[trustless://eil] Yoav Weiss - Making Ethereum Feel Like One Chain Again

Transcript

Hello. Hello. Who's excited for Trust List today? >> Like maybe one more coffee. Who's excited for Trust List today? >> All right. Thanks, guys. Thanks, guys. Uh my name is Britt. I'm really excited to be MCing today. And first up, we have Yo Weiss, EF researcher, head of the account abstraction team, and co-author of the Trustless Manifesto. and he's gonna be talking about making Ethereum feel like one chain again. Big round of applause for Yav Weiss. >> Hi, good morning everyone.

Sorry. Yeah, talk to Maurice about that. Okay. Yeah. So, uh today I'm today I'm going to talk about uh so in this talk I'm going to talk about uh making Ethereum feel like one chain again. And it's uh this talk is going to be is not going to be very technical. It's about the vision. It's about the vision behind EIL and uh not a deep dive into the protocol mechanics which we're going to do later in the day. It's mostly about what it enables and why it matters.

So, uh how many of you here use the used the internet before the mid 90s? Right. That's that's even more than I expected. Yeah. So, but um many people in this ecosystem were not even born then. So, um so I so I have I I used it and it wasn't great to say the least. It's I mean it was very innovative but uh usability was lacking. Um and the main the main thing that made it difficult was that each um each server and each protocol were an island. uh you couldn't so I could TN net to a server and use it and use it in various ways. I

could FTP to a server and download a file or I could go and go to an IRC server and chat with my friends. We even had the gopher which looked a little bit like a website but it was always local on one server. So there was no there was no composibility, no way to uh no way to move between the apps. like if I I can download a file and then I can go to a then I can switch to another app and uh I can switch to another app and go chat about it but not in one not in one place and the friction of moving between them was um uh was quite high. Uh I would have to switch an application on my computer just like move between I don't

know between TNET or FTP or an IRC client and I would have to reauthenticate myself. So I have to authenticate myself separately to each server and then it has a single use and the app has to do everything. It's not it does not combine functionality with other apps. Then HTTP changed everything. Uh one of the great things that HTTP did is to abstract away servers. So we can have one applica so one application can span across different servers. uh it could load an HTML, it could load the content uh text content from one server and then load the pictures from another and then

even uh code like JavaScript libraries that come from various servers and later later we even started seeing specialized services running on different servers such as you could have a secure payment uh could have a secure payment server for which enabled e-commerce and you don't have to implement your own uh h your own payment system in every website. And we even got identity with OOTH meaning servers can uh meaning that uh you can have your identity in one place and use it to authenticate to many different apps on different servers. We had video streaming services. So this really made the web what it is what it

is today. And what and what ties it all together is the browser. The browser is the user agent that this is the the user agent for everything. So now we have one application and a infinite use infinite use cases instead of having one application one use. So now uh going back to Ethereum um we so rollups rollups helped Ethereum scale. This is Ethereum's scaling strategy is based on having on having rollups on having layer 2s and it's great. We now have abundant us abundant blocks block space uh spanning across different layer tools but we broke UX.

So now for each for each L2 you can use it with a reasonable uh uh convenience uh like you would at least like you would Ethereum L1. But once you need to move between them, we have a high friction. So uh we need we need to switch to another RPC. We need to bridge our assets. Uh we need to um so uh we need to configure various things. We have like there's a there's very high very high friction in uh switching between them. So we don't see a lot of applications that combine uh things developed things that live on different chains which is unfortunate since uh one of the things we've been seeing is app

chains. So we have a a chain we have a chain that uh that hosts a single application but then the friction of using this application is quite high and uh so so we need to solve this problem of rollups feeling like an island and make it feel like one chain. Now the process uh the process nowadays is uh if I want to uh move between I want to do something on a different chain is I need to sign a transaction in my wallet and then I need to sign a transaction which sends assets or something to to a bridge then I need to wait for the for the bridge which is often often a trusted bridge. I send my

assets to someone and I wait and then uh I and then I'm prompted to switch to another to another network in my wallet and maybe I need to configure that network if I haven't used it before and then I need to sign yet another transaction on the new network to do something. So that's not what a what a should uh should feel like. It's uh it it feels a lot like the days before HTTP. It was all manual, errorrone and quite frustrating to use. So our vision is to have is to have Ethereum feel like like a single chain across all L2s. And what this what this

requires is using one signature per operation, not per transaction on each chain. If I'm running something that involves five different chains, I still want I want to see a list of what I'm doing. I need the clear signing but I want to sign to see the list to sign it as a single operation and then it should and then my wallet should take care of the rest and uh I also and I also want to pay Gaz once because maybe I don't hold ETH on all of these network. I have it only on one and I want to pay the gas for all of these uh for for my for my transactions on all of these different networks. I want to be able to use my funds, my assets everywhere. I don't

want to care about whether my assets are on optimism or on arbit room. I just want to use them. And we want all this to all this to work with the same security and trust assumptions. We want to give I mean Ethereum is Ethereum is all about censorship resistance and trustlessness. So we want to solve the UX problem but not by compromising and not by compromising and having a weaker censorship resistance for example and so this is the so this is the experience that we are building it's u let's say Alice wants to send uh Bob so Alice on arbit room wants to send uh to send USDC to Bob on base she signs once

there is no bridge She doesn't even know that that he's on another chain. She just pastes Bob's address, which is uh which contains all the information that that the wallet needs in order to know how to get there. And then she clicks send and Bob just uh and and Bob receives the funds immediately. Now, Ethereum is not just about transfers. It's about an so uh Ethereum was built to uh to run to run DAPs to run we want the decentralized applications running these are smart contracts you want to send transactions to them so we want it to be equally equally easy and equally trustless uh to make calls to make calls with a single

operation on multiple chains and this happens again with one signature so let's say Alice let's say Alice lives is on orbit room and she wants to mint an NFT on on optimism but she don't she doesn't have ETH on optimism and so and still she only signs she only signs once and her wallet and her wallet takes care of everything. So a wallet sees that she has ETH on on orbit room. It it enables her to use the ETH on on optimism both for paying the gas and for buying the NFT and she just receives the NFT uh instantly uh back on uh back on orbit room.

So that's uh so that's like a one click one signature and the wallet executes the operations everywhere. So HTTP unifi HTTP unified the web and made what it is today. And we think that EIL could do the same for could do the same for Ethereum. You'll be able to use your wallet as the your wallet becomes like the browser in the mid '90s where it allows you to it allows you to compose the functionality from all the different uh chains and use it as a with one UX. You never have to switch. which you never to have to bridge and that's just work and most importantly we are

preserving Ethereum's Ethereum's core value which uh which is to also give you censorship resistance self custody and all of these important uh traits. Uh one way to look uh one way to look at it is to compare centralized exchanges to the central to the centralized ones. So we everyone here used the use centralized exchanges and they're generally convenient but they require trust assumptions. You need to send your uh you need to send your assets to a to a to a third to a third party that lives offchain. You basically send it to a send it to a server and then you swap or you do you

do whatever uh uh with the assets and finally uh you can withdraw it back to the blockchain uh assuming they let you and then if they go down uh if they go down they become unavailable or maybe something along the way uh becomes unavailable like uh this morning a cloudflare is down globally and uh maybe you're not able to use uh to use your centralized exchange. So now you don't have access to your assets. Now compare this to that to a DEX. So in the centralized exchange you swap directly onchain. It's selfservice. You don't have a counterparty offchain.

So you retain self- custody. You you hold your assets and then you send them to a smart contract which which instantly within the same transaction gives you uh the other asset. So it's a so if we look at interrupt today at you know bridges relays solvers there are they resemble a centralized exchange you also uh you also send your funds and your calls to another you send them uh to an off-chain server. Now you have this offchain you have this offchain entity that's supposed to be to be your agent and act on your behalf but maybe it doesn't have your best interest at heart. It is a third party server

operated by someone else. So you submit your you submit your funds and your calls to someone to an intermediary. The intermediary runs some off-chain logic. Often you don't know what exactly it's going to do. It determines how to best fulfill uh to fulfill your intention. And then it and then it submits transactions on your behalf at the destination. Now compare this to EIL which I'm going to talk about. So with EIL you submit your own transactions everywhere. Nobody ever submits a transaction on your behalf. It's it's self-service just like just like a DEX. you retain self- custody the because funds are moved atomically similar to atomic swaps

and uh there and there's no middleman involved. So EIL is accountbased interop means that it puts the wallet it puts the wallet in the center. Yay lives locally. It runs locally on your machine inside your wallet and not on a not not on a remote server. You don't delegate you don't delegate anything to a to a server that does something on your behalf. Um users um users perform their cross L2 operations themselves.

Now uh we still I mean you can execute uh yourself but you still need to pay for the gas and to move liquidity from one place to another. So you need someone's help. We have uh in in EIL there's something called an XLP a crosschain liquidity provider but the XLP this uh liquidity provider it never um it never acts on your behalf. It will never transact. It doesn't know what you're going to do on each chain and it doesn't do it. It doesn't do it for you.

What it does is it sells you a gas voucher or asset voucher which is an atomic it's a form of an atomic swap. You send a so what so the only information you expose which is all onchain because there are no off-chain interactions. The information you expose is you say I need I need to use a 100 USDC on this other network. So I want to use a 100 USD USDC and I need to pay a gas a certain ETH amount for gas. That's all you that's all you have to reveal and then uh and then permissionless uh liquidity providers compete on chain on giving you a on giving you a voucher and they get a fee for it. So you get this

voucher and now and now you use it yourself. you put it inside your h inside your transaction for the uh for every destination chain you use and the voucher is used to give you the voucher is used to give you assets that come from this XLP's balance which means that uh so it works like an atomic swap the XLP did not do anything you executed the transactions on both ends and uh and the assets remain in a in your custody because until they give you a voucher the assets are yours and once they give you a voucher this voucher is I mean the only way they claim the funds is with this voucher and the voucher is now used on the uh used on the other side the

same message that they sent that they had to sign in order to claim your fund is the message that you use to claim theirs so this allows us to retain Ethereum's core values because you are censorship resistant because you execute your own transaction everywhere. So nobody can delay or stop your transactions and uh you don't have to so we don't introduce any trusted parties. There's nothing permissioned in the system. What does it feel like for the user? I mean the user doesn't want to feel or you know to to even know about all this just wants to use the just wants to uh to use the network. So it's a it's one

click to a it's one click to do a to do any operation regardless of the number of chains. The user never sees different chains. You don't switch between them. You don't even know where the DAP lives. What you what it feels like is one Ethereum that just works. You can you use your assets everywhere regardless of where they are and use applications regardless of where they deployed. You can use contracts on any number of chains. you don't know which chain they are deployed on. So, everyone wins. The users, the uh the users, developers, and the Ethereum ecosystem.

For the user, it's uh for the user, it's a very smooth UX. You just um so you can you can just uh sign something and everything works. And you don't need to add any uh to add any new trust assumptions, any new security considerations. So for the user, it's a clear win for the wallet. It's uh the wallet is empowered to become the user's uh the user's multi-chain agent. It becomes a multi-chain native and does not add dependencies on third parties. So if you're a wallet developer and you uh and you add support for EIL, you can just um uh you become the browser for the user

on this new uh on this new internet and for applications it for applications it's great because they become available to everyone. They are not there's no friction in using application. So you don't need when you develop a new contract you don't have to start thinking okay where where am I going to deploy this? you can deploy it in one place and it can be used from others without uh without friction which also saves you a lot of effort of doing interrop plumbing. I mean nowadays uh if you want to if you want your DAP to become available to other chains you have to do some work on the DP side. Uh with EIL there's going to be a there's going to be a standard that extends

extends the ETH send calls uh batching ERC. it extends with a new capability allowing allowing the DAB to just say here is the list of things that I want to execute. Here's a list of calls that I want to exit on each of the chains and that's it. The DAB doesn't know what EIL is. The wallet just gets this uh batch of uh let this multi-chain batch and shows it to the user gets it gets it approved and uh executed and the whole ecosystem wins because now it feels like one chain again. We have a unified Ethereum without giving up on a decentralization.

So today um today we are launching a test we are an experimental test net uh we released the uh release the experimental code it's uh by no means mainet mainet ready but we already we can already invite wallets and dabs to test it to start integrating and experimenting with what it enables and we'll only get to u we'll only get to mainet after getting after getting some initial feedback on this, making improvements, and then of course getting a getting an audit because this needs to be secure and then we'll get to mainet.

So now is your now is your opportunity to imagine uh with one unified Ethereum but with abundant block space and and the smooth UX, what can you build? uh what can you build that uh now that Ethereum is unified the HTTP enabled enabled so many new use cases and made the internet what it is today and uh it'll be interesting to see what people come up with what builders can do when we have something equivalent for Ethereum yeah so HTTP made the internet usable by everyone suddenly my mother was able to use the internet

and uh EIL can do the same for the for the L2 for the Ethereum L2 ecosystem. So, let's help users forget which chain are on. Uh right, thank you. Um right, do we have any questions? Yeah.

>> Oh, there's a Oh, wait. There's a We have this uh >> We got fancy tools for this. >> Try. >> So, there's actually been a QR code on stage all day. >> Okay. >> So, we've got a couple good ones. >> So, did you send it? Okay. Okay. Please. >> All right. So if you can deploy on any chain seamlessly, would this not result in everyone deploying on L2s to save on deployment costs and no one employing deploying on the L1? >> Uh no, because L2s have a so each L2 has its own trust assumptions. They're not created equal and and luckily we have L2 which helps us uh which helps us know which chains we should use and I think wallets should reflect that. So one part of the vision that I haven't talked

about because you can't cover everything in one talk is that uh when when we abstract chains and make the wallet discover it the wallet should also be able to look up on chain and at a station by an entity like L2 bit that checks like is this a stage one roll up stage two or maybe it's just some side chain pretending to be an L an L2. So, um, so the wallet should warn you actually, and if you're going, if you run a transaction and like, let's say you just use $10, maybe you're going to be okay with that. You say, "Yeah, that's fine. I'm I'm going to take this big risk on on $10." But if you're going to send $10,000, maybe uh this warning will mean something to you. And I said,

"No, actually, I want I prefer to use a DAP that lives on a chain that is more secure, whether it's a secure L2 or in some cases even L1." So the user gets to choose. You need to inform the user and that's the responsibility. I mean the wallet gets a lot of new responsibilities here and this is one of them. >> Awesome. Thank you. >> I have one more and just if anyone has any other questions, please feel free to sk scan the QR code that I'm sure will pop back up there in a moment. Um and you can chime them into the chat. Uh second question is can you explain the difference Oh, we got a couple coming in. Uh between EIL and OIF.

Yeah. So the difference between uh yeah between EIL and O and OF is intents versus transactions. So first we need to understand what is um what's an what's an intent? What's a transaction? Transaction is this imperative thing where we say exactly what should be executed where it's like I want to execute this call data. I want to call this function in that in that smart contract. Intense on the other hand very high level construct. With intense you can uh with intents you abstract it. You say what you want what what you want the end result to be not how to achieve it and then a solver will figure it out for you. It will the solver will know like I say okay I want to have funds on that

chain or I want to have that NFT and now a solver figures it out figures it out on your behalf and they will get it done. It's very convenient. It's so OAF does intense and EIL does transactions and the reason we chose to do it this way is we wanted to create this transport layer which you could actually build something like OIF on top. So we wanted to have this transport layer that doesn't interpret doesn't let a third party interpret anything on your behalf. You don't nobody interprets your intention. You decide yourself what's going to uh like what's going to be executed. But then uh so so you lose the convenience of having someone figure it

out for you. Your wallet now needs to figure it out. Whereas um with intense with intense you have so you have this flexibility but you pay with adding this uh with with adding this external dependency. So there's room for there's room for both approaches. They just make they make a very different tradeoffs in terms of how you could uh I mean in terms of who do you need to trust. >> Wonderful. All right. And just um so what if a transaction puts money on a chain not supported by the wallet?

>> So that's another part of the vision that I haven't covered that I haven't covered here myself but Marissa is going to talk about later. That's uh uh part of uh the vision requires multiple standards. One of them is a chain specific or interoperable interoperable addresses where you have to send someone. So so like when Alice sends to Bob on base she sends actually to 0x124 at base at base.eth. So that's how that's how Alice does it know about base. But what if her wallet doesn't know what base uh what base is? That's fine because there's going to be an onchain config. Every roll up can and that's also something that we've been

working on a standard for. So every rollup is going to have a a standardized form of a config contract that has everything you need to know in order to start interacting with it like the official RPC. And so you have the official RPC, the official bridge, official light client contact for verifying what this RPC tells you. So now when you when a wallet needs to interact with another uh with a chain that it never heard about, it can get all the information from a contract on L1 to that is registered to the ENS address or somehow reg some associated officially by the rollup team and that's how you get to that's how you can interact with any chain without knowing

about it. Now, of course, it means since your wallet doesn't know what this chain is and maybe this chain is actually like a fancy multi-ig and you don't want to trust it with large amounts. So, that's again why you need the uh why you also need this attestation contract where someone like L2B can tell you uh this is uh uh this network is safe to use and if not the wallet should warn the user about it. Is the EIL only compatible with EVM compatible L2s?

>> Um, yes. It's a technically it could be implemented for other chains, but our focus right now has been EVM chains. That's what we we implemented it as EVM contracts because it also internally it uses ERC 437 which is also built for uh it's also this account obstruction for EVM. So this remains so this remains a requirement for now. I assume that we'll see implementations of EIL for for other networks hopefully compatible ones but we are not we are not working on one right now.

>> Awesome. Let me see if we have anything else left in here. Also anyone in the audience please know you can upvote questions. So if you don't want to ask your own question feel free to contribute to um the popular vote. Um, how can we participate in the test net integration? Is there any doc link? >> Um, yes. So I think we're going to talk about this uh later in later in the day but uh we have a so we have a test net on we have a we have it deployed on sepoleia in a couple of rollups and we also and for the hackathon since there's going to be a there's going there's going to be a a hackathon at E global where we want to have people to build on

it it's easier to build on a it's it's easier to uh to build when you have actual real protocols deployed. So we have a test net we have a test nets for various rollups that are actually that actually forks off their mainet. Uh this is hosted by Tenderly and we have a we have EIL deployed we have XLPS running so you'll be able to start experimenting with integrations as a wallet or as a DAP developer by uh using these uh mainet forks of various networks and interacting between them. Awesome. Thank you for your time today, Yav. This has been amazing. I hope everyone's really

excited about what we heard today. Can we give a big round of applause? Thank you.