The multi-node Ethereum

Transcript

Well, thank you. Um, let's see. Slides. Perfect. Um, wait. Right. Um, yeah. So, I'm Casper, um, one of the co-founders at ETH Labs. Um, it's a new independent, um, nonprofit R&D lab for Ethereum and ETH. Um and our mission is basically to to make Ethereum the global settlement layer of the uh economy and with ETH at the at its center. Um and today I'm going to talk about the multi-node Ethereum future. Um as kind of already introed a bit, lots of things changed um happened over the last year. Uh and I want to kind of talk a bit about the

different dimensions across different at both the technical stack uh social layer and how this all feeds into our mission. Um yeah, so want to take a moment here basically um and talk about um the social layer of Ethereum which I think basically took another big step this year in in terms of decentralizing um the ecosystem. Um the protocol is has always been kind of very decentralized and the social layer has progressively kind of uh decentralized as well. Um and so as these labs we're like one um node in a much larger network of of um uh

stewards here. And so this is basically the multi-node uh future. And this is importantly it's it's not like this new big step. It's kind of this very progressive progress that has been in the works for many many years at this point. Um and just want to take a moment to uh take a look at that. So one example here is that GU um kind of the major execution client was built inhouse at the Ethereum Foundation um and was kind of the only major client for a long time and for proof of stake the Ethereum Foundation then made this strategic decision to actually not build any uh consensus layer clients inhouse. Um and

the goal here really was just basically um decentralizing the ecosystem and instead of building it in-house um helping the ecosystem spin up um such clients uh outside. Um another example here is that solidity language has um kind of moved out of the f and now lives at an independent organization called argot. um the all core devs uh call which is basically the governance um forum for kind of discussing the technical upgrades of Ethereum um now for the first time basically has uh moderators that also sit outside of the Ethereum Foundation with Anscan Nixo and um last

but not least um you you've seen basically many many new independent organizations um uh spun up this year um many coming also out of the Ethereum foundation itself such as Ethereum institutional e systems and also us um at ETH labs and the way I think about it is that it's basically like client diversity but at the social layer um so as I mentioned like we have basically we have many many clients implementing the same spec uh the same specification of the protocol and this is one of the super powerful properties of Ethereum that make it that

makes it very resilient. Meaning that um if one client has a bug, it's not really a problem because there's many others um without sort of any uh implication. And obviously this doesn't come for free like it has costs. Um it means uh there's higher coordination friction perhaps funding is fragmented. Um and it's just like yeah you have many uh kind of different organizations implementing the same thing gives you a lot of resilience and redundance and that's part of what makes also Ethereum so special. Um, and so we have many clients with one spec. And I think basically now what we're doing at the social layer is that we have many orgs

with the same mission to make Ethereum win. And um, it's yeah, I think very powerful. And to kind of just briefly visualize this like we have Ethereum the protocol here um with all these clients um I think 12 if I'm not mistaken um all implementing the spec um across the execution and and consensus layer and this gives us this resilience and now basically we also have this this other layer of like all these organizations um of kind of independent organizations um all kind of working towards this shared mission um of Ethereum. And a bit

later in the talk, I will also kind of add one more ring, but we'll get there in a moment. Um and yeah, basically I think Ethereum has always kind of done the hard but resilient uh has gone down the hard but resilient path. Um and it is really that uh what makes sort of Ethereum so special. And I think as long as we're all working towards the same goal, which I we very much are, I think this kind of increased diversity at the social layer is a very exciting kind of uh development. And again, it's not new, but it's just a continuation of that trend. Um, one kind of I think nice mental model here at this point is is is trend talks

a lot about it is basically thinking of Ethereum as a common. Um the point here is basically that uh anyone can come to Ethereum and start contributing. Everything is happening in the open. Um just last week we've had two new clients being announced. Um it's very exciting. Um you can see the discussions happening online. And you can you see basically the rules by which it's governed all play out. And it's really this um yeah collective um process. And um kind of another mental model and then I'll leave you with the mental models here is this uh kind of quote unquote Amazon test. Um where what's the Amazon test? Well,

basically Amazon is very famous for their kind of ruthless um reluctance to be reliant on um sort of any external organizations. Um so basically they they want to control and vertically integrate very strongly. Um and yeah, basically like Amazon is obviously not building a parallel internet. That would be kind of silly. um but it's refusing to want to rely on Nvidia and so they go at great cost to kind of try and avoid this um and so for Ethereum here the question then is basically can Ethereum be so neutral that entities like Amazon would be happy to rely on it um and I yeah and so I think the EVM is maybe an interesting case study here um

it's been kind of very successful in that regard and you have competing ecosystems like Monut and tempo willingly take on that kind of strategic dependence and that's a pretty big deal. Um, and so here you can really see Ethereum being um, yeah, just this uh, passing the Amazon test if you will. Um, and I I mean it's not black and white. There's questions like I think ideally you would want those chains also to settle on Ethereum rather than spinning up competing uh, L1 ecosystems. Um and I think that's exactly our ambition and the world um we're working towards too.

And yeah, this has to be the ambition in my opinion. Um which kind of yeah brings me to the next kind of point here of like what will this architecture look like of the many chains. Um and I would just want to argue that basically the future is multi-chain full stop. Like in my mind that's kind of a given. And the way I think about it is that the economic activity of the world comes in all kind of um shapes and forms and and basically different economic activity has different requirements. Some might uh want a high high frequency trade on a very fast chain. Some might have privacy

requirements or compliance requirements. And so basically this activity um is met by heterogeneous infrastructure and protocol chains to kind of be able to accommodate this. Um, and so in that sense, I think multi-chain in a in a way is given is a is a given. And so then the the the real question is like how will these all relate to each other? And at the high level, I think uh you can think of it as um there's like two outcomes that are possible. uh on the one hand you could uh live in a world where you have all these independent um L1s that obviously they can talk to each other um but they have to connect uh

basically through these brittle links and we'll talk about it in a sec um or you can have this uh world where you have many chains but they use one shared route of trust um and um just to kind of briefly double click on that Um in this world where you have many independent chains um you they basically each chain has their own heartbeat if you will meaning like their own validator set that build block by block um and if you want to then link these ecosystems um because obviously that's uh yeah um you you would want to have this

economic activity be able to talk to each other. um you always trust someone and you will always inherit the weaker trust assumptions here of the valid data sets and guarantees at best. At worst you have some sort of uh uh multisig uh going on. Um and in this shared root of trust world on the other hand you can have heterogeneous chains that configure in any sort of way optimizing for speed or privacy or whatever else and they still they settle on this one route of trust and they share this heartbeat basically um of the settlement layer. Uh and in this world you end up having shared security and liquidity and um

yeah basically moving across these chains you have the ability to have heterogeneous activity but with a shared strong um uh root of trust basically. Um and in this yeah in in this regard I think also an important takeaway is that you can replicate execution but you cannot replicate trust and so what do I mean here is that these execution environments like think of all these fast chains um they're great at what they do um but it's kind of becoming also commoditized right like it's uh kind of not quite copy paste but like we're getting to a point where you can um spin

up fast chains in a yeah relatively easily. Uh what you can't copy is trust. Trust is something that you build up over many many years uh with network effects um and it's a very high bar to clear um that yeah it's not something you can buy as well right and so um I think to become a root of trust uh for the entire world and I just want to kind of emphasize that like I really mean this like in my mind like I think this is in an incredibly ambit ambitious goal um because when I say root of trust in this context I really mean for the

entire world and historically obviously humanity has found ways to scale trust but it's kind of always found its limits at certain borders or maybe collection of borders but for example basically like you have payment systems today um that that uh don't fully scale globally they they cannot of run into their limits and other ecos other nation states might not want to rely on some others and so um it's it's an incredibly hard thing to actually achieve and I think Ethereum here is in a unique position that uh can kind of

be this shared neutral layer across the globe. Um and so what are the requirements basically for this even to be a viable path right um it's obviously neutrality security uh and then also having the actual right architecture for this um so neutral means basically that yeah you can have kind of friend and foe uh that don't trust each other uh still settle on this shared uh route of trust because it's so neutral. uh you need obviously immense security uh for all the value that kind of will be built on top of it. Um and then you also need the right architecture um in the sense that you need to be built for other like not

only uh to kind of verify and settle your own activity but also that of other chains and so um I think for the first two there's effectively two candidates Bitcoin and Ethereum um and even there you can like kind of argue when it comes to like postquantum and things like kind of diminish finishing block rewards where the Ethereum maybe actually is is uh kind of leading there. Um but I think importantly Ethereum is really the only one that meets all three, right? Like it's the only one that is even playing this game. It's basically playing a different game from everyone else where it's really trying to be the settlement

layer at this very global uh scale that has never really existed. Um and so it's a very very ambitious goal that we're going for here. Um and basically yeah just to kind of conclude this uh graphic here is that um we have this protocol that is built in this very resilient way um by all these different organizations and clients and whatnot. And this is kind of this resilience, this neutrality, this kind of open ability to build is also what then enables um all these different actors like apps from a to unis swap to uh other chains like

arbitum um Robin Hood chain and all these kind of companies and organizations kind of accepting Ethereum as the root of trust and basically passing what we kind of talked about as as the the Amazon test. Um, and yeah, again, like I think the Ethereum path like I think sometimes people forget just how ambitious our path is and how exciting it is. Um, and uh, yeah, we should just go get it. Um, and this is kind of how I visualize that kind of world. um where uh this resilience uh will enable to have like a very strong Ethereum with

ease at its center in kind of as this gravity well where all these other activity happens around it passes through it um and and kind of uses Ethereum as the shared root of trust. um and kind of I just hinted at it like I think ETH plays a very important role here um in that world as well and so let just want to briefly talk about uh ETH the asset as well um and why why it's so special um I think just briefly talking about Bitcoin like it's Bitcoin is also special and it it basically was the first hard digital asset um and I think Ethereum is

um special because of Ethereum's role, right? It's not just a ledger. It's because of Ethereum's um unique ability to be this shared root of trust not just for itself but for the global economy. Um and ETH in that EOS in that world is the maximum by definition the maximally trustless asset and will be at the center of this first truly global settlement layer. Um and that basically makes it in my opinion the hardest programmable store of value and also the bedrock um of the internet of value. Um and yeah I think really ETH uh in this world um can can become the the most important asset in the world. Um and

just to kind of yeah wrap up and like this is kind of my takeaway uh of of of the talk is on the left maybe the more abstract version of like what enables this architecturally this neutrality built through kind of redundancy and resilience and openness um and then sort of the how this looks like uh when it plays out. Um yeah, thank you. Um and I guess we have some time for questions as well.

>> Yes. Question timing. WOOHOO. >> So amazing. Let's talk exactly with that first one. Feel like this root of trust is new form of centralization. Ooh. Can that open up other centralization dynamics again? No. >> Right. Yeah. I mean it's it's a reasonable point, right? like it's sort of um yeah like instead of uh um it's basically like many independent chains versus one shared trust assumption across everyone and I think here really the point is that Ethereum

um is the most decentralized ecosystem and so it's basically relying on that decentralization that even enables it in the first place and so I I think it's it's it's a reasonable question and it makes sense. Um but at the end of the day I think in my mind the reality of the world is that economic activity will interact and will want to talk to each other. And so you have either the path where um kind of you have many independent chains but then you end up kind of accepting worse trust assumptions or you kind of go to the strongest most resilient decentralized root of trust um and and use that as a vehicle to kind of settle

everything. And in my opinion that that is the the better outcome for for the world and users basically. Yeah. Oh, let's go to the next one. Uh, if the argument is that west and others adopt their own system, then how can we scale Ethereum to the global level and build trust, >> right? Um, yeah, I mean, I think um this like I think this is will have to play out over time and I think is already playing out. Um but uh yeah I guess I'm not not quite sure like the question is asking what exactly like

>> yeah they are changes >> then how how can we scale >> I mean I think Ethereum is already in a way there and now we have the infrastructure and now we're at this place where we've built out the infrastructure over the last 10 years and now we're moving to this uh period of adoption and obviously the infrastructure is not done we're laser focused on on uh be becoming postquantum and all these things. So it's not done by any means. Um but I think over kind of now we're in this phase of adoption and I think we'll see these dynamics basically play out over the next years and so they're very critical and we shouldn't assume anything by default. Um but I think yeah an interesting litmos

test is basically seeing these dynamics play out of um certain types of activity both being settled on Ethereum um across across the entire globe. >> Yeah. Okay. Is there a space for more clients? How many clients is too many clients? >> Right. Yeah. Um again like again very reasonable question like these things are not free, right? like we have uh 12 clients, they all need to coordinate and they're like just like like in any meeting when there's too many people like maybe it becomes harder to to arrive at one thing but at the same time the hard path is also the resilient path. Um how many

clients is too many clients? Yeah. Uh I think clients also come and go like you you have the new cool kids and then they drop out. And I think the the the beauty is that um yeah you find this uh very healthy dynamic of of uh uh you need to also prove yourself as a client in a way to to kind of be viable. Um and so it's like a natural kind of process that happens I think >> even like how many node operators you adopt that kind of client. Yes.

>> Yeah. you must to take the advantages of that as well. And uh that was so a broad question but what specifically do you think we need to do next? >> Right. Um yeah, I mean I think uh again I think we've had this infrastructure built out and now we need to be uh in the mindset of bringing the world on chain in the right way and kind of also explain in a much better way what are the tradeoffs and what is the value proposition um and where do these trust assumptions accumulate in in if you do it in different ways. Um and it's yeah it's really about uh bringing the world

onchain in in a way that is uh incentive aligned for everyone. But I think we have a lot of explaining to do um towards in that regard to kind of explain the the power of Ethereum as well.