# Interoperability in the Ethereum Ecosystem — Radina Talanova | L2BEAT

- Speakers: [Radina Talanova](https://streameth.org/speakers/radina-talanova)
- Channel: [ETH Belgrade Community](https://streameth.org/eth-belgrade-community)
- Date: 2026-10-06
- Duration: 19:25
- Topics: People & Blogs
- Watch: https://streameth.org/watch/yt-Y9HAXMkxuns
- YouTube: https://www.youtube.com/watch?v=Y9HAXMkxuns

## Transcript

Right. Um, hi everyone. I'm Adi, product manager at L2B. Happy to be here in Belgrade. And, uh, today I'm going to talk about Interrop within the Ethereum ecosystem. Um, I'm going to go through some things that we've been working on at L2B. Um and then the bridges that are currently present out there and some developments that are in progress and would uh improve the interop. All right. So very quickly going through the basics what is interrop and what is the problem that is trying to solve. Uh the main problem is that the different blockchains are not aware of each other's state. So chain A only knows what's happening on chain A and same for chain B. Um and then in the context of moving assets, the thing is that you cannot really move the assets. Something needs to happen on chain A and then something else happens as a result on chain B. And the interop solutions are trying to answer the question um how does chain B know that something indeed happened on chain A? And there are different ways that this uh question is answered. Um oops. We started looking into uh these solutions back in 2022. Um with the first version of how to beat bridges. Back then it was focused mostly on the token bridges or the ones where you lock assets on one chain and the asset is minted on another chain. And for this reason we also use the total value locked as the main metric for the adoption. But at some point we realized that the space has changed a lot. So this is no longer uh that relevant and we decided to do a second attempt um on interop. So a couple months ago we launched um L2 beat interrup second version and we made a few changes compared to the first one. Um the main thing is that here we use volume as the main metric for adoption. Um this uh graph is animated on the website so it's way nicer but um it already shows how value is moving across the ecosystem and something that you could see is that um Ethereum is the top chain by volume and this has been the case every single day ever since we have launched the feature. So it's already functioning as um an economic and settlement hub within the ecosystem. Uh but of course this could also mean that there are no good bridges to go between the other routes. So you always have to go back to Ethereum and then go somewhere else. Um another important change that we did is in the categories of bridges that we identify. So there are three main categories and we split them based on the types of transfers that they do. First one is intent bridges that do non-minting transfers and I'll go deeper into that in a bit. Then we have the token frameworks. Uh they do burn and mint types of transfers and we have the cononical bridges that are the lock and mint type. Um and you could um already see from the data that um these are to be leaked looked into in a different way. Um because for example the intent bridges account for the majority of the transfers but at the same time it's very small amounts going through them. um usually for the largest amounts and I mean uh millions of dollars in tokens people are using either the canonical bridges or the uh burn and mint token frameworks. Uh but if there's an issue controlled option uh they do prefer this option. This is the case with uh CCTP that's the official bridge for uh USDC by circle. Um so the reason is that uh with intents you need the liquidity to already be on the destination chain and sometimes there's not enough liquidity to breach millions. Uh so then you have to fall back to the um canonical beach or the slower option. Now I'm going to go um through the different categories and the differences between them. Starting with the canonical bridges. By cononical bridges we usually mean the official or native bridges the ones where the chain operators uh are maintaining the bridge. So there is no additional trust assumption introduced uh apart from trusting the operators that you're already trusting by moving your assets to this chain. Um they usually use the lock and mint uh types of transfers. You lock the assets on one chain mean the assets on the other. And this means that the assets on the destination chain inherit the risk of the canonical bridge. Uh good thing is that this is limited only to uh the amount that is already locked uh in the escrow on the source chain. Um usually they don't have any protocol fees. You only pay the gas fee. Of course on Ethereum it could be a bit higher. uh but they are slow uh because they are waiting for a couple of blocks to avoid reorgs um and deposits take minutes. Uh withdrawals it depends if you're using zero or optimistic row but for the optimistic ones they could take days. um zk takes minutes to hours because um zk proving steel um it takes some time for the cononical bridges we already have the stages framework that some of you are probably familiar with um it evaluates the decentralization and the maturity and people usually think of it as um let's say base chain is stage one but what we're really looking into are the properties of the canonical bridges So uh stages depend on these properties. Um currently there are I think six stage one projects. There are some projects that are already stage two. Uh but it's important to note that stages do not reflect security. So it's not necessarily that stage two rollups are more secure uh to use. Then moving to the intent bridges. Um the way that they work is that there's already liquidity on the destination chain. You lock your assets on the source chain and then a solver or liquidity provider um fulfills your intent by sending you the assets on the destination chain. Um here the risk that you bear as a user only lasts while uh funds are in transit. Once you get them on the destination chain, you don't really um care about the risk coming from the intent bridge and it's the solver who takes the uh finality risk or um the possibility of the chains having curly and for this reason they also take some fees on top um for taking on this risk. Uh these bridges for this reason are super fast. Uh relay and the cross usually take two to three seconds. Um so yeah the risk is very shortlived for you. However you still might be interested in um some of the properties uh of these bridges. Uh one that you as a user care about is what happens if my intent is not filled. Can I request my funds back? Um this is what uh user recovery shows here uh on our intent pages page. And then as a solver uh if you want to start participating in any of them you might be wondering if you can join permissionlessly. And then um what is the mechanism that proves that you have fulfilled the user's intent and you can be repaid and get your rewards. Um and I don't want you to understand bits but I just want to say that even though the risk is shortlived there are still some risk. That's an example with a cross where you as a user log the funds in a contract and then there are multiple operators and then at the end there's a challenge um system uh responsible for making sure that the solver has completed the intent. Um there is a good onchain trace of everything that's happening. Uh and for comparison, that's relay where you uh send your funds to an EOA and then you get your funds from an EOA on another chain. Uh you're essentially trusting an EOA here. Um yeah, there is some risk involved. Uh in this case is the risk of the EOA just uh stealing your funds. And then the last type of bridges are the token frameworks. uh these are projects like layer zero or chain link um that uh have token standards where the interop features are embedded into the token standard. So whenever you create the token it comes with this interoperability by default. Um they use burn and mint uh types of transfers. Token is destroyed on one chain and is minted on the other. And uh because the interop is kind of enshrined into the token, this means that the entire token supply is affected and it's affected at all times. So the risk is embedded into the token. Um they also have some protocol fees on top and are usually fast. Uh transfers take seconds to minutes. Uh but this is probably in our opinion the worst type of bridge because um yeah you uh take on the bridging risk and unless you um migrate your token away at some point which involves some operational costs uh you cannot get rid of it easily. Uh plus all these solutions are very complex. uh we have already the research for layer zero and chain link and soon we'll uh look into the rest uh but if I have to um briefly give you a sense of the complexity within layer zero this is an example of an onchain deployment um on a single chain and if you want to enable um users to bridge the token between let's say Ethereum and arbitrum you need one config for this route and that's the thing with the ping border on the right and then if you want to allow them to bridge back you need another such config. So for every route that you want to enable you need a separate config that includes some libraries validation rules and so on. So um this mess here that you see in the background is an example of the configs for a single of token. Uh and the red lines are all the configs that you need to configure separately for the chains. Uh that's a lot of complexity. is very difficult to monitor. Uh we tried but it would be very expensive to do it for all tokens. Uh and that's why we call this fragility at scale. So the bigger this gets the more fragile it becomes. As if one of these configs fail then the entire system um could be affected. Uh there was this uh the kelp hack that uh you probably heard about. It happened because of a one of one deviian that got compromised. It's a good thing that after this hack many projects started switching to higher thresholds. So uh previously as you can see on the small charts uh the blue one represents messages that required two deviants. Uh after the hack people started switching to three devians which is the pink area on the chart but if it's not the devian uh it could be something else just because the systems are very complex. Um that's on the different categories of bridges and I want to give a quick update on the interal between the canonical bridges as interal between the ro ecosystems was a core premise for chains to join these ecosystems. Uh currently the only ecosystem that has interop is a layer but just because the chains within the a layer um do not have high activity uh it's not really used. uh plus for some reason they prefer using external bridges for most of the assets there. Um opt is currently working on something called op uh super node. Uh this is going to be um an implementation similar to op node but if the chains are now deriving their own state with op super nodes they have to derive the state of all the chains that are participating in this interop set. Um so it will be broadcasting messages and then it will be on the applications to decide how to consume these messages. An example of an application on top of it is the superchain ERC20 standard that they uh created. So that's currently in progress. I think soon they're going to uh test it out on a test net with unit chain and I think optimism. Uh so there is uh still some work to be done before it hits the mainet. Um, elastic chains uh recently uh announced the deprecation of gateway which was their interop solution. They said that they'll focus on interrup between the private chains or the previdiums that they're building. Uh but since then uh there haven't been any updates so we're not sure if that's still in the works. And arbitrum uh and the orbit ecosystem they never really had big interop plans. Um couple months ago they announced that uh they would go with intentbased bridging and promoted this crosschain broadcaster standard but they haven't followed up um since then and um last week uh they announced that they are working on ZK proofs which is it won't improve probably the inter interoperability between the L2s but would improve the one between the chains and Ethereum because it would allow for faster withdrawals. Uh so that's still something. Um if I have to summarize uh what I said, I would say that interop is sold because you can move messages and transfers around, but it's not trust minimized. Um and the idea of um having the Ethereum ecosystem as one chain, having this user experience is still not there. Um, so if I have to give you some suggestions on how to act, I would say that for developers, know who you're trusting when using bridges. If possible, do not use the token frameworks or the multi-chain frameworks because they enshrine bridge risk into the asset and you don't need that. If you really have to use it, if you have a business case for it, uh you could deploy the asset on Ethereum and then wrap it in an let's say OFT adapter and limit the risk to um specific supply uh but it won't be enshrined into the entire token supply. Then as a holder once again know who you're trusting and for the bigger amounts um go to the canonical or the issuer controlled option um because the intent ones usually won't support that. Um something that we've been working on to help you better understand who you're trusting uh is this dashboard of token relations. Here what you'll be able to see is the original set of deployments and then any derivative tokens that have been created by bridging the original tokens to other chains or wrapping them in some way. What you could see here is um the USDT token there is this big USDT0 section um which is created by locking the original USDT in an adapter. So these tokens uh within the section are interoperable between each other and yeah you could see that there are many different versions of USDT. Uh we're also looking to integrate with wallets and interfaces to surface this information directly to users about their portfolio whenever they are trying to make a decision. Um and just a few words on um some developments in the space that would improve interrop when it comes to trust minimization and faster withdrawals. That's for sure zk proofs. Um we already have zk catalog and we can see that many projects are moving to zk proofs. Uh b was the latest one. They announced multi proofs with the idea to reduce the withdrawal time to one day. I think currently it's five days. Um and then Arbitum said that they're uh working on extending the boat proof system with ZK proofs which would allow once again for faster settlement um on Ethereum. Then native verification and native rollups uh our research team is actually working on that and it solves uh similar issue. It will basically improve the risk properties um because the chains will no longer have to maintain their own proof system. They'll be able to use Ethereum's native proof engine, but this engine is still not on mainet. Uh so there are probably a couple of years until uh that's possible. Um there is the Ethereum economic zone which got announced uh I think in Khan this year. Um, it aims to solve the issue with the onechain UX. The idea here is that if you're an EEZ rollup, you'll be able to call a smart contract on Ethereum or on another easy rollup and then get the result of that and act on it in a single transaction. So there will be atomic execution uh between these chains. Uh but uh this is uh overall very ambitious um and will take some time. Currently I think there is a definite um and last week uh Gnosis DAO passed a proposal for Gnosis chain to become the first easy roup um so they'll probably experiment with that um is the only solution that's trying to solve this problem uh at such a scale right now. So we are hopeful that uh it would work. Last but not least, the Ethereum stroll map contains many improvements on the core protocol uh which would benefit not only bridges but all kinds of applications. Uh an important one is that uh Ethereum will at some point get fast finality and if you have faster blocks then you'll be able to get fast confirmations and as I said being able to confirm that something happened on a chain is at the core of Interlope. That's it. Um, thanks everyone and [applause] happy to answer any questions.