Scaling Ethereum Together with Brendan Farmer, Harry Kalodner, Alex Shevchenko, Lea Schmitt

and Alex from Aurora and when diving in so the panel will be focused on scaling ethereum together with like the emphasis on together and if we think about the recent narratives it has been very much like focused around L2 Wars and like who's first which scaling solution will make it and when you introduce yourselves I kind of want you to like obviously like describe your project like really shortly but then also pick one of your favorite scaling solution project that's not yours and give them a little compliment um so hello everybody my name is Alex I'm the CEO of a company that is called Aurora Labs we're the main developers of the Aurora protocol Aurora is an evm compatible Solution on top of near protocol which is horizontally scalable L1 and Aurora allows to launch your solidity code on top of near so in terms of the scalability solutions that I really like and kind of Believe In the Future these are ZK EVMS that are moving forward we know that ZK sync just recently rebranded in Era right they launched their solution so pretty cool stuff I think that in the future the KVM would make a lot of sense yeah that's it hey there uh my name is uh Harry I'm the one of the co-founders and the CTO of offchain labs building arbitrim um and we were kind of we're the optimistic role of building on ethereum having a uh really exciting uh exciting couple weeks since we for the first time surpassed ethereum and the number of TPS and seeing a huge amount of growth there which has been amazing um and in terms of I mean there's so much cool development going on in the layer two space that it's really hard to you know pick and you know talk about any one project but I mean certainly yeah I think the uh certainly kind of seeing the the you know amazing development on the zke VM side that's going on is kind of really uh really cool and it's going to be really interesting to sort of like get to start actually playing around with these Solutions because they they still feel kind of magical in a way uh everyone I'm Brendan I work on uh ZK r d and ZK VM stuff at polygon um and so yeah we're building a ZK VM we're trying to uh basically run the evm inside of zero knowledge proof and sort of get cool security and scalability guarantees uh and I I actually have been really impressed by arbitrim um being like very pragmatic very shipping focused and like obviously doing really good work so so I'm diving in so obviously everybody has like a little bit of a different approach on scaling ethereum we have like the alt L1 approach facing kind of this roll-up Centric future that was laid out by vitalik a couple years ago in your opinion like what do you think um will be the approach that will succeed or will we see fragmentation or will we see um more of like a complementation as the L2 space becomes really crowded whoever wants to start I mean I definitely think that there's going to be a lot of different solutions like it's the interesting thing is that like there are there are really fundamental trade-offs here like in no way do I think that any one technology is going to kind of excel in all ways like I think that they're going to be sort of really fundamental and kind of interesting sort of applications that are kind of have pros and cons and kind of gravitate or more towards kind of one solution or another I think that we will hopefully see um and what I very much look forward to is kind of standardization around things like interoperability um so that even while kind of individual chains might have their differences they'll all kind of be connected and kind of be able to have a very smooth user experience traveling among them yeah so from from my point of view yeah we are here on the stage three projects that are actually taken three different approaches so our approach approach of Aurora is launching multiple EVMS we just launched a thing that is called Aurora Cloud it's something like it's the similar approach of to the approach of polka dot with power chains or Avalanche with subnets right so this is one way in order to scale evm you need to have many ebms right that's that's the typical one typical approach and then there is optimistic Roll-Ups and ZKA Roll-Ups uh there are indeed trade-offs and independently of anything fundamentally evm is extremely hard to parallelize right just because because of the concept of synchronous execution of the transactions right so I believe that in future we are going to see many different EVMS that are implemented in one way or another and yeah I'm absolutely I'm absolutely with Harry that we need to work on the on the interoperability in between this uh in between these different EVMS so we need to work on development um of the pre-compiles the proper communication message messaging in between these two these different EVMS so that is super super important uh because yeah we are here already for many years with clogged ethereum and we really need to fix it off and move towards the mass adoption so uh I I think if I'm understanding the question correctly it's like Roll-Ups versus Alta once is that or only I have the microphone now yeah but but okay but uh so so I I think for me like like one thing that we think a lot a lot about is like uh so we have like the existing internet and like we have this like like proto-internet of value and and like somehow like crypto doesn't feel like the internet because you have to like we have this like fragmented environment and we have to like Bridge stuff and like like there's so much work that is required by the user uh and so I think like if we think about like what the steps are to make the internet of value feel more like the internet of information I I think it's about like sort of providing security to the user in a way that they don't have to like think about it and so I think that like the roll up if we think about like L2 guarantees is providing this like trustless movement and interaction between chains like I actually think that that's really important and I think that that sort will sort of like dominate in the future versus like side chains or Ulta ones or three um I would disagree a little bit with this stuff because uh um with the development of the technology the boundaries in between side chains and l2s are actually diminishing um I will take an example of of the Rainbow Bridge right so the way how the Rainbow Bridge Works in between ethereum and near is that there are light clients of ethereum implements to near there is a light client of near that is implemented on ethereum right and what it means it means that in order to break alt L1 which is near now you need to also to break the the client on the ethereum Chain right it is not the the the forward looking thing but you are able to submit also fraud proofs in the same way how l2s are submitting right so so it means that you have launched a thin that made another L1 and L2 so we can have the the differences in between them are becoming a little bit less strict so I think that the the technology is moving towards the direction of kind of unifying and and you know maybe in a year or maybe in couple of years we are going to stop just talking about l1s and l2s and this is one end on the other end if you will take a look at the validators who are validating ethereum polygon and other blockchains many of them are professional validators they are validating different blockchains it means that their reputation and their losses are interconnected so security of different blockchains though not in code but on the social level it they kind of interconnected so kind of the way how everything develops you can we come in as as you know to a social guarantees of security of multiple chains together this is extremely good for smaller chains that are just only starting so I guess just to I want to kind of disagree a little bit on some of that which is that that fundamentally kind of security is expensive security is very expensive I mean with proof of stake it became a lot cheaper but securing kind of consensus is is not easy and and the Rainbow Bridge is a really impressive feat of Technology but fundamentally one side could reorg and the other side wouldn't be able to react if the security assumptions are violated that the light client bridge is using because fundamentally they're not Atomic and that's kind of one of the Magics of Roll-Ups is that fundamentally they will reorg with the underlying layer they are fundamentally one in the same from the perspective of the consensus mechanism which kind of does create like this sort of much simpler and much cleaner model I think that lets you essentially create more value from one single piece of security layer as opposed to kind of needing to have independent security layers that require each individually to be protected yeah work um yeah so so I I actually think that that's a really like like to sort of echo all right I think that's a really interesting point and I can see that like we building Roll-Ups might be wrong about sort of like the minimal amount of security but I I think one really interesting thing that we've seen in the last year is during the Terra collapse like this you know these sort of Black Swan events that like aren't that uncommon in crypto uh like they had to stop the chain because validator like the security of the validator set was compromised and so like I do think that there there's like some indication that uh to the extent that we can uh like add additional security through fraud proofs or through uh validity proofs like that that is something to like aspire to and to to sort of build in if we want like to aim for fault tolerance maybe I can like expand on that or just like have a like another questions that's sort of going into that direction so if you like the ultimate goal if you want to name it like L2 or like other alternative scaling Solutions is to make ethereum accessible to the next billion users and we will get to a point where like the first entry point might not be ethereum L1 but it might actually be um like one of the scaling Solutions and um the challenge here though is that all of these projects are a little bit differently run than the EF they have different stakeholders VCS token distribution governance models and different security and risk assumptions like how would you what do you think is really important to ensure that we maintain the core values of ethereum thus censorship or systems and decentralization that was a long question sorry I mean I think it's a great question and I think it's something that like a lot of teams are going to be struggling with over kind of the over the coming years like I think that kind of at least sort of our view there is that like it fundamentally comes from the point of of kind of neutrality being to me sort of the most important value that we are building infrastructure and and that should not be sort of opinionated it shouldn't pick favorites it shouldn't kind of be biased it should be neutral and like I think that it it's I think that and I hope that it's kind of a a sort of philosophy that's relatively sort of nicely enforced by user choice and by developer choice that fundamentally kind of when when developers are thinking about what platforms they want to build on the kind of developers want to be where other developers are we build communities and we want and everyone wants to kind of be aligned and sort of having that neutral clean layer is sort of the Common Ground that brings developers together as opposed to kind of pitting them against each other I I think that the like ethereum ethos um of being like a net contributor I think is something that like doesn't like necessarily come naturally to um to people coming from web 2 or sort of like other areas but I I think it's really important to preserve that um and yeah like I wanted yeah I would second um this this idea infrastructure should be uh should be neutral right uh however it seems to me that the way how how the crypto industry is developing forwards is that we are going to see kind of permissioned environments that are going to be launched by some teams in order to comply with the regulations we are going to see fully YOLO environments when we would never know who actually launched these environments and so on the the most important thing is to give an opportunity for everyone to try to implement these different visions and try to execute there and there is one particular thing that I would like to highlight that unfortunately right now is missing and though kind of people starting to pick it up and this is the concept of decentralized front-ends right so so smart contracts introduce this kind of decentralized back-end thing where you are able to change the centralized backends of financials for example financial institutions into smart contracts right auditable transparent and so on and now Regulators are trying to crack down on the front ends for these things right while front-ends are not different from the back ends actually this is also the code in case we will provide an opportunity for people to create decentralized front-ends deployed on the blockchains and then retrievable from the blockchains and then rendered inside of the browsers then it is going to complete kind of the cycle of of everything yeah there should be also the index and layers and additional things blah blah blah yeah but this thing is missing right now and because of this all of the thesis of ethereum unfortunately for people who are willing to go YOLO they are not able to do this right and from my point of view this is something that limits the freedom of the creators and of the developers so once this piece is done then we can say that infrastructure is fully neutral and now and now developers are able to launch their networks the way they want to do it any response to that one okay good um I kind of have I wanna touch upon a slightly different aspect of the whole scaling approach as well so um coming back to the topic of the panel is like scaling it together or like scaling ethereum together and that is one question for you um Alex so Aurora implements the evm as a smart contract on near and near has been labeled as an ethereum killer in the past do you see a contradiction or like how do you see that so at the moment when I joined near like two and a half years ago uh I only heard of the times when somebody was naming near as an ethereum killer I can tell that from being inside of the near they were never this this aspect or this you know thought around killing ethereum and moreover we are kind of executing exactly opposite we were not putting for example inside of Aurora the base token near right the base token of Aurora is eth right so we are trying to give an opportunity for everyone to Simply kind of being able to access additional markets right for for ethereum developers um and this this kind of thesis I most probably evolved somewhere outside of near and maybe some Maxis either near Maxis or you know you know antineer Maxis started to started to throw these thesis in there near ecosystem is extremely friendly we are working together with ethereum a lot we have chats with ethereum Foundation we are discussing the plans and yeah my team is contributing a lot in in eaps actually development we are we are working right now heavily on um additional like the next version of the web 3 RPC that must from our point of view must be changed into into from from the pool the type of RPC into the push RPC uh yeah so we are here to help everyone um yeah I just don't the thesis is separated from the team but I guess there's not only to all all ones we've been chatting earlier a tiny bit like about like one of the articles that Danny Ryan published where he was also like talking about okay what does actually prevent an L2 to become parasitic and actually like sort of fork off and Fork into its own L1 how do you view that Arboretum and polygon um I think I mean so so I I can speak from uh like a little bit of experience we um uh we were working on like a ZK alt L1 uh in 2021 and then joined polygon and were extremely relieved to be working on sort of scaling ethereum uh especially an all-time one but um I I think for us like it it's just really difficult like like why would we Fork away ethereum when it's like providing security for us when it's like sort of like providing this um like this like rich and vibrant ecosystem and Community I I do think that there's a danger of uh l2s becoming like rent seeking or but it's it's really difficult to see sort of like the road uh to that like given that I think the community is very sensitive to it and like some of the network effects which exist in crypto like liquidity uh like they it like liquidity is fairly mobile uh and so I I mean I I hope that that just doesn't happen um yeah consensus algorithms are kind of a pain like that's one of the beautiful things of working on Roll-Ups is that you can like really sort of do a lot of interesting stuff while not having to deal with kind of all of the most painful issues that that l1's kind of the l1s have to deal with and so like I think that like fundamentally sort of it's it's a really interesting balance in the kind of thinking about like arbitrum as a chain versus like I don't it is a chain but it's also something I very much think of as just kind of an extension of ethereum itself and that like it's it's like kind of unimaginable to think about a world where like it wasn't like I think there is kind of like an interesting sort of other part of this which is which is kind of a more complicated interesting issue which is basically sort of traffic moving away from L1 into L2 and what effect that has on kind of the security budget and what effect it might have to have kind of most of the value of ethereum and most of the transactions of ethereum in in L2 which I think is also a world which like is is great and we're basically just making more efficient use of ethereum resources but it is a really kind of interesting thing to think about yeah for sure and I mean I guess there's also like a lot of like open research questions still and like a lot of like unknown dynamics that will probably only play out in like the next couple of years I'm also looking at the time maybe just like as a class the final closing question what do you what would you wish to see in 2023 like I feel like we have like a narrative already sort of like evolving around Sega EVMS but I mean there's obviously a lot more to scaling ethereum what do you anticipate or like what would you wish to see from the entire ecosystem uh I think for so I I think like people have this misconception that like ZK VM is like a silver bullet for scaling and it the reality is is that uh ZK VMS are still bottlenecked by the ethereum execution client and so it would be really cool to see uh like further improvements on sort of throughput and scalability for for clients and you know that might take us further to just go so vitalik I think maybe a month ago or something put out a a really great uh a really great article that was about the stages of roll-up decentralization currently where we are today Roll-Ups are definitely are not yet fully decentralized the plan is there the vision is there but there's kind of work to be done to actually get to the point of of kind of full and complete realization of the dream of Roll-Ups and and that's something that I think uh I want to see really driven at aggressively and that I think kind of a large amount of progress I think will be made yeah that's uh that's pretty cool topic here move towards the decentralized sequencers right and make sure that things are moving the there but for me uh the most anticipated thing um is improvements are on the user experience side for everyone who is working with the blockchain so things like protocol level meta transactions or account obstruction or whatever whatever solutions that are capable of delivering to the users the absence of per transaction payments which is a very um a very like a concept that really destroys the user experience this is something that I'm looking forward to all right cool thank you and I think if we're like perfectly on time thanks so much for coming [Applause] [Music]