of fees

and so on so on. So at the end, the important is understand what we are doing, where we are, what are the current technologies that are available, working together. I think we can go very far just to bring this technology to mainstream. And this is that this is why I think is good merge. There is very good feet in the management teams. there is a very good fitting there. And yeah, it's, for me, it's very exciting to be at this moment,

just part of this polygon mission there. So which is the same that the one we had. So it's, it's very exciting. We'll see, you know, this is at the end is. We'll see in the next month and years and see how the things evolve. But I think it's a very good, I think it's a very good. good thing that we can work together. Cool. So, I mean, the polygon network obviously is used quite heavily currently.

I see that there are different parts here that kind of where maybe it's not clear how they will get, how they will work together in the end. But do you have an idea of architecturally how it will work? So basically, will you, will polygon resort to the Hermes ZK roll up model? or is this not clear yet? I would say it's not clear yet. But, you know, at the end, it's what I told you before. It's just bring this technology mainstream. And bring this technology, when I'm saying bring this technology,

it means bring this technology with all the values that this technology adds. And if you have to do some concession at some point, it's important that we understand which concession we are doing in each site. And this is, yeah, this is, so Polygon will use the technology that's available at each point to have the most adoption they can and to try to keep, the values of decentralization as much as they can. And this is, so if this is the Hermes technology, that's great. But if it's somebody else technology, because it's better, it's going to be great too.

So it's, the idea is to bring this to, to what, it's, it's important that there are, that we have a lot of companies that are developing on top of Ethereum, that are thinking about blockchain, that is thinking about decentralization. And I remember that during the ACOs in the, in the 2017, in 2018, there was a lot. If you, if you check Ethereum, saw a lot of different applications, a lot of different people just building things on top of Ethereum. And I remember that a lot of those initiatives, a lot of those projects, they just stopped their business some way because they couldn't afford just, the Ethereum didn't scale. And so they couldn't afford paying the gases and make no sense just to use some insurance application just paying for it. every time you just hired the insurance, just paying $50 or whatever.

So it makes no sense. But they are very interesting applications. And at the end, it's just a matter of scaling. So it's important that these companies continue working, continue evolving, continue doing things on top of Ethereum. Because when Ethereum scales, whichever way, they are ready. they are ready running there so it's in some way it's just earning time in the in the future so that's why it's so important those uh side chase even if the properties are not as good as not the ideal as not the ones that we want to we want to have but they are doing their job they're

doing their their their mission okay even it's like it's not a test net but you know so test that is not enough sometimes. You know, this is some middle thing, you know, just, and this is, I think it's these side chains, you know, it's important for that, but you need to understand, and this is probably the key, the important thing, you need to understand what are the limitations of these side chains. Yeah, I completely understand. So, so let's talk about the roadmap and how this fits in with Ethereum 2. So basically the current vision for sharding is data availability shots,

which sounds wonderful if you're in charge of a ZK roll-up. So tell us about when you plan to, I mean, I know that these estimates are always fraud, and we've all been there. So when do you plan to go live with the ZK EVM and how will that change once data availability shots become available? Let's say, in our side, the ZKVM for us is more or less, it should be more or less a one-year project. Okay.

So maybe it's, maybe we can, for example, we are right now, we are accelerating and we're running a lot. But, you know, there are some, maybe there are some stages that we delay. So that's, it's difficult to, it's difficult to know exactly when, but it's not, it should be around a year. Okay. So this should be production. ready some way.

Again, things may happen is something that's new. A lot of challenges have happened. It's just like not fully coming. It just gives you what's in my mind, what should be. How this fits with Ethereum 2.0? I don't know. We need to see for us that the sharding is an important thing. Right now, I think they are working a lot

with a merch. Yeah. With a, so and probably sharding is going to come, it's going to come next. It's a path, you know. We will see and it's a stage to stage. Right now, the cool thing of Ethereum 2.0 is,

you know, I remember in the 19th, in 2019, 2018, that was not clear. And there was, every time that we talk about, Ethereum to the zero was not, it was like the research was not done and there were a lot of things

and even changing one from one side to the other, which is normal because when you are researching, you know, thing changes and you go deeper in one topic and then you you see that you can do it much better, so you just go back and it's normal. But from outside, it's like it was, we'll never had. When you see it from outside, we will never have Ethereum 2.0. This is never, we'll never happen. This is the feeling that a lot of time we had. At this point, I think that nobody doubts that Ethereum 2.0 is going to happen. Maybe we are talking

is this. We will take one year, two years or three years. But Ethereum 2.0 is happening. It's the way to go. We replaced the concept of this we call it before execution environments. No way call it ZK roll-ups or ZRolaps, at least, if you want. It's not exactly the same, but almost the same. And this evolution, you know, just this evolution of Ethereum

with all these zero-knowledge technology that has been evolving, all the work of the sharding, on the beacon net, on the consensus mechanism, all this, you know, all this is coming together. and I think that at this point nobody doubts that blockchain and concretes and concretely Ethereum is going to be a global scaling blockchain very soon. If you talk to me maybe in 2008, if you talk to me maybe in 2008, in when the scaling solutions were, you know, state channels or even these side chains

and things, it was very sceptical and for me it was not clear. But at this point, I think the path is quite clear. No, this is going to scale for sure. And it's just a matter of, at this point, it's a matter of engineering. It's a matter of just putting the pieces together, building. Of course, maybe improving some protocols and things, but it's there. It's more fine-tuning and building, but this is going to happen. Cool.

So maybe to conclude, let me zoom right out. So you've always been a person to think a lot about what's right and what the the right thing to do us, right? I mean, basically with the Robin Hood group and the whitehead hackers and Giveth and so on. So, I mean, currently you're working on technology that makes the blockchain more performant. But on a societal scale, what do you hope will come out of this? So basically, if you look at blockchain, say, in five to ten years, in what metric will humanity be better off because of blockchain? Look, at the end, I hope that the people get proud of the communities they belong to.

So here is empowering people that means that you are part of a society. You know, you were born. You will die and you don't exactly, you don't really know what you are doing here. in this world, but you know that you are surrounded of the other humans that are more or less the same than you. And you are trying to do the best thing for all of us. So at the end, you are just building communities and you are just trying to, in some way, work together to have more, to be more happy.

And this is what's life about. And the cool thing of a blockchain is that this should allow you to be part of the community, of the community that you want to be in the community. should be half opt in and opt-opt-opt-opt-opt-opt-opt-opt-opt-o part. That means that you should work for the community. You should... This is why governance, for example, is something that's so important.

But, you know, it's just governance, but also organizing or distributing, if you want, distributing the scarcity resources some way. This is why the money is about and why the tokens are about and so on. this is what can blockchains and this technology can help if you see

if you see the world right now just thinking how many states how many conflicts are right now in the world because a lack of a lack of trust in the central authority

and all these places you know places where there are wars right now or countries that, you know, let me put an example here, it's just Venezuela. Venezuela is a democratic country and the people just wants to be how they want to build, or it's just a dictatorship and who knows, you know, there is a central party that's organizing some elections. Nobody know if this is corrupt or not. It's a lack of trust.

Imagine that you can just organize a voting system where it's just real and you understand who it works. This will solve a lot of things right now in that conflict. And this is just an example. There are many conflicts and many things in the humanity that can be solved just by having to trust central. authority and this is this is what's the blockchain so this is what the blockchain will add to the society of course blockchain is a technology is not enough for this transformation of the

society to happen you know that it requires some social transformation to okay but definitely the blockchain can be the technology that triggers this change in the society for sure. Cool. Thank you, Jodi. So where can people go to find out more about Hermes? That's Hermes.io. I think you can find all the information there. Right now is Polygon slash Hermes.com. But yeah, yeah, both domains work. So just go there. We have our Reddit group. We have Telegram group. Of course, Twitter also there. I think it's just Hermes with Zit, you know, just, and that with Zid, just, it's easy to find us.

Cool. It's been a real pleasure to have you on. I've wanted to do this ever since you gave that really astonishing talk at ECC, which we'll also link to in the show notes. So if you're interested in the maths behind this, go check out that talk as well. It was fantastic. Thank you so much for coming on.

Okay. Thank you very much. you. 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, guest or other podcast

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.