A Fireside Chat with Vitalik: Coordination Challenges in the Ethereum Ecosystem

CN
3 hours ago

"In this fireside chat with Vitalik, we will explore coordination by studying the multi-layer coordination challenges and potential solutions within the Ethereum ecosystem, hoping to inspire everyone to think about how to manage coordination issues in decentralized ecosystems and provide insights for overcoming these challenges."

Organizer: LXDAO

In this fireside chat with Vitalik, we will explore coordination by studying the multi-layer coordination challenges and potential solutions within the Ethereum ecosystem, hoping to inspire everyone to think about how to manage coordination issues in decentralized ecosystems and provide insights for overcoming these challenges. Below is the text content👇

Bruce: Hello everyone, I am Bruce, a core contributor to LXDAO and ETHPanda. Today, we will explore the theme of "Coordination" and the multi-faceted coordination conflicts and possible solutions within the Ethereum ecosystem. Through this dialogue, we hope to provoke thoughts on how to handle coordination issues in decentralized ecosystems and gain insights to address coordination challenges, making the entire ecosystem operate more smoothly and efficiently.

We are pleased to have Vitalik, an important participant and observer in the Ethereum ecosystem, with us today. First, let’s invite Vitalik to give a brief introduction.

Vitalik: Hello everyone, I am Vitalik Buterin, co-founder of Bitcoin Magazine. I have been involved with Ethereum for the past 10 years. I initially worked in research but have also participated in many different aspects of the ecosystem.

Bruce: Today we are here to discuss topics related to coordination. From your perspective, how do you define coordination within Ethereum? Are there any good examples of coordination in the Ethereum ecosystem?

Vitalik: I think coordination can have many different meanings. In an abstract and general sense, coordination basically means that multiple people act together for a common goal, rather than acting in ways that ignore each other's needs or are antagonistic. This can mean many different things. For example, it might mean people contributing to certain public goods, like projects that are valuable to the entire ecosystem.

It can also mean people working together on some common standard, like people transitioning from speaking one language to another because the other language is better in some ways. This is essentially what happens every time there is an upgrade to the Ethereum protocol. It might even be a highly chaotic effort where people are doing different things independently, but still contributing to a common goal. Wikipedia editors are an example. No one is trying to direct others, and no one is forcing everyone to do things in the same way at the same time. But the result is that you still have many contributors working on something that benefits everyone. So I think all these different types of coordination are happening in the Ethereum ecosystem, which largely relies on this.

Bruce: Speaking of the Ethereum ecosystem, due to the concept of the "infinite garden," the Ethereum community encourages diversity. However, this diversity can lead to competition for resources and reputation or similar issues. So, what conflicts and coordination challenges have you observed regarding competition and cooperation among communities? What solutions do you think could promote better cooperation and development among communities?

Vitalik: So far, one of the challenges we have seen, which I think we have handled quite well, is the collaboration among different Ethereum clients to upgrade the Ethereum network and update their code simultaneously. Many different parts of the ecosystem are doing this, which is actually quite an impressive achievement. Ethereum is a unique ecosystem where the largest Ethereum client, Geth, is estimated to account for only about 52% of the network. This is not seen elsewhere. In most ecosystems, basically one participant controls almost everything. We see this in browsers, in Bitcoin clients, and even in many implementations of social protocols that attempt to be decentralized.

The challenge of this approach is that we still have to reach consensus on the next upgrade that happens every year. There are many structures within Ethereum trying to achieve this goal. For example, there are face-to-face meetings held every year, and there have actually been several of them. We held one in Kenya and had a smaller gathering just yesterday. Then there are all the AllCoreDevs meetings, online discussions, incentives, and so on. Initially, the Ethereum Foundation provided some very important funding to these client teams. Even today, it still provides some funding, but most of their revenue comes from the clients themselves, which is an example.

Another example is funding public projects for the entire Ethereum ecosystem. Historically, the Ethereum Foundation has been doing this, but now we see the emergence of Gitcoin funding, Protocol Guild, and other foundations. We released a transparency report about two days ago. One interesting statistic is that in terms of the amount of public funding allocated in the ecosystem for 2022 and 2023, the foundation accounted for only 49%, just under half. 51% came from other organizations. Now I think there are other challenges as well. One significant challenge is the standard cooperation between Layer 2s and wallets. This is an area that is currently starting to be discussed. And all these discussions about supporting public projects are ongoing. Now more people are trying different experiments because Gitcoin, Optimism, and Protocol Guild have started to take the lead in demonstrating.

Bruce: You just mentioned coordination conflicts among clients, and essentially, as the Ethereum ecosystem has developed, the EIP and ERC standard-setting processes have involved more stakeholders, making the process tighter and slower. What major conflicts have you observed in the standard-setting process? How do you think we can balance openness and efficiency to reach consensus more effectively and promote the development of standards?

Vitalik: I think there are three types of conflicts, and it’s important to think about them separately. The first type is when different groups try to push different standards because the standards they advocate benefit them. This situation occurs not only in Ethereum but also around the world. The second type of conflict is when people push different standards simply because they have what is called "not-invented-here syndrome," or because they want the pride and social status that comes from creating something that everyone uses. The third type is not really a conflict at all; it’s just that people have some minor disagreements, and you just need to work to get everyone to sit down together and set aside their differences to reach a solution that everyone is relatively satisfied with.

For the first situation, I think one thing we can do in a decentralized world is to establish basic norms about what types of standards might be adopted. For example, if you publish an account abstraction standard that requires sending account abstraction transactions to go through my server, then no one will accept it. People only accept things that seem genuinely neutral.

Another thing we are trying is that at Devcon and ETHcc, many people expressed dissatisfaction with having too many competing peripheral events. So we conducted this experiment: during Devcon, we discourage people from holding any form of peripheral events. Peripheral events can take place before or after Devcon, and during Devcon, we encourage people to establish community hubs within Devcon. If your community hub is not just a platform for self-promotion but can facilitate substantial cooperation among different partners in the same industry, then the foundation is more likely to provide support.

So basically, we won’t hold separate Optimism events, Arbitrum events, or Starknet events, but rather hold a Multi-L2 event as part of Devcon. In this way, we encourage cooperation to occur at least on a social level, rather than having one person push everything and then pass proposals as their own. This is also how we are starting to experiment more with standards, trying to establish this norm. I think this also helps address other issues. On one hand, there is the pride people want to support their own things. On the other hand, there is the very human aspect of not wanting to yield to someone else's "strong dominance," or resisting things imposed by others.

The way to address both of these issues is to try to encourage more cooperation, even from the very beginning of a process. As for the third issue, it’s that communication between people is still insufficient; we just need more people and organizations to create forums that can facilitate this kind of dialogue.

Bruce: Thank you. The next question is about Layer 2. Because we all know that Layer 2 solutions play a crucial role in Ethereum scaling. So, how do you view the coordination challenges between Layer 2 and Ethereum? What challenges or strategies exist in aligning the development and governance of Layer 2 solutions with other ecosystems?

Vitalik: Layer 2 initially started in a very independent way, with many people beginning to build their own tech stacks, just trying to create something usable, something that could scale Ethereum relatively quickly. And now the real focus of the ecosystem this year is basically that Layer 2s already exist, they are functioning properly, and they are achieving their intended goals. So everything is migrating towards Layer 2. But how do we ensure that Layer 2 actually feels like and operates as a cohesive ecosystem, rather than feeling like 40 different blockchains? There are specific examples. For instance, if you have tokens on Optimism but have some applications on Arbitrum, then the deposit process, which is transferring tokens from one place to another, becomes very complicated. There are many similar situations where there are too many non-standard aspects, and too many things lack unified standards.

So we have started discussing how to standardize these cross-Layer 2 aspects, which involves participation from Layer 2 teams and wallet teams, and this is an area that is making significant progress.

Bruce: Thank you. Due to time constraints, Vitalik, do you have any other thoughts to share regarding coordination?

Vitalik: When it comes to the topic of coordination, I think there are two important aspects: one is the social aspect, which is the communication between people, and the other is the economic aspect. Interestingly, people like me often tend to overemphasize the economic aspect. However, I feel that in this conversation, we are actually focusing more on discussing the social aspect, which is quite good.

But the economic aspect is also very important; you cannot try to force people to act in ways that severely contradict their incentives. Because, as you can see, if you rely too much on moral pressure, eventually people will feel frustrated and angry, and then rebel against you, sometimes even telling others completely different crazy ideas. So I think in this regard, what we have successfully done as an ecosystem is that we are good at funding small projects. For example, if you have an important public good that needs funding, and you only need $300,000 or $1 million to create a demo, many people will fund it, including the Ethereum Foundation, individual funders, DAO organizations, and individual Layer 2 projects. If they see the demand, many will give you $300,000.

The challenge we face is that when a project transitions from needing $300,000 for a demo to needing $30 million and serving the entire Ethereum user base as a mainstream project, the incentive mechanism basically shifts from complete socialism to complete capitalism. At the $30 million level, there is actually nothing that can truly incentivize you to act in a socially beneficial way beyond market incentives. Because everyone will basically say, "You already have enough money." And we want to fund those projects that would not otherwise receive funding.

Once you are already a company, you have users, you have investors, the next challenge is that when the incentive mechanism is fully marketized, continuing to act in ways that benefit society, such as adhering to standards, not trying to create vendor lock-in, and continuing to remain open source, becomes difficult as the incentive motivation starts to fade. So I think a fundamental challenge is how we can truly improve the incentive mechanism at the $30 million level. For me, this is an unresolved issue. I really welcome people to try different approaches to solve it.

Audience Interaction

Q: I really wanted to ask a question about standards, but some of the things you said really inspired me. You mentioned the sudden shift from small funding to $30 million. While these two situations have different operating mechanisms, isn't this sudden shift itself a problem? Can't there be a more gradual approach? What experiments can we conduct? For example, in Web3, I feel that we are not seeing enough small businesses. It seems that everyone wants to shoot for the stars, or they are just hanging around, enjoying saunas and other fun things. How can we support more small businesses? I'm curious how you think about this issue or what interesting attempts you have seen?

Vitalik: I think there are different types of support. One way to support is more proactive; if there is a promising project, it needs to be provided with a user base, helping the project to be used in a coordinated environment, allowing it to connect with reality and improve. For example, at Devcon, we practiced many things, such as ZK identity tools like Zupass and various on-chain or open-source projects. Part of the goal of doing this is to help projects overcome the barriers of network effects, meaning that no one has heard of them, and no one is using them, which is a form of non-monetary support.

On one hand, in terms of funding support, once a project reaches a high level of development, the problem is basically that we need to find a balance; you want a funding model that is not entirely charitable. Because even at the $3 million level, if you rely solely on charitable funding, it will quickly run out. You need a funding model that is not purely charitable, one that expects to see returns but is also not solely driven by returns. The key question is to find out who is willing to participate in this model in terms of funding.

I believe people are willing; there are many people, even many ETH whales, who hold ETH because they believe in this vision and are even willing to make some small sacrifices, but at the same time, they are not in a hurry to donate everything they have.

On the other hand, the question is what the actual institutions are, what the models are, and what kind of funding models can truly encourage projects to remain open source, stay friendly to standards, and remain decentralized. Ideally, if you end up being very successful, you can also give back to the next wave of projects.

I know there are various projects in the ecosystem that are basically trying to integrate the funding of different large ecosystem participants. The basic theory is that if you have this funding, then at least if they have confidence that everyone is participating simultaneously, they will be willing to invest in projects based on that. But so far, this is definitely still in the early stages.

Q: Another question is about people starting their own projects instead of working on existing ones; it seems there is no incentive. If you are part of someone else's project, the economic returns are also absent. What we really need now is user experience and onboarding guidance. But people, especially venture capitalists, if we want to do a bigger project, they only fund infrastructure. What can we do to improve this situation?

Vitalik: That's a good question. An interesting thing is that our private goods funding and public goods funding have both somewhat failed in this regard. Because if you look at projects like Optimism's retroactive funding or Gitcoin, one of their main weaknesses is that they have basically turned into popularity contests. To obtain a large amount of funding, you must have a high public profile; you must have your own marketing department, as if you are basically a political party. But many people are not interested in this model; they do not want to be full-time, self-promoting politicians. Moreover, this is essentially a model that conforms to existing social status trends, which clearly favors creators over maintainers.

I think at least in terms of public funding, if we can consciously work to create mechanisms to identify and support those maintainers, it could be very helpful. I have seen in public goods funding that retroactive funding community projects are trying to do this, basically trying to identify the downstream dependencies of large projects that everyone considers valuable, identifying dependencies, and then identifying the dependencies of those dependencies. This way, you can support the entire graph.

In Optimism, there are even people who are explicitly trying to dig up and find such projects. For example, the inventors of Keccak (a hash function used by everyone) are scholars who do not know how to showcase themselves on Twitter. A few years ago, they received a $200,000 retroactive grant. So I think at least there should be a conscious effort to create a public information graph that shows who contributed to what and makes it easier to be noticed. This in itself is an important dependency. Because once you have this, you can more easily support any mechanisms that try to improve it.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到support@aicoin.com,本平台相关工作人员将会进行核查。

Share To
APP

X

Telegram

Facebook

Reddit

CopyLink