Escaping the Splinternet with Vitalik Buterin* of Ethereum

Jeremy: I want to welcome and introduce Vitalik to the conference. Welcome, Vitalik.

Vitalik: Thank you so much, Jeremy. Happy to be here.

Jeremy: Awesome. So I was just sharing a little bit of the first time we met and where that was in time. And I think first I just want to express my gratitude for everything you've done for the space and making things like USCC even possible. So thank you.

Vitalik: Yeah, well, thank you for making USDC itself happen. It's been a really big contribution to the space for a lot of people.

Jeremy: Thank you. We're excited and I think just getting started. So I want to come back to a theme that we're really thinking about here at the event, which is this idea of kind of how do we move from speculative value to utility value. And I think it's something that has happened multiple times in the history of this space where people get over indexed to speculating and lose sight of the utility phase. But I think right now it seems like it's a really particular moment because there's so much attention on this. And I guess as you look at where a theorem is with the merge behind us, what do you think the next kind of critical layers or problems to be built are and what do you think that does in terms of bringing utility at scale?

Vitalik: So, to me, the most important problem now that the merge is done is basically the scalability problem. I think it's important to keep hammering on the scalability as an issue because the scalability challenges really are central in my view to the problems that are actually preventing a lot of the cryptocurrency and blockchain applications that we dream of from going mainstream and that are motivating both centralization and dominance of some of the big money speculative use cases that we're trying to move beyond. Right? And I think it's also a little bit easy to forget today because we are in a bit of a bear market and demand is lower, transaction fees are lower. But if you remember what things were like a year ago, right, it cost more than $5 per transaction to use Ethereum, and it depends on how complex your transaction is. Simple stuff might have cost three to five. More complicated stuff might have cost ten to 20 right. A new contract cost some amounts in the hundreds. So it was crazy. And I think it's important to just appreciate the extent to which that issue by itself can be a show stopper to an ecosystem. Right? So, yeah, I remember when I visited Argentina back at the end of last year in December, and just I got a chance to see firsthand how some of the people who were using blockchains and cryptocurrency there often even for very pragmatic reasons, like it was just the easiest available way to move money and save money, and how they were using blockchains.

And what was interesting was that there were people who use cryptocurrency. People did have Bitcoin, ETH, USDC, USDT a whole lot of things. But most people were not interacting through blockchains, they were interacting through a finance account. And I think the reason why finance is such a just popular way of handling cryptocurrency should be pretty clear, right? Like finance to finance has instance transfers. It's no fees and you don't have to wait, you don't have to pay anything. And especially for people who are not wealthy people and not in wealthy countries, even being able to save $5 on a transfer is a big deal. And I think for everyone, not having to wait five minutes for a transaction to go through is a big deal. Right. So after the merge, the thing that we're focusing on as an ecosystem, it basically is scaling us through this roll up centric roadmap. Right? So there is this layer two projects optimism, arbitrary, StarkNet scroll, loopring, polygon, the list just keeps on going. That are trying to create these second layer platforms that are more scalable, Ethereum-like networks that live on top of Ethereum and inherit security from Ethereum but have used the blockchain in a more optimized and more intelligent way so they could have lower fees. And then there's upgrades that are being planned to the Ethereum chain itself. So prototype sharding is and then followed by full dank sharding. 

These are phrases that you'll hear in Ethereum research circles that are upgrades that make the Ethereum chain itself more scalable and more able to absorb the load from all of these roller projects. So after all of these, the Ethereum ecosystem's ability to absorb and process transactions will increase by a factor of something like 100 to 1,000. And that's going to be a big deal both to the kinds of the number of users and the applications they can get on the ecosystem and the kinds of fees that they have to pay. Right. And so the question to ask is like, well, you know, given the applications that are possible today at, you know, thirty cents to three dollars a transaction, what kind of applications will be possible when we're talking about zero point three to three cents a transaction? And I think the answer is quite a lot.

Jeremy: Yeah, totally. I use the metaphor of dial up to broadband and we are now in the broadband upgrade cycle. And you really couldn't do all the ideas people had for what you could do with internet apps. You really couldn't do them on dial up, but you can start to do them. That's an exciting roadmap, obviously. And I think we want to make sure that USDC as a protocol is working seamlessly on all these layer twos for that utility value. I want to change gears a little bit and talk about what we're starting to see, which is as some of this starts to go more mainstream, there's just more laws and regulations that are coming and starting to interact with this whole ecosystem. There's laws and regulations that are starting to see governments have a view about how public chain infrastructure might work or the roles of different stakeholders in it. And I'm just interested at a high level as this becomes more mainstream at scale, what do you think of the sort of role of governments and policy with respect to how public chain infrastructure in particular evolves?

Vitalik: Yeah, it's a very good and I think, important question. I feel like my general hope is, I think, the same as what a lot of people in the ecosystem have been hoping for the whole time, which is that the base layers should hopefully be left alone as much as possible. Because the challenge with base layers, right, is that they are global, and if one jurisdiction starts censoring them, then that just immediately makes them much less attractive to people in other competing jurisdictions. And basically the hope is that the public blockchain space is a kind of humanity’s escape from the splinternet world and if the space just becomes part of the splinternet then there's almost no point to it left, right? But at the same time there's also the application layer and applications are the things that actually do things and some of the things that applications do don't fall under regulations and some of the things that applications do fall under regulations. 

And I definitely think that it's not realistic to expect the entire space, applications included, to stay unregulated because I think the pressure to regulate applications is going to be too strong and the kinds of activities that some applications are doing, just regulated activities under existing definitions. I think my hope is that the crypto space can both definitely kind of fight hard to preserve the neutrality and the fundamental security of the base layer, but also be proactive and actually kind of participate constructively in the conversation of how different kinds of applications can be regulated and what are the ways of both complying with regulations as written and satisfying enough of the goals that regulators and the constituents both do have and might have that hopefully both that satisfy their needs and at the same time, not preserve the properties of the applications that are the reason why people want to use them.

Jeremy: Yeah, I think the more that we can educate governments that public chains are like operating systems and it's a neutral technology and yeah, there may be financial applications and non financial applications and others it's really important for people to be anchored that I think oftentimes people just assume everything that this is is somehow like a financial application, which they're all not. It's sort of related. But I want to step a bit back and talk about kind of when we think about this new infrastructure layer, this new base layer. I think one of the things that really inspired me getting into the space and I think has been important from what I can tell in some of the origin ideas around Ethereum too, which is sort of in what ways can we build an infrastructure layer that allows us to have new building blocks for social, political and economic structures? Like new structures that we can build on? And I think back in college, when I was in college, I was pretty interested in the work of people like Herbert Simon and an economist and looked at sort of all the different ways in which organizations could be structured and the relationship of organizations to markets and other things. 

And I think embedded in Ethereum and embedded in this public chain infrastructure is this kind of social, political and economic philosophy that's there. And I'd be interested to hear you just talk a little bit about if there are things that you think could become possible in terms of social and commercial and even political organization that get built up on this infrastructure. And I think I'm thinking very long term about this. What are the materials that we now have to kind of work with as we think about new types of microstructures and macrostructures that we can build from the ground up on public blockchains?

Vitalik: Yeah, I think for me the idea of blockchains being this like base layer that can be used to actually like build and enable all of these different incentive structures and institutions and different ways of managing both in the economy and other aspects of society I think is definitely really fascinating. And I do think that there are a lot of opportunities for blockchains to be that layer. I think the main big learning that I've had over the last five years or so is this idea that if we want blockchains to be used for things that aren't basically just different versions of the concept of money and that's it, then there needs to be some way for blockchain applications to exist that are not completely financialized. Right. And what I mean by this is that you can't just create a currency and then say, oh, this currency is not intended to be a money token. 

This currency is intended to just be fun points and then expect that that's what the token is going to be. Right? Because if your token is transferable, then there is going to be an exchange rate on you to swap between that token and the USDC or ETH or whatever else and it just is going to end up having all of the properties of a money token. Right. There are a lot of good reasons, like very fundamental game theoretic reasons that you can't just wish away by having a slightly different culture or whatever for things like governance, for example, you don't want purely financialized infrastructure to be controlling it. There's actually a lot of problems with governance tokens and if you want to make a DAO, then you want the participants in the DAO to be aligned toward what the goal that DAO is trying to accomplish. And if you have a DAO that's fully governed by a token that's typically transferable, then some of that token may well get into the hands of completely unaligned participants. Some of that token may well get into the hands of participants who are willing to sell their vote to the highest bidder. And just all of these issues and basically everything just collapses into being either some kind of profit maximizing system or even it just gets kind of bought up and kind of hacked outright. There are even stable coins where somebody bought up like half the governance tokens and then made one proposal that voted for all of the money in the treasury to be transferred to them and they made something like $80 million or $40 million off of that. It was crazy, right? And so actually making applications that are non financialized in the sense that they actually try to prevent this kind of collusion where people do things in one application in exchange for some favor in another application, that's the big thing that blockchains need to be able to do somehow, right? And so this is where the ideas around sole bound tokens that I've been exploring recently come into play. This is also really important for voting applications, right? One of the important properties for voting systems in general is this idea called the coercion resistance. It's a very strong form of privacy that says no one can see how you voted and you don't have the ability to prove to someone else how you voted, even if you want to make that proof. Right? And the reason coercion resistance is important because it basically means it's needed to prevent people from selling their votes, right? And so that's making that sort of thing actually possible. Within a blockchain space, there are projects that are trying to accomplish this.

So there's a lot of the recent projects that are trying to combine together zero knowledge proofs with blockchain technology. So MPCs, CLR funds, there's just kind of growing list of them. I think it's fascinating and I really support it. So I think those kinds of technological base layers are important and I think things like proof of community, proof of attendance protocol, all of these protocols that try to actually put richer information about the participants on chain so that other applications can start actually plugging into that, right? So if you want to issue a new token and you want to issue 1000 units to every unique human, you should be able to, if you want to create a new Dow and you want the governance of that data to be run by people that have actually participated in some community, you should be able to right? If you want to do something where you issue some tokens to people that have completed a particular task, you should be able to do that, right? Creating these building blocks to enable that sort of stuff, that's something that I'm really excited to see happen over the next few years.

Jeremy: The surface area for experimentation, I call it, sort of describes like corporate forms, which are organizational forms, is probably one of the most, I think, overlooked and most important pieces of what's emerging here. And I've always been interested in are there different forms to how a corporation can exist? How essentially the effort of organizing labor and capital together and how labor and capital can work together? How these entities that kind of combined labor and capital can, can work across each other and across the world. And there are, you know, these sort of ideas of like, syndicalism which go back, you know, 100 plus years. And it feels like just as I listen to you and think about the building blocks, cynicalism as an idea, it was rejected, right? It was essentially like the Joint stock corporation and sort of what we think of as sort of modern state capitalism was what kind of manifest itself and grew. And then we've kind of evolved now where we have this Internet and we have building blocks like public chains, and we have the ability to start composing new structures where new types of entities where people can interact and where labor and capital and output can work and where you could connect these different types of entities together in different ways. These are these sub dows and dow linking and other things. How do you think about that at a broader level? Do you feel like the experimentation that's happening now, the building blocks that we're talking about here are going to over a longer arc, let's call it, whatever, 20 years? Are we going to be able to have a new shape to the very nature of the firm and the way work and production happens?

Vitalik: I think one of the really interesting things that's been really on the rise in the last ten years is the rise of informal cross organizational collaboration, right? So this is like cross organizational collaboration that really is not kind of mandated by any kind of grant strategy. There isn't a signing ceremony, there isn't a treaty, there isn't anything sharing agreements, no lawyers involved. It's just that people in different groups that are working on similar things this join in Telegram chats or signal chats or whatever else, and they talk to each other about the stuff that they have to talk to each other and from wherever they are. To wherever they are. And they just get a lot of work done, right? So I talked about zero knowledge proofs a lot, and I talk about this topic a lot because I do think that zero knowledge proofs are this super important technology that's even really complementary with blockchains and some very nice ways. 

But the other thing that's fascinating about the zero knowledge proofs community is just how much it depends on that sort of stuff already, right? So if you look like Plonk, for example, like Plonk is this very popular 0 proof protocol that a lot of people use. It was created by a couple of researchers from Aztec. And into that protocol, a lot of input came from researchers from the Ethereum Foundation and from academic institutions. Those people also just had some separate chats with people from Starkware. Then implementations of that were being made. One of those implementations was in Circom, and some of the people involved in that are now these days also involved in polygon. So there's just this kind of complicated mesh of people from all kinds of companies, all kinds of academic institutions, all kinds of countries. Like, I think even the groups I talked about already cover the US. UK. Germany and Israel. And I'm sure the list grows even bigger, right? So this stuff happens and it's just a natural way for humans to work together and it's actually an efficient way for humans to work together in the summer global context. 

So the next question is basically what kind of formal structures can we create that play more nicely with this kind of approach? Right. The issue that you sometimes get when you have bad formal structures and that is that these informal structures that just really make a lot of sense and really work nicely with how humans just naturally want to cooperate is that you end up getting this kind of friction. And there's always a trade off between being good at just doing research versus kind of satisfying each individual's, each individual formal organizations, incentives, and it's economic survival imperative. And there's all of these kind of open source versus closed source trade offs. And sometimes people want to stay in silos and don't want to collaborate with each other. And so the question is, can we create better formal structures that are more aligned with this kind of collaboration that's happening already and clearly is a very good model for cooperation? It's an interesting question.

Jeremy: I think the laboratory is sort of the crypto ecosystem where we've seen a lot of this and sort of the thing I'm interested in is when does this cross over, to quote unquote, outside the crypto ecosystem? When does this crossover into coordination and creation and output that is not just in the production of software or other things, but into even creating other types of services and products and so on? Where do you see that? Do you see that taking place or do you see examples of that already kind of taking hold in a meaningful way?

Vitalik: One of the earliest signs of that happening I'm seeing is in the DeSCi movement, the decentralized science. You have projects like Vida, DAO, and then they have the Bitcoin grants, decentralized Science grants. You have people trying to create these kind of institutions that perform the unavoidable and formal task of allocating capital from a limited pool of capital, because all of these projects do ultimately need capital to do stuff. But trying to do that in a way that's collaborative, in a way that encourages cooperation across all of these different institutions, and it tries to get different people from all kinds of places who have relevant expertise involved. So I think that's starting to make a lot of interesting progress, though. Also it's good that you get to see some of the limits of at least existing DAO models. Right, so one of the things that's interesting is that I think the concept of creating DAOs that are not just token governed is starting to gain legitimacy. 

So if we talk to the people in Ukraine DAO, for example, they don't have a governance token and they have some very deep reasons for why they don't want a governance token because they really value the idea that the people who are governors are aligned with the cause. Right. And I think that's something that more and more projects are going to realize. One of the ideas I kind of published in that recent post on DAOs that I made is this concept of pod based governance where instead of trying to decentralize by in a way that's still logically centralized, you still have this one singleton that's making all the decisions except you just try to have 10,000 token holders voting on what the singleton does. You actually try to split the work up into individual teams. Totally. Yeah. So that sort of stuff. It's good that that experimentation is happening and I definitely expect we're going to see lots of iterations on it both within the kind of crypto space for the crypto space and in some of these adjacent spaces. I'm really looking forward to seeing what comes out of that.

Jeremy: It feels like this is supranational. But when I think about it, the kind of natural course is that there are going to be these new organizational forums that are at scale, that have very large numbers of participants that are producing in ways that a traditional multinational corporation or other would have been organized not just to produce products, but to organize around a lot of things. I think one of the questions that I think it relates back to one of the earlier questions I asked you, which is sort of like what are government's going to say about all this? Which is at what point do we see sort of this crossover between essentially code is law and what happens in what's been developed and organized using a completely sort of on chain based systems of governance and incentives and the like when that sort of interacts with nation state law and when the coordination work is global, right, the coordination work and it's sort of its participants from it could be 190 countries or whatever, whatever it is. And and the substrate makes it possible to do all these things. But there aren't necessarily contracts. There's kind of code as law contracts. And do you think this emerges from, like, case law in different countries or what do you think the path is for these kinds of supranational on chain organizational forms to become sort of viable at scale and that entities, quote, unquote in real life or other forms of traditional contract entities, corporate form entities can kind of interact with these kinds of organizations.

Vitalik: Yeah, I think it's interesting to explore to see how in the 20th century the big corporations were a political technology in a lot of ways almost as much as an economic one. Right. Because the very dominant social pattern of every employee working for a particular large corporation and being in a particular formalized marriage and all of these things, they make society very legible in the kind of seeing like a state sense of the word. And they do enable very particular forms of tax collection. They enable very particular forms of regulation and the internet in general, like even forget about the crypto space in particular, just the internet in general. I think frustrates that kind of enforced legibility and that's definitely going to be discomforting to people who are either used to that kind of legibility or who just believe in that kind of structured kind of institutional style vision of providing force and managing things in society. But I think it's a necessary transition and it's just obvious that these newer and less legible ways of working together are just better forms of collaboration in a lot of cases and the different forms are going to adapt. I think there are different kinds of adaptation that work in different situations. Right. 

This question of how to regulate DAOs, that's just going to have to have a lot of case by case law involved, right. Because there is the concept of unincorporated partnership. But the challenge with DAOs is that they have this sliding scale of involvements, where you have people who are involved 100% of the time, you have people who are involved 10% of the time, you have people who are involved only by holding tokens, and then there's people who hold 11.3% of all the tokens, and then how often do they vote? And it really frustrates attempts to create clean separations between the directors and major shareholders and employees and everyone else and there are case laws basically going to have to figure that out somehow. One of the things that I think could make these kinds of transitions easier is that there are forms of both regulation and taxation that are not as dependent on legibility. 

For example, I'm a big fan of land value taxes, right? One of the economists that I've gotten to know recently, I'm a big fan of his name is Lars Dowsett. He has this site called gameofrent.com that's just an excellent introduction on why land value taxes are a really good idea. But one of the interesting side benefits of them is that they are very low on legibility because you don't have to know what people are up to. You don't have to know what people are doing with their land. You don't have to know what all of the different things that people have and why they're doing them. You just have to be able to estimate land values and you can estimate land values at a fairly macro scale. You only need a little bit of visibility into the economy and it's also at the same time if you're a good way of raising a lot of money. Right. I think being low legibility is like this good heuristic for good principles, for basically how to create regulatory and tax and political responses to how people organize that are very future proof and that aren't going to just be overturned in like, 15 years when people invent yet another new form of working together. Yeah, so I think there definitely are ways, and I think the 21st century is definitely going to have to look very different from the 20th, but I think the stuff can be figured out.

Jeremy: Yeah, so just in closing thoughts and kind of building off that, I think the invention of new fundamental political, economic, social systems, and it's happening and we're building it on top of technologies like Ethereum and super. Grateful for everything that you've been doing and sharing your thoughts with all of us today. Vitalik thank you.

Vitalik: No, thank you so much for organizing this conversation. Jeremy, absolutely.