Bankless - 138 - What's Next? Vitalik Buterin | Part I

Starting point is 00:00:07 Welcome to bankless, where we explore the frontier of internet money and internet finance. This is how to get started, how to get better, and how to front-run the opportunity. This is Ryan Sean Adams. I'm here with David Hoffman, and we're here to help you become more bankless. What's Next with Vitalik Buterin? That is the topic today. We're doing this episode in two parts. The focus is what's next in general, and specifically, what's next for Ethereum? In part one, which is being released today, we're going through the Ethereum roadmap from a technical perspective. This is the short to medium term. In part two, we touch on the social, the philosophical, some of the deeper questions around the thing that we've built called Ethereum and its impact for the world.

Starting point is 00:00:54 That's going to come out on the following Monday. I'm going to give you a teaser of part one before we get into it and also part two. Part one, that's today's episode. We're going to get Vitalik's reaction on the merge. It took eight years to get here. How did he feel? We ask him that question. Ethereum also no longer burns energy.

Starting point is 00:01:15 What are the green arguments for proof of stake? Vitalik makes them. And is this whole sustainability thing really a big deal? We get into that subject. Why did proof of stake take so long? It's in the back of all of our minds. We ask Vitalik that question as well. And he describes the journey to get here to the merge,

Starting point is 00:01:33 to proof of stake. What's left is another question. We review the entire roadmap from a technical perspective. We also ask about some concerns. Is Vitalik concerned about staking centralization, for instance? Is he concerned about Ethereum's future censorship resistance? That is all in part one. That is what you are about to hear today. In part two, we get into the social and philosophical. We zoom out. We talk about Ethereum holistically. Is it a network state? Is it like a country? Is that what we've created a whole digital country? Or is it a Dow, a decentralized autonomous organization? If it is a Dow, how in general should Daos be governed? Should we learn from the corporate world and institute some sort of corporate governance policy? Or are there lessons for us from the

Starting point is 00:02:18 political science side of things? We ask Vitalik about that. We also end with, which I think is a really interesting topic. I've never heard Vitalik cover what's next in AI, in artificial intelligence. We asked Vitalik to speak about AI. Apparently, he just went on a recent trip to Silicon Valley, talked to some of the foremost experts in the AI field. So we ask him, I mean, the basics on everyone's mind. Is AI coming to destroy humanity? Or will there be some peaceful coexistence? Will there be an intersection with crypto? Is it crypto versus AI in a cage match? Crypto representing decentralized interests and AI on the centralized side? Also, this was interesting to me, why Vitalik thinks advanced AI might actually be happening sooner than he previously

Starting point is 00:03:04 thought. We talk about all of those things on what's next. That's in part two. That is coming out the next Monday. David, what are your thoughts in this episode? Yeah, well done on that intro, by the way. That was the longest intro that we've done on the podcast. Thank you. Yeah, what's next is a really good question. And it's definitely the thing that people are asking, the current state of crypto, as there doesn't really seem to be too much. much on the horizon in terms like a narrative. The inflation narrative is gone. You know, COVID's over. The merge is over with. So what is next? But there's plenty of things that are next on the Ethereum roadmap. So this is an updated review, part one, updated review on the Ethereum roadmap just to get

Starting point is 00:03:46 that re-sinct and downloaded into your brain. But I really think it's that second part, the episode that's going to come next week, that I think is the more interesting things. And as the world seems to be accelerating, Ryan, like technology sees. seems to be getting faster. I know ARC is down in price, but doesn't mean it's down in thesis. Kathy's word thesis, that the future is coming faster than people expect it to. I definitely subscribe to, and I think Vitalik does as well. And that start to be shown in some of his conversations with people outside of the crypto space, like you alluded to in the AI space. All of these things are around the horizon. And so we need to be aware of them because if you're not seeing them

Starting point is 00:04:23 coming at you, they're going to blindside you and you're not going to be ready for it. And so understanding these conversations is the intersections of crypto and AI, how DAOs operate and how they should be governed. And what is the long-term network state roadmap for Ethereum and what it may become in relation to that? I think if you understand these questions, you are so far ahead of the rest of the world. And so that's the episode that we try to produce here on this week and next week. I will say in the first part, there is a section where the terms get a little esoteric and technical for you. That's where we go through the roadmap with the merge, the splurge, the purge the verge, etc. You'll hear it when we get to that part. There is another episode

Starting point is 00:05:02 that we did with Fitalic going through this entire roadmap, more in layman's terms, and we'll refer to that. There'll be a link in the show notes for that. Some terms like DVT, which we didn't defined or SSV. This is distributed validator technology, shared secret validators. This is about decentralizing the validator network. Of course, we've also done episodes entirely on this EIP. That's an Ethereum Improvement Proposal called 4844, which is basic sharding, proto-dank sharding. So this scales up the transactions per second on roll-ups that Ethereum can support. One last thing for you premium subscribers, a special gift. We've actually released both part one and part two today for you. So you don't have to wait.

Starting point is 00:05:43 It's all available on the premium feed. You also will have raw access to the debrief where David and I give our thoughts on the Vitalik show. These are our unfiltered thoughts on the episode that was. And there's a link in the show notes. If you have not gone premium, upgraded your membership from free to premium, you can go do that and get access to the full episode now. Well, without further ado, our episode with Vitalik Buterin. But first, a word from our sponsors. ZK Sync is an Ethereum Layer 2 network that is pushing the frontier of high performance blockchains that don't compromise on security or decentralization.

Starting point is 00:06:18 ZKSink has combined the power of zero knowledge roll-ups and the Ethereum virtual machine, enabling developers to build the greatest Web3 projects possible, ones we haven't even seen yet. its killer applications to onboard the world, but crypto killer apps need ZKSink as a platform to build on first. It's generally accepted that zero knowledge roll-ups are the conclusion of crypto blockchain scaling technology, and ZKSink is leading the charge into the final frontier of crypto economics. So if you're a developer who wants to build your app on a future-proof foundation, which gives your users the best Ux possible, check out ZKSink's website at zKSink. And yes, there's also going to be a token, so give them a follow on Twitter 2 at ZKSink.

Starting point is 00:06:57 is bringing crypto-friendly banking straight into your tracking account. With Juno, you can send money from your Juno checking account straight onto a layer two, like Polygon, Optimism, Arbitrum, and they have ZKSink and StarkNet support on their way.

Starting point is 00:07:09 You can skip the ACH wait times, you can skip all the gas fees and go straight from your check-in count to an Ethereum layer two in seconds. Inside Juno, you can buy and sell crypto with $0 fees, and your Juno checking account comes with a metal master car

Starting point is 00:07:21 that gives you up to 5% cashback on your spending. Juno is also giving you $10 cashback on your first cryptocurrency. the deposit and $100 when you set up a direct deposit. This ad just writes itself, so go sign up at juno.finance slash bankless. Rocket Pool is your decentralized Ethereum staking protocol. You can stake your eth in Rocket Pool and get our ETH in return,

Starting point is 00:07:41 allowing you to stake your ETH and use it in Defi at the same time. You can get 4% on your ETH by staking it with Rocket Pool, but you can get even more by running a node. Rocket Pool is the only staking provider that allows anyone to permissionlessly join their network of validating Ethereum nodes. Setting up your Rocket Pool node is easier than running a node solo, and you only need 16Eath to get started. You get an extra 15% staking commission on the pooled ETH that uses your node to stake. You also get RPL token rewards on top.

Starting point is 00:08:08 So if you're bullish e-staking, you can boost your yield by adding your node to the decentralized rocket pool network, which currently has over 1,000 independent node operators. It's yield farming, but with Ethereum nodes. You can get started at rocket pool.net, and you can also join the Rocket Pool community in their Discord. You can find me hanging out there sometimes in the chat, so I'll see you there. Bankless Nation want to introduce you yet again to Vitalik Butrin. He is a crypto economic researcher. He's a builder. He is a philosopher. This is the first time he is appearing on bankless in a post-merge world. He also has this fantastic blog on his website. You got to check out. I read whenever an article is published at Vitalik.ca.

Starting point is 00:08:46 And there's a book out called Proof of Steak, which is a compilation of many of those blog posts. It just came out earlier this month. He also happens to be the guy who wrote the Ethereum. white paper. Vitalik, welcome to bankless, post-proof of work, post-merge. How you doing? Great. I'm, you know, obviously, we're really happy that this big, long, eight-year journey finally at its completion, you know, really happy that proof of stake is here and it's fully here. It's running. It's actually powering the Ethereum chain. The whole transition seems to have happened even far more smoothly than pretty much anyone, including myself, expected. And, you know, just happy that that whole chapter and, you know, the chapter of all of the

Starting point is 00:09:40 various trolls on the yet Twitter doubting whether or not proof of stake is even possible is over in the Ethereum ecosystem now gets to, you know, move over and start focusing full speed ahead on the next thing. Yeah, so what are the trolls going to focus on after the merge is done now that we have proof of stake, because they're not going to just, you know, wave the right flag, right?

Starting point is 00:09:58 They're going to move on to the next thing. That's a good question. Yeah. I mean, I think if I was a troll, I would probably be waving one of two flags. So, like, one of the flags you can wave is obviously that proof of stake is actually bad.

Starting point is 00:10:09 And the other flag you could wave is that sharding will never happen. Right? And, you know, if you want to wave the flag of sharding never happening, then, you know, there are arguments that you could make about how,

Starting point is 00:10:20 like, oh, you know, the peer-to-peer networking of data availability sampling is actually complicated and that stuff is not going to happen and actually Ethereum is going to hit scalability bottlenecks soon and you know blah blah blah blah blah but like betting against the Ethereum ecosystem's ability to produce technology is something that clearly does not have a very good track record so I think yeah I know if I was a troll I would definitely kind of stick to the ideological side and just try to argue that proof a stake is bad somehow

Starting point is 00:10:52 Vitalik, you said that the merge was even smoother than you expected. So I'm wondering if you could just put us into the mind of Vitalik as we were going through the merge because this has been, I mean, the beacon chain launched, of course, in December of 2020, but proof of stake as an idea has been along even longer than that. And efforts on the beacon chain has started before the beacon train launched, of course. So this has been a long time coming. So how does it feel to be in the post-merge world? And what was it like to be watching this thing go through?

Starting point is 00:11:22 And also what was it like to be watching the world watch Ethereum in this moment? Yeah. So I think one of the reasons why we're all surprised that the merge went really well is that none of the test debt merges really quite went that amazingly smoothly, right? Like the most recent merge before the real one, I think it was, I think the Robsden merge. It even did not finalize for about an hour because slightly more than a third of. of all of the nodes did not make it through the transition, and there was some code glitch and they had to update.

Starting point is 00:11:58 One of the merges before, like, it only had something like 80% online. Like, there was always some client combination that just did not survive the transition. And, you know, there were various bugs. I think one of the reasons why that ended up not carrying over to the main net, and the main net ended up going great, actually, is that a lot of the bugs were not really bugs with the implementation. There were bugs with the settings, right? as the test nets, they all tend to have, you know, various weird and like settings choices

Starting point is 00:12:26 and people are just less used to connecting into test sets and making sure they're connected in the right way than they are with the main network. And so we ended up having a lot of bugs on that side. But if you're just watching as a more naive observer from further away, you might think like, oh, you know, every one of these test that merges had at least a bit of an issue. And, you know, surely the main one is going to have even more problems just because it's a much bigger ecosystem and it's all of these thousands of people

Starting point is 00:12:56 and we don't have as good an ability to immediately reach everyone. But, you know, no. And actually the main net merge just went incredibly smoothly. So the process of waiting for the merge to happen, you know, it was in some

Starting point is 00:13:12 ways surprisingly uneventful. Like you know, there was a call. There was a Zoom call. And everyone will, or at least lots of people were on that call, and I'm sure lots of people were on other calls as well. And we were just, you know, talking about what the merge is, talking about what the merge means for the world. And then, you know, after we got through a few speeches, it's like, oh, wait, okay, now we're less than five minutes away. And then everyone just started waiting for the countdown. And eventually, you know, the terminal block, which is the block that surpasses the total

Starting point is 00:13:48 terminal difficulty was hit. So that's the block that surpassed the threshold for basically how many hashes the proof of work chain is supposed to do before it turns off. And that was hit. And then immediately after that, the first proof of stake block happened, and everyone celebrated. And then we waited, you know, 12 minutes for it to finalize. And then we finalized. And then everyone celebrated again. And like, that was basically it. That's really to. This is not so simple. Yeah. Easy, right? We should have done this years ago. But Vitalik, to David's kind of second question, part of, I think, the interest from the Ethereum

Starting point is 00:14:24 community is watching the world react to the merge, right? And you were mentioning the trolls earlier, and, you know, if you were a troll, what you'd do is probably, like, argue that proof of stake is a bad idea entirely. As I was, like, looking at the world's reaction, there was one thing that kept popping up in mainstream media. And that was this energy reduction narrative, this decrease in electricity cost. And I do think that trolls, at least from a mainstream perspective, right, are going to have a hard time arguing with mainstream that proof of work is superior to proof of stake from a, you know, a green perspective, from a climate change perspective. But what's your take on this? The whole ETH energy reduction thing, do you think that is overrated? Do you think

Starting point is 00:15:09 that it's underrated? How much does this actually matter in the scheme of things? Yeah, I mean, I think there's a couple of levels at which you can look at that question. One is the object level issue of like how important the climate issue is and how big of an impact the switch to a proof of stake is on the climate, right? So I think my view on the climate issue is kind of fairly boring and moderate, which is like it's not a literal existential risk in the sense that, you know, it's not going to kill anywhere close to the number of people that like, you know, something like World War III would. But at the same time, it's like a really crappy and terrible thing. And if you look at, you know, the studies of what worst case global warming

Starting point is 00:15:50 could lead to, it's like reducing the GDP of some of the already poorest countries in the world by like a factor of two or three, right? So it's a, you know, quite a big deal as far as big deals that actually exist in the world are concerned. And I think anything that we can do to try to move beyond our carbon-heavy present is definitely a good thing. And, I mean, obviously there are all these arguments that proof-of-work proponents like to trot out that, like, oh, you know, actually this is incentivizing green energy production, or actually this is using energy that would not be used for other purposes, because proof of work is really good at, like, finding the energy that would not be useful

Starting point is 00:16:37 anywhere else. But, no, generally, when I look at those arguments from far away, they always have this kind of feeling of self-serving bias. It's like it just feels like the algorithm that these people are running is like not like, oh, you know, let me neutrally figure out whether or not proof of work is good or bad. It's like, oh, let me try my best to come up with arguments for as, you know, like soldiers in this rhetorical war of, and, you know, it helped me justify the cause I already decided a long ago I stand for. You're saying it's a convenient rationale. Like, it was really convenient that proof of work people think that proof of work is green.

Starting point is 00:17:13 Totally. The arguments to me feel like, imagine if someone from Philip Morris came to you and said, like, hey, you know, Philip Morris is actually an incredibly ethical company because we're contributing to the incentive to cure lung cancer. Right. Like, this is basically what happened. Slow claps. Yeah. And then, you know, there's all of these like methane, like, flaring things. And I think the challenge with that. is that, like, you can always identify, you know, like these really specific situations where there's something really unusual about the energy market. But, like, the more you zoom out and, like, think about this in the long game, it's like, you know, no, if you consume energy, then that increases energy consumption and energy consumption for the next while is going to contribute to climate issues. And, you know, there are going to be, like, these little discrepancies in both directions, but the little discrepancies over time, they're just going to, I think, average out,

Starting point is 00:18:16 but just like the one big fact in the center of it is the thing that's going to dominate. So, you know, people argue that, you know, proof of work is like good at sort of grabbing those little nooks of energy that nothing else is because proof of work has this special property that you can do it from anywhere. But at the same time, proof of work has this property that if you want to make good use of your hardware, you have to keep it running 24-7, right? like this concept of, oh, I'm going to only turn on my hardware at those specific times when there is spare electricity and otherwise I'm going to shut it off. Like, that's total BS, right?

Starting point is 00:18:48 Just because like the hardware cost is a big portion of all proof of work cost. And so if you want to be competitive, you know, you have to use your machine and you have to make good use of your machine until Moreslau causes that machine to expire and requires you to replace it with a better machine. So, yeah, in general, I'm, I guess, on the whole, not convinced by any of that. And I think, you know, reducing energy consumption is good and great. And I'm immensely glad that we as an ecosystem have done it. So that's the first level of the question, right? The second level of the question is, I guess, from a kind of more selfish Ethereum ecosystem point of view,

Starting point is 00:19:31 like, obviously, the fact that we're not contributing to breaking the end. environment anymore is good for Ethereum from a narrative perspective. And I think there are lots of signs that this is actually true. Like I've personally talks to a bunch of people in, you know, corporate and government contexts and just lots of different places that are like, hey, we want to do blockchain stuff, but there's just too many people within our organization that are really don't want to be contributing to, you know, making the planet blow up. And, you know, if the merge completes them, like a lot of them have even said, like, you know, yeah, I'll be much more excited about using Ethereum, and they think a lot of that's true. I think the subconscious line of thinking

Starting point is 00:20:15 that at least some of these people that I'm dismissively referring to as trolls as referring to is trying to make is like, oh, you know, this environment thing is like a centralized world economic forum ideology. And chances are, if you're into that stuff, then you're into centralized world economic forum stuff generally. And so like most of people who care about that are going to also care about, you know, making sure we have global financial surveillance because they think that's the only way to prevent crime or whatever. And so why are you even trying to go after those people? And so, you know, the people that crypto should be going after are like these sort of steely based people who understands that like that centralized world economic forum ideology is total crap. And like actually the environment's going to be fine. and like methane is great for the environment and everything's going to be solar within five years anyway. And so like I think there's this sort of implied subconscious viewpoints that, you know, really there's just two kinds of people one or the other. And the world economic forum, bug eaters are a lost cause. And so you might as well go after the base school people and the

Starting point is 00:21:23 base school people don't really care about the environment. And so you might as well use proof of work. But I feel like that viewpoint is just false. Like, I've just, I don't know, I've, you know, talks to lots of people. I've been to, like, 53 countries now and, you know, chatted with all kinds of people in various industries and governments. And, like, I don't think it's true that, like, there's this binary. And I think there's plenty of people that, like, really do deeply believe in the, you know, the freedom thing and the open source thing and the, you know, decentralized, like, censorship-resistant, global neutral platform thing. also do care about not blowing up the planet at the same time. And like I feel like the Ethereum community in general is probably sort of more, you know, on average, tilted toward that

Starting point is 00:22:16 middle, though there's definitely people on all sides of the spectrum. So I mean, you know, it sort of ties into the social and cultural bet that I think Ethereum is making that some of those kind of hardcore proof of work proponents are taking the opposite. side of the bed. So, you know, I'm happy with the side of the bed I've made and I, yeah, plain plan to keep making it. So we'll see. By the way, for YouTube viewers, the reason David and I have a smile in our face is because we just noticed Vitalik's login name is Adam Back, which is maybe one of the CEOs. A big proof of worker. Yeah, the Philip Morris is fighting for proof of work? But Vitalik, can I ask you, is it material? Like, is the difference between

Starting point is 00:22:57 Ethereum switching from proof of work to proof of stake? Does that have an actual material? impact. I think you tweeted out a Justin Drake figure, and I'm not sure where he got this number, where he sourced it, something like a 0.2% reduction on worldwide electricity consumption, which feels very material. What kind of numbers have you seen? And are these actually material numbers? Or do you think it was like much ado about nothing all along? I mean, the 0.2% figure definitely roughly lines up with other statistics that I've seen. Right? Like, we've seen statistics that, like, Ethereum mining uses a about as much electricity as my thank you as either half of Singapore, sometimes, or Austria

Starting point is 00:23:37 sometimes. And like, those countries are about 0.1% of the world population, but they're a wealthy 0.1%. So I think if you just look at all those different estimates, like somewhere vaguely in the 0.2% range does make sense. I mean, the one argument that I have seen people make is that, like, oh, that hash power is not going away. It's just going to go to Ethereum Classic. But, and that's clearly true to some extent in the short term, but they just obviously can't be true on the medium or long term, right? Like in the medium or long term, like we've basically reduced the amount of reward that gets paid out to people who participate in like GPU style of proof of work by probably a factor of 10, right? And so in the medium and long term, the amount of effort going into that style of proof of work is also going to go down by a factor of 10. Yeah, yeah, we've certainly reduced the incentive.

Starting point is 00:24:28 therefore it should follow that there will be less miners in the future. We were talking to Danny Ryan and Tambako yesterday on the live stream, and Danny gave out this metric that I want to run by you, where, yes, we killed 0.2% of global electricity consumption, but he said that if ether price went 10x, then it would have actually have been 2%. And I'm wondering if that's how the math checks out. And so, like, really it's like, yes, we killed it at 0.2%.

Starting point is 00:24:54 But if we believe that Ethereum is going to be the global settlement, of the future of everything, then the future energy consumption that we eliminated is actually much more significant and is actually the bigger story. I'm wondering if that math checks out and what you think about that. Yeah, I think I fully agree with that. I mean it obviously all depends on your projections for just how big the Ethereum and crypto thing is going to get. But in the bulk cases where Ethereum keeps becoming a more and more significant thing, then yeah, absolutely. it's a much bigger savings than 0.2%. But, like we alluded to earlier,

Starting point is 00:25:32 the road to proof of stake has been long and winding with a few dead ends, a few backtracks, and then moving forward to where we ultimately got here today. But for the listeners that came in during the 2021 bull market or later in 2022, they might not know why Ethereum proof of stake took so long. So I'm wondering if you can kind of just, for the listeners that weren't here in the eras of 2017 to 2020,

Starting point is 00:25:56 kind of give a recap to the journey thus far. Like, why did proof of stake take so long? And what were the dead ends and the roads that we almost went down? What were that path like? Yeah. So the roadmap really did change a huge amount over the last few years, right? Like, in 2017 to the beginning of 2018, there was this whole Casper FFG concept where we would basically build the proof of stake system as a smart contract on top of the proof of work chain.

Starting point is 00:26:30 And then step one would be to kind of get that running. And then step two would be to have this kind of separate charting chain and then we would really slowly migrate the thing. And then before 2017, the roadmap was something even more different. It's like, hey, let's start a new chain from scratch and then basically just completely retire the existing one and demand all of the applications to actively move over. And then in the middle of 2018, Justin Drake made this post on a pragmatic BLS aggregation, where he argued that the Ethereum blockchain should make really heavy use of BLS aggregation in order to support as many validators as possible and be able to be a more decentralized chain. and that I think ended up being a really a prescient decision, ended up really simplifying the architecture and helping us a lot. But one of the consequences of that meant that we would not be able to actually build the thing as a smart contract,

Starting point is 00:27:33 and instead it would make much more sense to build a new chain, and then find some way to migrate the Ethereum system over to it over time. So work on the beacon chain started. Now, then simultaneously there was this parallel. all track of working on different ways of how to transition existing Ethereum into the Beacon Chain, right?

Starting point is 00:27:54 So in 2018 to 19, there was this concept of execution environments where it's like you can think of them as being sort of like roll-ups except much more enshrined and where the chain would not just provide data.

Starting point is 00:28:10 It would also provide a minimal execution layer. And the idea is that the existing Ethereum system would be one of those relaps. I think back then we were also really contemplating trying to upgrade from the EVM to WebAssembly. And that's one of those things that we ended up dropping over time, right? Like one of the unfortunate things that I think was necessary as part of the merge was that there were a lot of dreams that kind of got dropped and the ambition did end up decreasing a bit. So the desire to improve the VM ended up kind of dropping off. And part of that was definitely just

Starting point is 00:28:45 us noticing that out of all of the East Killers, the only ones that really had significant success were the ones that just accepted the EVM and the ones that tried to argue that, oh, the EVM is total crap and will replace it with something much more performance and it better. Those,

Starting point is 00:29:01 maybe with the exception of Solana ended up kind of falling off mostly. There's obviously the question of like just how many other things we could use the merge as an opportunity to reform. And I think we did managed to reform some things, but a lot of other things did manage their, like, basically stay as

Starting point is 00:29:21 they are. I mean, right now, the Ethereum clients are still processing the pre-merge chain, right? And there was a desire, I think, to use the merge as a reset opportunity and allow Ethereum clients to not have to process that history. And I think that still will happen with EIP 444, but that's something that we're going to have to wait a bit longer for. So there's just a lot of these little things on which the ambition of the project decreased a bit, but it slowly turned into a more and more realistic thing. And then, you know, finally we settled on a design and we even settled on a very simplified design for the merge,

Starting point is 00:30:02 and then we went for it. And the design that we finally settled on, the one that we know of today, which is now current Ethereum, do you think if we had, like, waited and thought about it more and done some more research, we would have found a different design, or do you feel like the design that we settled on is like the final logical conclusion of Ethereum proof of stake? So there's two questions there, I think.

Starting point is 00:30:23 One is what is the better long-term form of proof of stake? And the other is if the research team and only the research team had had 500 years to spin its wheels, could it have come up with a better way of managing the transition? So for the first question, I think the answer is like, no, we are currently very far from what a truly optimal Ethereum and purestake would look like, right? And, you know, recently we've been having these discussions about single slot finality, which is just a big redesign of the consensus. There's all of these designs around like Merkel trees and different ways to implement

Starting point is 00:31:02 charting and different ways to implement deposit in astrology and all of this stuff. So still a lot of improvement work to go. And I think we do still have the opportunity to take. keep working hard over the next few years to make that improvement actually happen. But then on the question of was the merge as it was done the best way to do the transition, I would say maybe yes, maybe no, right? So I think if I could like give myself a research time capsule back to 2014, I probably would have said, don't bother with the Casper FFG design and instead implement a simpler

Starting point is 00:31:40 version of chain-based proof of stake first, and so that you can move to that in, like, maybe around 2018, and then, like, properly, yeah, improve proof of stake later on a more relaxed schedule. Where, like, I do think that we set ourselves goals that were a little bit too high for the proof of stake that we moved to for the first round, and we even ended up not accomplishing in some of those goals, right? Like, there's a lot of these security issues that arose from the specific design of the KASPRAFG system in the way that the LMD ghost and Fork Choice side and the BFT consent side, like, interact with each other. And that's one of the bigger things that we're going to have to resolve with a single SWAT finality change. Yeah, basically, I think create a simpler

Starting point is 00:32:28 and the less powerful group of stake and move to it earlier would have been one of the bigger changes that I would have made. But the format of the merge, I think, was excellent. Like, I I still don't think that there is a better merge format that we could have come up with. And I definitely hope that other chains like Zcash and Dogecoin are probably two proof of work chains. Who is core deaf communities I talk to a lot and I'm pretty friendly with. I hope they learn from the merge and they move over to proof of stake over the next four years or so. So Vitalik, if you could go back in time, it sounds like one thing you might change is you would have simplified the proof of state kind of conception and maybe got it. it in the roadmap a little bit earlier, deployed a little bit earlier. I mean, as you mentioned,

Starting point is 00:33:14 we're here now, and that's fantastic. But we did have some false starts and some meanderings along the way. And I think for everybody involved, it took longer than we thought, I mean, eight years. And proof of stake had been in the Ethereum social contract from the beginning. Yet there was also a benefit. And maybe that benefit came in the form of like proof of work distribution, right? So more ETH stake was distributed over. a greater number of entities, I guess we would assume or suppose, with proof of work distribution. And so that was a benefit. I mean, overall, is there anything else big that you would go back and change? Are you satisfied with how this turned out? I mean, definitely, had we known the results

Starting point is 00:33:58 of the meanderings, it would have been great if we could just skip to the results immediately. right so the design of the proof of stake system and all of the choices of like should it be a chain or a smart contract or something else has one side of it the other side of it is the layer two scaling realm nap like i think optimism and arbitram could have been finished and fully trustless by now had we known from the start that roll-ups started the way to go and you know we had not spent as many resources kind of going down the state channels and plasma rabbit holes like i think going down those rabbit holes was useful from the yeah point of view of like inventing ten thousand ways not to make a light bulb. And I do think that state channels make a lot of sense in some specific use cases, but I definitely think that we made a false step in, like, at first, really, yeah, hoping that that would be the way that Ethereum applications would scale in general. And then with plasma, like, there was this hope that we could find some way to generalize it, but that ended up just not working. And then two years later, I think we ended up learning that, like, well, actually pretty much nobody wants,

Starting point is 00:35:02 a scaled Ethereum that's just a payment system. Everyone who is in Ethereum already wants something general purpose. And so there's no point in making even a roll-up unless it's a general-purpose one. Fidelic, you know we use this going west metaphor. We're on a journey into the frontier. There's a dark fog of war ahead of us. But some of the guests on the podcast are people that have scouted into the unknown and come back to tell us what's in the future.

Starting point is 00:35:28 And so as somebody who knows the history of Ethereum very, very well, and also understands cryptography and crypto economics very, very well. Can you, like, place us in history or on, like, you know, a map or a journey of where we are going and where Ethereum is? I remember you saying something along the lines of post-proof of stake Ethereum is about 55% completed. Could you talk about, like, where are we in Ethereum's trajectory and where are we in crypto's relevancy to the world? Sure. So I think one way of looking at it from a trajectory of the product. call point of view was to look at the roadmap from last year and basically ask how much progress

Starting point is 00:36:08 has been made, right? So if let me just Google this now and let's check Ethereum roadmap on Google images and I'm sure it'll be up there, there it is. So just looking at the five tracks, right, we can go through the tracks one by one. So the merge is completed now. The things in the merge section that are not completed are, one is the post-merge hard fork with withdrawals, right? That's obviously the one of the next big priorities right after the merge. It's a pretty simple hard fork, and like we basically can know what to do. There's a spec. I think the main debate at this point is to do it at the same time as 48444 or do it earlier and then do 484 a bit later, but otherwise, it's like, we're not far from it. Distributed validators, I think,

Starting point is 00:37:04 still progressing. Actually, haven't checked up on that team for a while, but I imagine they've made quite a bit of progress, though it's not quite 100% done yet. Is that DVT from Oble? Is that what that is? Yes, it's a DVT, aka SSV. SSV, right, right, right. Yeah. Cool. Right. So, and then if we look in the longer term extras, single secret leader election, there has been some great work on the cryptography there. The cryptography itself has been published. Single SWAT confirmations, or what we now call single SWAT finality, that's a bigger item. And I think it's an item that has been really moved earlier in the roadmap, because people just recognize the value and the importance of it much more. And I think this is one of those things that we are going to have to kind of have a big public

Starting point is 00:37:52 discussion with the community at some point because there are huge benefits to say a lot of suede finality but at the same time there are some costs so like for example some compromises either to the 32eith validator balance property or to the level of economic finality that we can expect and just like different tradeoffs of what kinds of things would have to be sacrificed if we want to be able to actually have finality in, like, say, 32 seconds or something like that. And then better signature aggregation is just a subset of single-s-lot finality at this point. So that's the merge. Basically, the entire left half of that box is finished and the right half of the box is still to go, which is great. The surge. So there,

Starting point is 00:38:51 I think things have obviously been reshuffled a bit, right? So back then there was the whole 4488 versus 4844 stuff. And I think lately 4844 has been winning. And it's basically fully specified. It's just waiting in implementation. And so that would be, I would call 4844 the equivalent of basic sharding. though I think one of the interesting things with dink sharding, which happened after this roadmap, is that dink sharding really moved us away from the concept of having discrete shards, like shards as the usual units. Instead, we're just moving to this more amorphous concept of like there's these big data blobs, and there is a distributed procedure for verifying the data blobs.

Starting point is 00:39:39 But, you know, otherwise there's just blobs and there isn't like a hard point where one of the one section of a blob ends in another section of a blob begins. I would say the short term called it expansion box got short-circuited a bit. The basic sharding box, that green bar is probably a bit further. And then the basic sharding and data availability sampling, like that's a really big research area. So, you know, it's still quite a bit of stuff left to go. Then the verge of birkle trees. That one is interesting in that I think there's been a lot

Starting point is 00:40:18 of progress on implementing verical trees. The main sticking point at this point is that the transition from our existing tree structure to a vertical tree is going to be something that's like a big engineering challenge to implement, and there's still debates about how to do it. And I think in general, it's been deprioritized a little bit relative to getting scalability out, because scalability is just like so incredibly important to core to Ethereum. The Purge is, so history, expiry, P4444 making some good progress. Banning self-destruct, I think, at some point we just have to pull the trigger and say we're going to do it. State expiry.

Starting point is 00:41:00 State expiry has been deprioritized a lot because of Proposier-Builder separation. Basically, if you have PBS and you have statelessness, then the number of actors that actually have to hold the entire state is really tiny, right? regular validators don't have to hold the state because regular validators would just have to verify other people's blocks. They're not creating their own blocks. And so that's something where the order of operations also got reworked a bit, right? But, you know, which is a good thing. Like, I think state expiry being deprioritized that does give us a lot of freedom to figure out all the other stuff first.

Starting point is 00:41:39 And then all of these other things about just like making the Ethereum protocol cleaner and like getting rid of RLP, cleaning up the block structure. Like, there's people who, like, I think wants to do them, but they're kind of low priority. And I think Ethereum is going to be in a place where it's able to do these sort of more aesthetic things that just have to do with making the protocol simpler and look cleaner, but probably would need another, like, a year or so of firefighting. Right. Look, this is one of the big things that big changes that's going to happen with Ethereum

Starting point is 00:42:12 protocol development over the next five years, which is. is really moving away from firefighting mode and moving from, you know, we got to get this fast, fast, fast, the community's angry at us. You know, people really, really want this stuff now to this kind of more calmer, a trajectory of, you know, really valuing stability and sustainability. And I think the merge already, like, really reduces the pressure on the Ethereum ecosystem a lot. But I think a successful transition, at least to proto-dank sharding, is going to be the rest of the way. And then, you know, full-dank sharding can happen on its own time.

Starting point is 00:42:50 Cleanups can happen on their own time. Virgil trees can happen on their own time. And so the pressure on the ecosystem to do those things quickly can at least be a bit lower. And then... Vitalik. Yeah. Before we get to the splurge, can we just kind of summarize for folks? And I'll remind bankless listeners that we did an entire episode on the Ethereum roadmap

Starting point is 00:43:10 where we went through the merge, surge, verge, purge, and splurge. in extreme detail, so make sure you catch that episode. But high level here, the functions of these various categories, the merge is about transitioning to proof of stake. And that's pretty much done. There are a few things to tie off, but we're getting real close. It's almost done. The surge is about scalability of the Ethereum network, Ethereum economy, in the form primarily

Starting point is 00:43:38 of roll-up scalability. And the big functionality to watch there is the protodendium. dank sharding EIP 4844, which is maybe upcoming. Don't know when yet, but that's the search. And then the verge is about statelessness, and that helps us further decentralize the network, allow individuals to run nodes. And so does the purge. It increases the, it eliminates some of the technical debt, allows us to run nodes better.

Starting point is 00:44:07 And when we have the merge, the surge, the verge, and the purge, that's kind of the base functionality. I wanted to just pause there and reiterate that before we get to the splurge, because it feels like the splurge, that's the goody bag. Those are all of the extra things that we'd like to implement, but the core of this thing from a roadmap perspective is probably the first four categories. Is that, you know, approximately, right? Is that in the ballpark, Vitalik? Yeah, I think that's fair. I mean, I do think that one connotation that it's important to get away from is the connotation that

Starting point is 00:44:40 the sporge is or anything in the sporge bucket is less important than the other four. I think there are, the individual items are smaller, but a lot of them are quite valuable. So like, for example,

Starting point is 00:44:53 account obstruction is very valuable. PBS or some kind of stabilization of the MEV situation is obviously super valuable. And with the discussion of censorship resistance, it becomes even more important.

Starting point is 00:45:10 EVM improvements are, like, that's a smaller one, but, you know, like, EVM big into arithmetic. It can be a big deal. Basically, let us do a lot more cryptography inside of the EVM and probably get half the benefits of what we wanted to do with Wasam back in the day. ZKK snarking everything is, like, actually a really big deal. So, like, I think for if I had to choose, between giving up the verge and giving up Ziki Snarks, I would probably

Starting point is 00:45:41 give up on the verge. Wow. Whoa. Yeah, like the one of the reasons why is because like Ziki Snarks could give us a sort of poor man's verge where basically we don't bother with verical trees and we solve the

Starting point is 00:45:54 witness size problem by snarking the witness instead. And now that approach like, it's very ugly. It has a lot of weaknesses. It'll require of far larger amount of like work for the snark provers than would otherwise be necessary

Starting point is 00:46:11 but like it at least kind of works and gets a stay objective and snarking in general is just incredibly valuable for making it easier for people to verify stuff on chain yeah so that's something that's really valuable like iziki synarchs are I think I've said this before in my opinion

Starting point is 00:46:29 they're as big a yet and as important to technological breakthrough as blockchain so wow and we're going to start seeing them more and more over the course of this decade. But, you know, it still would take a while until the technology gets to full maturity. So, you know, fingers crossed and hoping that it just keeps on progressing peacefully. And we'll see. Lens Protocol is an open source tech stack for building decentralized social media applications. It is the new era for social media.

Starting point is 00:46:54 We all have toxic relationships with our web two apps. We want to break up with them, but we can't. These applications own our digital lives and all the relationships that we've made. We need to break through to a new paradigm of social network. networking applications that we control rather than them controlling us. Lens isn't a social media app, it's a protocol to let a thousand web three social apps bloom. Lens is a permissionless and transparent social graph that is owned by the user. In crypto, we say not your keys, not your crypto, and on Lens, we say not your keys, not your profile. With Lens, your followers go with you to whatever social media application you want to use, and instead of being trapped

Starting point is 00:47:26 by an algorithm chosen by that app, Lens lets you choose the way you want to experience your social media. Lens is the last social media handle that you'll ever need to create. So in order to get started, there is a secret code word in the show notes. Enter that code word in the Google Forum linked, and you'll be well on your way to entering the world of Web3 Social. Arbitrum is an Ethereum layer-toe scaling solution that is going to completely change how we use Defi and NFTs. Some of the coolest new NFT collections have chosen Arbitrum as their home,

Starting point is 00:47:52 while Defi protocols continue to see increased liquidity and usage. You can now bridge straight into Arbitrum for more than 10 different exchanges, including finance, FTX, Whoobie, and Crypto.com. Once on Arbitrum, you'll enjoy fast transactions with cheap, allowing you to explore new frontiers of the crypto universe. New to Arbitrum, for a limited time, you can get Arbitrum NFTs designed by the famous artist, Ratwell and Suoy for joining the Arbitrum Odyssey. The Odyssey is an eight-week-long event, where you can play on-chain activities and receive

Starting point is 00:48:19 a free NFT as a reward. Find out more by visiting the Discord at discord.g.g. You can also bridge your assets to Arbitrum at bridge.arbitrum.io and access all of Arbitrum's apps at portal.arbitrum.1 in order to experience Defi and NFTs, the way it was always meant to be fast, cheap, secure, and fiction-free. The Brave browser is the user-first browser for the Web3 internet, with over 60 million monthly active users. And inside the Brave browser, you'll find the Brave wallet, the secure, multi-chain

Starting point is 00:48:46 crypto wallet built right into the browser. Web3 is freedom from big tech and Wall Street, more control and better privacy, but there's a weak point in Web3, your crypto wallet. And most crypto wallets are browser extensions, which can easily be spoofed. But the Brave wallet is different. No extensions are required, which gives Brave browser an extra level of security versus other Wallets. Brave Wallet is your secure passport for the possibilities of Web3 and supports multiple chains, including Ethereum and Solana. You can even buy crypto directly inside the wallet with

Starting point is 00:49:13 ramp. And of course, you can store, send, and swap your crypto assets, manage your NFTs, and connect to other wallets and defy apps. So whether you're new to crypto or you're a season pro, it's time to ditch those risky extensions, and it's time to switch to the Brave wallet. Download Brave at brave.com slash banklists and click the wallet icon to get started. So that is the roadmap progress graphic, and we've looked at that kind of lens on what Ethereum looks like next, what Ethereum looks like moving forward. There are a few other lenses we might want to put on this topic and this question, Vitalik, before we tie it off. One is the economics, maybe the economics of staking. Are they now set in stone the economics of Ethereum, or are they, you know, scheduled to

Starting point is 00:49:55 change at some point? If change, what sort of changes might you envision? That's definitely a good and very important question. So changes to the economics of staking that I could see happening. One of them might be that I think there are ways to make the deposit and withdrawal cues faster, at least in the normal case. So things like allow a huge amount of deposits and withdrawals to happen if the chain finalizes. So that will just make the experience for validators easier. and that will also reduce the incentive to participate in stake pools,

Starting point is 00:50:34 and it'll make it easier to have smaller and more decentralized stake bowls, for example. So that's one. Then number two would be changes to the MEV architecture would obviously affect the economics of staking. Like even today, the MEV revenue is a pretty significant portion of staking revenue. right and like there is some economic benefit to participating in MEV boost but that's something that will obviously

Starting point is 00:51:04 change a lot and especially if the Ethereum Protocol includes some form of enshrines proposal builder separation and then if Justin gets his way and we also add MEV smoothing which is like a change that basically

Starting point is 00:51:20 forces MVV revenue to get redistributed to all of the ECA validators instead of being concentrated in one of them, then that will reduce the variance of staking revenue, which also reduces the incentive to be part of a stake pool. So a lot of little changes like that, I think, are going to happen as a result of some changes to staking that can happen. The system requirements of having a staking node hopefully will decrease over time, especially as things like statelessness come in and probably even more once Zekisnerks come in.

Starting point is 00:51:56 I think my longer term dream for Ethereum is basically that all that a staker will have to do is just download and verify a bunch of data, verify once an arc, and then sign something. And if that's the case, then, like, staking on mobile phones is going to be extremely viable, right? Like, even if you have a phone on the go most likely, because it will not consume a huge amount of battery, the only thing it will consume a huge amount of is bandwidth, right? But I think it will take five or ten years to get there. but, you know, I'm hopeful, I think. And we also have BLS signature aggregation, which can allow us to go from the 32 ETH stake down to 16, maybe down to 8. Yes.

Starting point is 00:52:36 So like the future sci-fi version of Ethereum is I'm just like walking around with my phone with eight ether, verifying the network. Right. So this actually gets into one of these bigger challenges and like tradeoffs in the single-swat finality roadmap, right? So in the single-s-slaught finality roadmap, basically, the idea would be that, like, instead of all validators participating over the course of 32 swats, you would have validators, every validator staking and sending a message during every individual slot.

Starting point is 00:53:10 And if you do that, then there's basically two ways to do this, right? One way is what we call supercommittees, where basically you don't literally have every validator staking in every swat. you have some subset, but it's a pretty big subset. Like it's about 4 million etha worth of validators. But then you just rapidly rotate all of the validators in and out to try to make it fair over like even pretty short time scales. So that's one approach. And if you take that approach, then it should be very possible to reduce the minimum

Starting point is 00:53:44 each deposit size at the same time. The other approach would be to say, well, you know, we are just going to literally have every validator stake in every slot. And this is something that realistically would do some combination of increasing the load on the network and reducing the number of validators that the network can support. Now, one nice thing that it will be able to do is it'll be able to basically remove the need for individual validators to have multiple slots, right? Like someone who has 10,000-eath, they'll be able to just have one slot that's just 10,000-eath.

Starting point is 00:54:19 And there's some technical reasons why we can't do that today that have to do with Lugshot. in committee assignment. But both of those things would go away if we have a single-sword finality and if we have dank sharding or even proto-dank sharding. And so then like basically you'll be able to have, you know, validators with like 10-eth alongside validators with 10,000 eth. So the one nuance, though, is that we would not be able to like have a hard guarantee of 32-Eth, being like of it being possible to stake with 32 eth basically because like we will not be able to know what the distribution of other stakers is right like we'd be able to

Starting point is 00:54:58 support a maximum of you know maybe about 130,000 validators or maybe a bit more but if it just still happens that everyone who has a lot of youth still chooses to split their eth up into like 32 eth just you know either because the wealth distribution turns out that way or because like some people just want to be mean and take up more slots then, like, unfortunately, the minimum staking amount might end up, like, increasing a little bit higher than a 32-Eath, right?

Starting point is 00:55:27 And that's one of the risks of that approach. But at the same time, the benefit of that approach will be a lot more protocol simplicity, a lot more security. The chain would, like, finalize, and you would get up to maximum security after, like, 32 seconds will be a number that I think could be realistic to hope for. So, you know, there would be a lot of benefits, but at the same time, there would be this cost. But on the other hand, you know, if you wanted to try to optimize for having some native ability to stake with like one ether, four-eth, then we would need to talk about much bigger validator set sizes and bigger, like, needing to handle a larger number of validators at the same time, but also having this like super committee mechanism, and then there would be more

Starting point is 00:56:14 complexity, and there would be more different kinds of tradeoffs. and yeah so basically you either get one really good thing and you sacrifice a bit on the other side or you get a different really good thing but then you sacrifice in other ways. If you want a more detail of that document that describes this tradeoff, I think if you just search paths towards single swat finality, then like we talk about the two strategies in there, the supercommittees versus global validator set. and I try my best to explain to readers what the benefits and the disadvantages of each approach are. But I think because this affects not just the technology, but it also affects the economics and it also affects the staker experience, I think it's important for this to be a community-wide discussion.

Starting point is 00:57:04 So I would really welcome people to read that document and kind of start making their own views heard on what things they value more and what things they value less. and we can keep trying to improve these proposals into something that can hopefully come close to satisfying everyone. Let's flag that for the bankless community, the Ethereum community, which is, we'll include a link in the show notes to the post past a single slot finality. It sounds like there's going to be a lot of conversation around that in the months ahead as we figure out as a community what to do there. And those tradeoffs are real, Fatalic. It is always difficult to adjust, you know, magic numbers like 32Eth up. and yet there seem to be some real benefits to doing that. And there are also benefits in terms of decreasing the minimum amount to stake as well.

Starting point is 00:57:51 But I want to ask you more broadly from another lens on this. I mean, we've gone through the product roadmap lens from almost like a developer's perspective, the features that need to implement. But I think there are like maybe three questions in people's minds or three different topics that have come up recently about Ethereum's future. The one is this concern or question over centralization, of stake, which we can get to. Another is censorship resistance, which has become a popular topic over the last few months. And then the third is looming in the background is layer twos and how have they lived up to the hype.

Starting point is 00:58:26 Are they really working? Let's start with the first. Centralization of steak. And the concern here of Italic is Coinbase, Cracken, large staking pools. They are going to own all of Ethereum steak or a majority or some number that kind of like breaks the security guarantees of Ethereum. Are you concerned about this? And what are your thoughts on how the roadmap can mitigate it if you do have concerns? Yeah, I think it's definitely always a concern. I do think that the concern is overhyped a bit, right? Like the trolls do sometimes, for example, try to kind of sneak in this depiction of Lido as being a single centralized actor.

Starting point is 00:59:12 and I think a lot of people in the Ethereum research team are critical of some aspects of LIDO, but I think it's important to defend it in the sense that it's definitely not true that Lido is a single centralized actor, right? Like there isn't one, you know, owner or sysadmin or dev that has the ability to flip the switch and turn all of the Lido actors into participants in some kind of attack, right? like it's a protocol that distributes a stake between like somewhere between like 19 or 21 or like some pretty big number of a subvalidators. We are each of those subvalidators only has a few percent of the stake. So Lido does have some pretty decent decentralization internally. Obviously even taking Lido out than like Coinbase Plus Cracken plus a bunch of others do add up to quite a bit. and it is a concern.

Starting point is 01:00:08 I think in the short term, the good news is that, like, you know, these are good people and they're people who love Ethereum and they do want Ethereum to prosper. And so I think the risk that they're going to do something terrible in the short term is pretty low. I mean, Brian has even been willing to say that, like, he would rather shut down the Coinbase staking his service than turn it into an engine of censorship, right? So obviously, like those kinds of commitments are commitments that we should hold the ecosystem to, like every single centralized dust-taking provider even. But obviously the good intentions of specific people are not something that we want to rely on in the long term because we are about being a decentralized ecosystem. And I think in the longer term, there are some good solutions, right?

Starting point is 01:00:58 So one thing to point out is that like Bitcoin proof of work. decentralization and even pre-merge Ethereum proof of work decentralization. Like, they're not that high either, right? Like, I think Bitcoin is, uh, well, a kinch rolled by, like, more than 50% of it is like two or three mining pools. Um, let's see. I'm going to go up on the internet again and check, btc.com stats pool. Foundry USA 23.

Starting point is 01:01:25 And pool 18, F2Pool 17.1, Binance pool 14.3 via BTC. 9.5. So like three pools control more than half the Bitcoin network and five pools control like 80% of it, right? So those are pretty scary numbers. They are no less scary and probably even more scary than Ethereum proof of stake. Now people do sometimes reply by saying, oh, come on, you can't say the proof of work pools and pro steak pools are equivalent. Proof of work minors can switch pools at any time, but stakers can't. But the reality is that we're like months away from withdrawals. And withdrawals are going to mean that, you know, yeah, if you're not happy with your staking pool, you absolutely can withdraw and you can immediately redeposit into a different one.

Starting point is 01:02:12 There's actually even the possibility to add a feature to the protocol, which is even better than withdrawal, like basically give stakers the ability to transfer their staking power immediately. Now, the one caveat of that approach is that the staker would still be vulnerable to being slashed by their previous staking provider for some amount of time, right? But aside from, like, in terms of the kind of power in the network, like, with this kind of feature, they would be able to actually move it over very quickly, right? So ProofStake absolutely can be upgraded and can be designed to make it possible to move stake between different pools and different providers and even change your own keys very quickly.

Starting point is 01:02:59 So that's something that will increase practical decentralized. quite a bit. Withdraws being enabled is also something that will practically like really increase the ability for people to solo stake because as a solar staker like you're not going to have this liquidity problem like you'll basically have the same ability to withdraw your money quickly that even or almost like the same ability in the normal case as a participant in Lido does right. In an exceptional case where the chain is like under attack or whatever then if you signed up on your estate, or at the same time as an attack is happening, then like, you know, sorry, yes, this is what you

Starting point is 01:03:36 signed up for and you have to stick around and validate for a couple of weeks. But aside from that exceptional situation, you know, you will be able to just go and leave, right? So as proof of state gets upgraded, I think over time, you know, we are going to keep improving this stuff more and more. And I think proof of stake is going to get pretty close to a system where you will be able to just like start and stop staking whatever you want. And I think people, are underestimating the extents to which the very real weaknesses of the current proof of stake system are something that's going to stick around for the long term. Vitalik, would you make the claim that just as soon as proof of work was kind of switched off,

Starting point is 01:04:16 we fully transitioned to proof of stake, that proof of stake, even in its current form with maybe some centralized actor controlling more stake than we'd want, is still more decentralized than the former Ethereum proof of work network. Would you make that claim? I would say absolutely. I would say that for a couple of reasons. I mean, one is that once you take into account all of the solar stakers and Lido's internal decentralization, I think it's just true as a matter of fact that post-merge Ethereum is better on any reasonable decentralization metric than proof of work Ethereum or even proof of work Bitcoin. The second thing I would say is that proof of stake has much better recovery from 51% attacks, right? Because if a 51% attack happens, then, you know, you can socially recover from it. You can partially automate recovering from it. You can do a lot of, there's a lot of recovery options that exist that just don't exist at all in proof of work, right?

Starting point is 01:05:22 In proof of work, like if half of the miners start attacking, you're pretty much screwed, right? like there is this argument that like oh you know the good guys are going to all come together and they're going to offer hundreds of millions of dollars of like capital and try to like start bringing the chain back toward the good guys but like come on you know it's like just why would you even make a system that relies on that when you can make a system where if an attack happens you can just delete the bad guys they can keep going right it's like just the practical ability of doing it i think is just a much stronger So the, you know, if a 51% attack, a stake attack starts, like, the amount of actual lock-in that it has is much lower. What's always interesting to me about, like, popular crypto narratives is, you know, you always have to look at them from first principles and kind of think rationally about them. Because as a matter of fact, it does seem like people were not complaining about how centralized proof of work Ethereum was in mining pools, et cetera, in similar form to Bitcoin. And yet, Ethereum through proof of stake has become more decentralized. and now people are complaining about it. And that's not to say that Ethereum doesn't have a ways to go and needs to work on this

Starting point is 01:06:32 problem. I think it absolutely does. But if you just look at the upgrade that just happened, you'd think that Ethereum just became kind of a centralized, you know, controlled apparatus of the WEF in the state. Like, with the transition, that has not happened. But can we talk about a second maybe looming threat, which is people are wondering about censorship resistance. What about censorship resistance? Is that still a problem Ethereum needs to solve? I think censorship resistance is absolutely an important issue. And I think there is absolutely a risk that

Starting point is 01:07:07 we should be worrying about that, like, particularly with the switch to like MV boost type architecture, where validators are going to be outsourcing responsibility for which blocks to produce, that basically we're going to enter a world where almost all blocks get produced by a couple of centralized actors, and those centralized actors end up censoring. So this is something that has been talked a lot about internally, both in the Ethereum research community and in the flashbots and MVP community. And the good news is that there are a lot of mitigations, right? So one of the mitigations is this concept of transaction inclusion lists,

Starting point is 01:07:49 a.k.a. CR lists, which allow validators to say, hey, I'm going to outsource responsibility for creating a block, but whoever does it has to include this set of transactions. Right. And so you as a validator, if you identify transactions that are being censored, you can add them to an inclusion list, and they just are going to be included in whatever block you accept. There is even a version of this where the validator adds their transactions to the end after the, yeah, builder. has already submitted their block, right? So basically giving the validator extra freedom to add whatever they want to the end of a builder's block.

Starting point is 01:08:28 So the builder doesn't even have to realize that there's particular transactions that the valigator is intent on adding in the same slot. There's the longer-term stuff around decentralized builders and SGX builders and MPC and Zegasnerk builders and all of these different concepts. So basically there's a lot of these different ideas that try to bring back more autonomy to the validator and make sure that the validator still has control and the ability to force transactions that they saw to be included despite builders having control over transaction ordering. So that's one major mitigation.

Starting point is 01:09:11 Another major mitigation, I think, is just ensuring that the builder in real ear markets are as competitive as possible. and that's something that I think we've seen a lot of good work on over the last couple of months, right? Like there are also the blocks route builders. There's a couple of other builders. There's a couple of other reliers that are involved. I think there's some dashboard somewhere. Let me guess see if I can find it again. Mavieboost.org has one.

Starting point is 01:09:37 If you look at top reliers and top builders, so the flashbots relay is making 81% of the MEPBust blocks, but then blocks throughout max profit is making more than half of the remaining ones, and then followed by blocks throughout ethical, followed by blocks throughout regulated, and then block native manifold and even are still managing to make blocks at least once an hour. So Vitalik, like David and I run our own nodes via rocket pool, and it's been really interesting because you have to pick your relay and your block builder, right, of this list, basically. These are some options. there's an interesting question is like blocks route regulated is like o-fact compliant your version of a

Starting point is 01:10:23 builder and a relay and blocks route max profit is not for example and so the individual validator chooses now are some of these decisions going to be like enshrined in protocol so that a builder a relayer cannot choose to be kind of like o-fac compliant or something like this because what happens is it does seem to present a choice to the validators. Do you want to do the blocks throughout regulated or do you want to do the blocks route max profit? Like OFAC compliant or not? And if you're Brian Armstrong or a coin base and you're under some regulatory pressure, I'm betting you're going to pick the regulated option. Does this get solved in protocol in some way? Good question. So first of all, I think the concept of reliers is going to go away when we have in protocol PBS, right? This is

Starting point is 01:11:14 I think something that's really important, right? So relays are this thing that's like very public and they have like these publicly available names and you have to consciously choose them because relays are this trust demanding role where if you as a validator sign up to a relay that's cheating, then that really can do a whole bunch of nasty things to steal money from you or cause you to lose money.

Starting point is 01:11:40 With in a protocol PBS, that trust component is going to go away, right? And so relays are going to go away and the only thing that will be around will be builders. So builders, I mean, even today the builders, they are just like zero-ex-something and they don't really have public identities, right? I mean, obviously, you are going to be able to discover the public identities of some of the builders. And, you know, there may be some validators that will decide to refuse to accept blocks from some of the builders. but otherwise, the market of builders is already looking quite a bit more competitive than the market for relays, and hopefully the market for builders is going to get even more competitive, right? Like any competitive builder market, the amount of room that even winning builders have to censor is actually very small,

Starting point is 01:12:32 because if they start censoring more than a tiny amount, then that decreases the maximum bids that they can afford to make. And so if you censor it more than a tiny amount, even as a winning builder, then you would just naturally start producing lower bids than some of the non-censoring builders. And so validers would just automatically start choosing blocks from non-censoring builders just because the bids that they offer are going to be higher. So with relays as this kind of unfortunately centralized, that trust-providing function removed, I think a lot of the problem is going to go away. That's good to hear. All right, let's get to the last lens of Ethereum problems that need to be solved. And this is one that was already underway pre-merge, but maybe with proto-dank sharding, 4-844, gets kind of accelerated. And that is the roll-up-centric roadmap. And I just want to ask you a question about L2 progress and in general. Have L2s so far, layer twos, that is. Have they lived up to the hype? Have they lived up to maybe even your hopes? I think my answer is mixed. I think they've, they have done a lot in like actually having a thriving ecosystems around them and better and better ability to bridge between them, starting to add some of the ideas around compression. The ZK EVMs have amazed me. You know, they've gotten to a point of where they are much faster than I was expecting that they would. So there's just incredibly smart people that have been doing incredibly good work there. The one thing that's probably happened slower than I expected by quite a bit is work on fraud proofs, right?

Starting point is 01:14:13 Like both optimism and arbitrarum, like they haven't taken off the training wheels yet from what I understand. And it's even in some cases looking like it's going to be quite a while until the training wheels come off, right? And so this even makes me wonder, like, if this is the case and like if it's, turns out that making a good fraud group is like actually almost as hard as making a good ZKVM, then are we even going to see this like leapfrog situation where some of these optimistic roll-ups are just going to at some point move straight to ZK even before the training wheels fully come off on the optimistic roll-up? I mean, if so, then that would obviously be a really fascinating outcome. But, you know, at the same time, then that's obviously a huge win for like ZK

Starting point is 01:14:59 roll-ups as a concept over optimistic. roll-ups as a concept. But we'll see. So, Vitalik, we've done a good job summarizing the Ethereum Roadmap, talking about the post-merge world, and also the things that have been in conversation lately, censorship positions, layer 2s, and centralization of stake, of course. And this really, I think, does a good job summarizing the last two years of Ethereum, I think. There was the era of proof of stake and shipping the beacon chain. That was like 2020 to where we are now in September of 2020. what do you hope that the era of 2023 to 2024 is for Ethereum? What should the developers be focusing on?

Starting point is 01:15:40 What should the paradigm of development be? And also, what role should the community have? People that are not technical, people like me and Ryan and the content creators and the content consumers and the generalized Ethereum community, what role do they have over the next couple years? Sure. So I think there's two big priorities. One of the priorities is to get scaling figured out.

Starting point is 01:16:00 and I mean that at all layers of the ecosystem. Like getting the Ethereum protocol of fully roll-up ready, which includes things like Prototank charting, getting roll-ups themselves to be fully ready for users, getting applications on top of them, getting good bridge infrastructure between them, getting all the wallets to support them, just helping the transition to a fully roll-up centric Ethereum complete.

Starting point is 01:16:22 And then the other one is a transition from Ethereum being in rapid developing, firefighting mode to Ethereum being instability mode. I think it's a transition that has to happen. And like, I do think that to some extent it's an inevitable transition because as the ecosystem grows, the cost of changing things increases and then there start being all these regulatory concerns and lots of existing stakeholders. And so there is this kind of narrow window to get a lot of important changes through. But at the same time, like the ecosystem really needs to get out of this kind of firefighting, like, you know, hey, the community is yelling at us to get something

Starting point is 01:17:03 now, now, now. And so, like, you know, let's make a totally stripped down pragmatic version of it and ship it mode into a mode of, like, much more deeply caring about making sure that every single step that the roadmap takes is on the path towards some kind of more stable form of a long-term roadmap that leads toward sustainability. And... Aside from those two, there's obviously some smaller items like, you know, the verge, the purge, the splurge, account abstraction, ZK snarking things, continuing to improve proof of stake. But those are kind of separate. They're also kind of a, I mean, they're kind of the same thing, right? There are protocol changes and in a lot of cases that are geared more toward long-term sustainability than toward, like, putting out fires that exist today.

Starting point is 01:17:55 but, you know, they need to happen and, you know, they will happen and we can just kind of be fairly calm and careful about it. Make sure not to put too much strain on the development teams, both because that'll cause them to burn out and because we don't want it to, like, become, not become possible to create a new client at some point in the future because there's so much work to do. So, yeah, I know, those are the goals as I see them. So I think what you're saying is, and what I'm also seeing is the area, of when merge is over, but also more broadly, the era of when is over. And I think that's because, like, you know, communities, the masses, they like the token stuff. They like the prices.

Starting point is 01:18:38 And the merge was fundamentally from the community standpoint about reducing the eth issuance. So that's going to be the thing that impacts the thing that they obviously care about the most, which is ether. And now that's in the rearview mirror. Like, when merge, like, we already did it. So, like, when charting doesn't really have the same ring to it. I don't really see as many people going when charting, when charting. Not with gas prices, this low, David. Yeah, not with gas prices as well. That's a good point. But I think, Vitalik, what you're saying is that

Starting point is 01:19:03 like the whole when, when, when part of the Ethereum communities in the rearview mirror. And now that we've shipped proof of stake, we shipped a good version of it, it's working well. And now the meta of Ethereum developers can be a little bit more relaxed and intentional as to what actually, and maybe even more of a perfectionist as to what actually does get shipped on chain. And so we can make sure that we get 4844 absolutely perfect.

Starting point is 01:19:31 We can make sure we get PBS absolutely perfect because we are in a slower and more intentional paradigm of Ethereum development. Would you relate and resonate with these statements? I would say so. I say that's a very good summary. Beautiful. Hey, Bankless Nation, as we alluded to in the beginning, we're cutting up this episode because it was so long. And we've gotten the feedback that y'all don't like the long episodes. So we're cutting it up, splitting it into two different episodes. The second half of this conversation is coming a week from today, next Monday, and this is the conversation about Ethereum as a platform for network states, Ethereum and its intersection with AI. And overall, how DAOs should be governed and what a DAO really is.

Starting point is 01:20:12 So stay tuned for that conversation. It's happening in seven days if you are listening to this on Monday. But also, if you're a bankless premium subscriber, it's in your feed right now. So switch over to the premium RSS feed to finish that conversation. Otherwise, I'll see y'all in a We're going.

Bankless - 138 - What's Next? Vitalik Buterin | Part I

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.