Fireside Chat with Vitalik：Coordination Challenges in Ethereum

https://www.youtube.com/watch?v=D6-nQyihVRw

In this fireside chat with Vitalik, we will explore coordination by examining the multi-layered coordination challenges and potential solutions within the Ethereum ecosystem. We hope this will inspire thoughts on how to manage coordination issues in decentralized ecosystems and provide insights into overcoming these challenges. Here is the transcript of the conversation.👇

Fireside Chat

Bruce: Hello everyone, I’m Bruce, core contributor of LXDAO and ETHPanda. Vitalik is a key participant and observer of the Ethereum ecosystem. We are thrilled to have you here. Let's begin with a brief introduction from Vitalik.

Vitalik: Hello, I'm Vitalik, co-founder of Bitcoin Magazine. I also have been in Ethereum for the last 10 years. I started off doing research, but also participated in many different parts of the ecosystem.

Bruce: Because we are here to discuss the topic of coordination, so first I want to say from your perspective, how do you define coordination in Ethereum? Are there any good examples to have effective coordination within the Ethereum ecosystem?

Vitalik: I think coordination can mean many different things. So very abstractly and very generally, coordination basically means that multiple people acting together for some kind of common objective instead of acting in ways that either ignore each other's needs or go against each other. And this can mean a lot of different things.

For example, this could mean people contributing to some public good, like to some project that is valuable to an entire ecosystem. This could mean people working together on some kind of common standard. So basically imagine people switching from speaking one language to speaking another language because the other language is better in some way. This is basically what happens every time the Ethereum protocol upgrades. And it could just even be this highly chaotic effort of people doing different things completely separately, but still with a result of contributors towards something common goal. So people editing Wikipedia would be one example. There is no attempt to order people around. There is no attempt to force people to do things in the same way at the same time. But the result is still that you have many contributors to something that benefits everyone. So I think all different types of that coordination happen in the Ethereum ecosystem. It depends heavily on it.

Bruce: Speaking about the Ethereum ecosystem, the Ethereum community encourages diversity due to the philosophy of the infinite garden. However, this diversity can lead to competition over resources and reputation or something like that. So what kinds of conflicts and coordination challenges have you observed in terms of competition and collaboration between communities? And do you think there are some solutions to make it a better collaboration and growth among communities?

Vitalik: I think the challenges that we have seen so far, one of them that I think we've actually resolved very well is just the different Ethereum clients working together to upgrade the Ethereum network, to update their code at the same time, a lot of different pieces of the ecosystem doing that. And this is actually a pretty impressive achievement. Ethereum is the one ecosystem where the largest Ethereum client-Geth, at least according to some estimates, only has about 52% of the network. And that is not something that is true anywhere else. In most ecosystems, there is basically one actor that takes over almost everything. And we see this in browsers, we see it in Bitcoin clients, you even see it in implementations of a lot of social protocols that try to be decentralized.

The challenge of this approach is that we still have to agree on what is the next upgrade that happens every year. And there's a lot of these structures inside of Ethereum that try to help do this. So there's in-person gatherings that happen every year. There's actually multiple of them. So we had one in Kenya, also another one that's a smaller one that happened yesterday. Then you have already, AllCoreDevs calls, online discussions also incentives. So at the beginning, the Ethereum Foundation gave all these client teams some very significant grants. And even today, it has some, but it's like the majority of their revenue still comes from the clients themselves. That's one example.

Another example is just funding public projects across the Ethereum ecosystem in general. And historically, the Ethereum Foundation has done this, but now we've seen Gitcoin grants, ProtocolGuild, other foundations. We published our transparency report about two days ago. One of the interesting stats was that in terms of the amount of public funding capital allocated in the ecosystem which is in 2022 and 2023, only 49% slightly less than half was the foundation. 51% was other organizations.

And now I think there's still other challenges. One big one is cooperation on standards between Layer 2 and between wallets. That's an area where discussions are starting right now. And all of these discussions around supporting public projects that's all actually continuing. There's even more people trying different experiments now that Gitcoin and Optimism and Protocol Guild have started showing the way.

Bruce: Actually, you just mentioned the conflicts of coordination between clients or something like that. So the next question is regarding the standards. As the Ethereum ecosystem grew, the EIP, ERC standard setting process involves more stakeholders making the process compact and slower. So what's the major conflicts have you observed in the standard setting process? Do you think there is how to balance the openness and the efficiency to make a consensus more effectively and drive the evolution of the standards?

Vitalik: I think there's like three types of conflicts that happen. And I think it's important to think about them a little bit separately. One of them is where different groups are trying to push different standards because the standards that they push are benefiting them. And this is something that happens not just in Ethereum. It happens everywhere in the world.

The second type of conflict is people who are pushing different standards just because they have either what gets called Not Invented Here syndrome or because they want the pride and the social status of being the one that creates something that everyone uses. And then the third is not even a conflict at all. It's just people have small differences of opinion, and you just have to put in the effort to get people into a room to actually put aside their differences and agree on something that everyone is reasonably happy with.

I think for the first, in a decentralized world, one thing that we're able to do is there's basic norms on what kinds of standards are likely to get adopted. Like if you publish an account abstraction standard and it says to send an account abstraction transaction, you have to go through my server, then nobody is going to accept it. People only accept things that actually seem neutral.

Then another thing that we are experimenting with is so for Devcon, at ETHcc, a lot of people were unhappy about just there being so many side events that were competing with each other. As a result, we did this experiment where we basically said that during Devcon, we would discourage people from doing any kind of side events. And so side events would happen before Devcon, after Devcon, and during Devcon, instead, encourage people to make community hubs that are inside of Devcon. And basically, the EF is more likely to support your community hub while it's really only likely to support your community hub if it's not just an advertisement for your thing, but if it actually includes collaboration from different partners that are in the same industry.

So basically, instead of making an Optimism event, or an Arbitrum event, or a Starknet event, we would make a Multi-L2 event. And that would just be part of Devcon. So actually trying to, at least socially, encourage collaborations to happen instead of one person pushing everything and then that proposal being granted as their proposal. That's something that we're also starting to experiment with more standards, as well trying to set that norm. I think that also helps to deal with the other issue. One side of it is this pride aspect of wanting to support a thing that is yours. And then the other aspect is this very human unwillingness to submit to someone else being the alpha male that's not wanting to, or wanting to resist something that's pushed by someone else.

I think the solution to both is to try to encourage more collaboration to happen, even at the very beginning of a process. And then for the third one, it's just not enough people talking to each other. We just need more people and more organizations actually creating those forums where that kind of talking happens.

Bruce: Thank you. So the next question from my side is like Layer 2. Because we both know that a Layer 2 solution plays a crucial role in scaling Ethereum. So how do you see the coordination challenges between that and Ethereum? And the challenges or strategies involved in aligning the development and the governance of Layer 2 solutions with the other ecosystem?

Vitalik: Layer 2 is definitely started in this very independent way where a lot of people started just building their own stacks and just trying to make something that works and something that can scale Ethereum reasonably quickly. And now the problem that the ecosystem is really focused on this year is basically that Layer 2 exist, Layer 2 work, Layer 2 are going to achieve. Everything is moving to Layer 2. But how do we make sure that Layer 2 actually feel like and act like one ecosystem instead of feeling like 40 different blockchains? And there are specific examples of this. It's like if you have coins on Optimism, but then you have some application on Arbitrum, then the process of depositing, like basically moving coins from one to the other is very hard. Like-minded are like non-standardized. There's lots of things that are non-standardized. So we've been starting to have these discussions on standardizing this cross-layer2 stuff. And that includes both Layer 2 teams and it includes wallet teams. So that's been an area where there's a lot of progress being made.

Bruce: Due to like a limited time, so Vitalik, do you have like any additional thoughts to want to share with coordination?

Vitalik: I think one important thing about coordination is that there's like the social people talking to each other aspect. And then there is the economic aspects to it. I think ironically is that like people like myself have a good history of like overweighting the economic side. Though I feel like in this conversation, we've actually overweighted the social side, which is kind of good.

But like the economic side is important to talk about as well. Like basically, like you can't try to like force people to act too strongly against their incentives, because as you see, if you try to like rely on like moral pressure too much, then eventually, like people just like get frustrated and they get angry. Then they rebel against you and just sometimes go and tell people a different crazy interaction. And so I think in that regard, the thing that we do successfully as an ecosystem is that I think we are good at funding small things. Like if you have an important public good that needs to be funded, and like all you need is like 100k or 300k to make a demo, there are lots of people that will fund it. There's Ethereum Foundation, individual funders, DAOs, individual Layer 2, ETH whales, like lots of people will give you 300k if they see the need. The challenge that I think we have is like when a project switches from that mode where like it needs 300k and it's a demo to that mode where it needs 30 million and it's like a big mainstream thing that needs to handle the entire Ethereum user base, the incentives, like they basically flip from like full socialist to full capitalist. It's like there's nothing really that like gives you incentives to act like pro-socially beyond market incentives at the 30 million level, because everyone basically says, hey, you already have enough money, right?

And that like we want to fund things that otherwise would not get funded. And like you're a corporation, you have users, you have investors. Then the challenge is that when incentives flip to 100%, then the incentive to continue to act pro-socially in the sense of like following standards, not trying to create vendor lock-in, continuing to be open source and all these things, like those incentives start to disappear. And so I think one challenge is basically, and how do we actually improve incentives like also at that $30 million level? And like to me, I think this is an unsolved problem. And I really welcome people trying out different things to try and solve it.

Audience Interaction

Q: I really want to ask a question about standards, but you said something that really triggered me. You were talking about there's flip, switch that gets kind of flipped from like small little pots of money to then like $30 million. There's like kind of different like un-new regimes for both. Isn't that a kind of problem that there is a switch that there's not like a more continuous way? What are the experiments where we can actually, like, for example, like small businesses in Web3. I feel like we don't see nearly enough of them. It's like everybody's trying to go to the moon or like we're hanging out here and going, hanging out in saunas and doing fun things. How can we support more small businesses? Curious how you were thinking about this or what kind of fun things you've seen.

Vitalik: I think there's like different types of support. One type of support is just more active. Like if there is a project that's promising then just like actually giving them a user base and like actually helping like that project get adopted in some coordinated setting where it gets some level of contact with reality and it can improve. One of these examples is like at Devcon, we've had dog food, ZK identity things like Zupass to pass a lot of these on chain or open source different things. And the goal there is that in part to like actually help projects like get past that network effect hurdle of nobody hearing about them and nobody using them. That's one type of non-monetary support.

I think another on the monetary support side, like once a project gets to that higher level, the question is basically I think you want to get equal, like you want to have some kind of funding model where it's not like full charity, because like at even the $3 million level, like if you just use charity capital, you're going to burn through it very quickly. You want a funding model that is not pure charity that like expects to see a return, but at the same time is not just motivated by a return. And the question is to find like one is what kind of actors are willing to participate in that on the money side?

I think people are. I think that there's a lot of people, like even a lot of ETH Whales who are keep ETH because they believe in the vision and they're even willing to make some small sacrifices, but at the same time, they're not as eager to just like give their entire lot away. And on the other side, the question is like what is the actual institution, like what is the model, like the model of funding that actually encourages things like staying open source, like staying standards friendly, being decentralized, ideally even that if you end up being super successful, giving back to the next wave of things.

I know that there are various projects in the ecosystem that are basically trying to like get together different funds of like various large ecosystem actors. Basically, the theory is that if you have these funds, at least if they have the confidence that they together are participating in this at the same time, then they would be willing to like invest in projects on that basis. But like so far, this is definitely one of the early stages.

Q: This is a question with regards to people starting their own projects instead of working on an existing. It seems like you don't have an incentive. If you're part of someone else's project, like the economic incentive is not there. You want to like what we really need right now is the user experience and the onboarding. But people only like especially VCs, if we want to have a bigger project, they only sponsor infrastructure. What can we do to improve the situation?

Vitalik: That's a good one. I think one of the interesting things is that I think both our private goods funding and our public goods funding like somewhat fails on this, because if you look at even like Optimism Retro Funding or Gitcoin or any of those, one of their big weaknesses is that they're basically popularity contests in the sense that they get in order to get significant amounts of money. You have to also have like a big public profile. You have to have your own marketing departments like you're basically a political party. And that's not like there's a large group of people that that model really doesn't appeal to like they don't want to be full time, not self-promoting politicians. And also like that is inherently a model where like it goes along with existing social status trends, which definitely favor the person who creates and not the person who maintains.

I think that if at least on the public funding side, we can work like intentionally and try to create mechanisms that identify and support people who maintain that could help things quite a bit. I have been seeing within the public goods funding, like the like the reject community projects that are trying to do this, like basically trying to identify downstream dependencies of like big things that everyone agrees are valuable.

Identify dependencies and then identify dependencies of those dependencies. And then you're actually able to support that whole graph. There's even been like very explicit attempts to try to like dig up and find things in that category in Optimism.

For example, the people who invented Keccak which is like the hash function that everyone uses, but it started by like academics who have no idea how to show themselves on Twitter. And like they got a 200K retro grant a couple of years ago.

So I think like trying to intentionally like at least create the public information graph of who contributes to what and make that more watchable is like one of those big, itself like dependencies. Once you have that, then you're able to more easily support any kind of mechanism that tries to improve it.