In the midst of the chaos, re-examining Ethereum's positioning and roadmap

Author: Mike Neuder, Researcher at the Ethereum Foundation

Compiled by: Azuma, Odaily Planet Daily

Editor’s Note: This article is a personal opinion piece published today by Ethereum Foundation researcher Mike Neuder, which mainly re-examines Ethereum's positioning, roadmap, value capture, and more.

According to Neuder, while this article represents his personal views, it has received supportive comments from several prominent figures in the Ethereum ecosystem, including Vitalik, during its writing process. In the current climate of widespread controversy surrounding Ethereum, this article may help the market further understand Ethereum's operational thinking and development context.

1. The Essence of Ethereum: Property Rights

Ethereum is fundamentally a protocol about property rights. The Ethereum protocol creates a digital form of self-custodied, permissionless assets that can be transmitted globally and cannot be confiscated or censored. Ethereum's steadfast pursuit of decentralization is precisely to achieve this goal; any compromise on decentralization could become an opportunity for confiscation or censorship, fundamentally limiting the effectiveness of this property rights system.

The three main pillars of this argument are:

• The biggest difference between blockchain and traditional finance lies in property rights, meaning users have inalienable rights regarding the storage and transmission of value;

• On centralized blockchains, certain powerful entities can influence the outcomes of the chain.

• The value stored within the property rights system is directly related to the credibility of that system's property rights.

Overall, a centralized system that may be coerced by centralized censors cannot provide the same property rights effects as a decentralized system, and thus its value is lower. There is a common misconception that Ethereum's decentralization is only valuable in a "third world war" or "post-dollar era," but this is actually incorrect—decentralization is crucial even now.

The attack model of blockchain must consider not only those adversaries who wish to revoke transaction finality but also those more subtle actors who attempt to control economic outcomes without completely destroying the system. This form of attack manifests in various ways, including coercing validating nodes (see the recent employee report from the New York Federal Reserve Bank) and imposing strict KYC/AML requirements on on-chain activities (see details of Blackrock's BUIDL fund).

Solana's claimed goal of achieving "the best, most permissionless, and most accessible financial market globally" and "a globally shared state accessible to anyone without permission" cannot be realized without a clear strategy to maintain the credible neutrality of its block production. If this cannot be achieved, the chain may ultimately become a regulated but transparent financial transmission layer—potentially subject to government censorship. This prospect seems far less attractive, impactful, and valuable than a property rights system centered on "anti-censorship" and "self-custody."

In addition to the validating node set, Ethereum has effectively decentralized many other parts of the ecosystem, including (i) early distribution of ETH based on crowdfunding and PoW mining; (ii) decentralized staking distribution; (iii) meaningful activity and transaction volume on L2; (iv) continuously improving client diversity… Ethereum's decentralization initiatives at the "human" level are also impressive—this network is openly built by individuals and teams from around the world, allowing many to contribute to and invest in the future of the protocol. This true decentralization of value, power, and intellect is difficult to replicate. Furthermore, since most technologies are researched and developed in open-source and public domain environments, Ethereum can also inherit some advantages from ecosystems focused on execution scaling. Technology can be commoditized, but Ethereum's decentralization cannot.

However, it is worth noting that the market, not values, determines the outcomes of these ecosystems. If the marginal costs of decentralization in areas such as L1 execution, user experience, and value accumulation are too high, then the value of the most decentralized blockchain may also decrease. The bullish logic for Solana, Monad, BSC, and Tron is that these blockchains can provide relatively sufficient property rights utility for most users and applications with a lower degree of decentralization.

I tend to believe that, in the medium term, censorship, asset seizure, KYC/AML, node coercion, and other behaviors will lead people to begin questioning the robustness of centralized systems, potentially limiting the market for such systems to a single jurisdiction. In a multipolar world where there is a lack of trust between nations, and they attempt to regulate and monitor their citizens through capital control and financial surveillance, it seems unlikely that global economic activity will naturally occur through a single system. However, Ethereum has a unique claim to credible neutrality, and ETH is an asset that derives value from this credible neutrality; it is also the preferred means of truly permissionless value storage within the system.

In contrast, dollar stablecoins issued by centralized institutions do not provide any property rights guarantees to holders. As Eigenlayer founder Sreeram stated, any holder of USDxxx may face the risk of being "harvested" by Circle or Tether at will—you cannot truly own programmable money in the presence of counterparty risk. I hope that ETH and stablecoins and derivatives collateralized by ETH can become the default option for protecting digital property sovereignty.

2. Ethereum and Rollups

Ethereum's neutrality and anti-censorship characteristics make it the best place for settlement, storage, and expression of value. However, relying solely on L1 settlement does not fully describe Ethereum's roadmap centered around Rollups. Ethereum also serves as the settlement and data availability layer for Rollups.

I view Rollups (and their corresponding Rollup platforms, such as Optimism Superchain and Arbitrum Orbit) as independent territories. Each territory will compete to provide users with what they want—fast transactions, low fees, simple on-chain processes, etc.—but this comes at the cost of a certain degree of decentralization.

I call them territories because, as it stands, the teams responsible for creating and expanding the ecosystem will continue to have significant influence in their respective domains, which seems acceptable. The significance of Rollups lies in their willingness to make trade-offs that Ethereum L1 is not willing to make; if Rollups needed to be as decentralized as Ethereum, why establish this symbiotic relationship in the first place? Rollups rely on the security and decentralization provided by Ethereum, while Ethereum relies on Rollups to expand economic activity within the ecosystem.

An important premise here is that Rollups must reach Stage 2, meaning the rules for upgrading bridge contracts are robust and provide a clear exit path for bridged assets. However, it is important to note that Stage 2 does not emphasize (i) the degree of decentralization of Rollup sequencers; (ii) the destination of fees and MEV (miner extractable value) generated by Rollup activities; (iii) interoperability between Rollup ecosystems.

Stage 2 sets a standard for how Rollups can utilize Ethereum's security and decentralization, while there are not many stipulations regarding other dimensions of Rollup design. I will not engage in the debate about how or when Rollups should implement sequencer decentralization (although I generally agree with Max—I see no motivation for them to do so). Nevertheless, I still agree with Vitalik that this should not be a top priority. I believe the most important tasks for Rollups right now are (i) to inherit Ethereum's security by achieving Stage 2; (ii) to inherit Ethereum's anti-censorship characteristics by having a transparent and efficient forced inclusion mechanism (unlike the current time delays). In my view, these are the key elements, all of which return to the theme that Ethereum can provide the most robust property rights system for L1 and L2 assets.

2.1 Ethereum's Data Availability (DA)

A key element in the design of Rollups is the location of transaction data publication (i.e., which DA service to use). In fact, we can see that some new projects have already started using alternative data availability layers (alt-DA) from the beginning.

I do not support the approach taken by some community members to use social pressure or coercive means to force projects to use Ethereum's data availability layer (DA), as this approach is unsustainable regardless. Instead, we should examine what unique advantages Ethereum's DA service can provide and consider potential network effects. The main advantage of using Ethereum DA is inheriting Ethereum's property rights utility and anti-censorship (do I sound like a broken record…). I like to describe this feature as enabling the "free flow" of Rollup assets. As an on-chain user, if I know my assets will not be confiscated and I can enjoy the same level of self-custody protection, I would be more than willing to conduct most of my daily financial activities on a Rollup that is slightly less decentralized than Ethereum. Based on this, let’s consider the following scenario:

• Scenario Design: For a user bridging ETH to L2 through a standard smart contract bridge, under what circumstances can they withdraw funds back to a different address on L1 through that bridge?

The escape ability of L2 depends on where L2 publishes its data.

• If the L2 is a Rollup based on Ethereum DA and publishes transaction data to Ethereum's Blob, then users will be able to unconditionally use the "escape" mechanism. Because every state update on the bridge contract is backed by data submitted to Ethereum's Blob, this ensures that Rollup users can prove the validity of withdrawals and utilize L1 for transaction packaging (they always retain sovereignty over L2 assets).

• If the L2 chooses to publish transaction data to other DA solutions, then the "escape" mechanism will only be available when Rollups are active. By publishing L2's transaction data to different chains, the state updates on the bridge contract on Ethereum need to be associated with the availability of transaction data on the alt-DA chain. In other words, if someone publishes an invalid state root to the bridge contract without publishing transaction data to the alt-DA chain (commonly referred to as a "data withholding attack"), then L2 users will not be able to prove that their withdrawals are valid, and thus cannot withdraw ETH back to L1 (they will lose sovereignty over L2 assets).

It is worth noting that the second outcome would require L2 to permanently stop producing blocks in order to seize all assets on the standard bridge contract, which is a rather extreme level of intervention. Based on the above scenario assumptions, we can draw a simple conclusion—only Ethereum Rollups that reach Stage 2 and publish transaction data to the Blob can provide the same level of property rights protection for assets bridged to L2.

The above scenario highlights the first network effect of Ethereum's DA service (which I consider to be the best effect): a Rollup that publishes data to Ethereum DA can benefit from other Rollups that also do so, as all assets on the chain will share the same trust assumptions. Sreeram calls this a "trustless composability network effect"—I like this name, although from the user's perspective, the value of this point is still unclear. We are still in the very early stages of L2 adoption, and it seems unnecessary to speculate too much on this. What may be more important right now is to ensure that Rollups have no immediate motivation to use external DA services. The goal of expanding Ethereum DA performance through PeerDAS and Danksharding aligns closely with the vision of providing a large amount of Blob for Rollups, making it a straightforward decision.

In the future, we can imagine that Ethereum DA will generate other network effects. For example, in scenarios where transaction validity can be proven in real-time and pre-consensus (preconfs) is applicable, Rollups using Ethereum DA may have better cross-chain user experiences, more liquidity, and more users. These arguments may be somewhat futuristic, to the point that many people cannot hold firm beliefs about them.

The network effects of DA will become crucial only when we truly view DA fees as a core component of ETH asset value. Let’s delve deeper into this issue.

2.2 Value Capture of ETH

So far, we have not discussed fees and how they bring value to the Ethereum (ETH) asset, even though this has been a major topic over the past few weeks. In the structure of this article, I believe the importance of this point ranks after (1) Ethereum's property rights utility and anti-censorship characteristics as a settlement layer, and (2) Ethereum's role as a DA layer extending security and decentralization capabilities to Rollups. That said, it is still necessary to consider more "direct" forms of ETH appreciation.

Personally, I resonate most with Dankrad Feist (another Ethereum Foundation researcher) regarding DA fees in a recent AMA:

"I do not believe that fees from the Blob will be Ethereum's best value capture mechanism. The data availability market is too unstable—while Ethereum provides the best security, it is too easy to obtain something 'close enough,' which will never be a good way to extract value."

Fundamentally, I do not believe Ethereum DA will have strong user stickiness. The aforementioned network effects are not strong enough to sustain L2's payment of high Blob fees, but I do not see this as a problem. By providing inexpensive DA services for Rollups, Ethereum encourages them to build and develop the volume of economic activity within the Ethereum ecosystem. Therefore, proposals seeking to drive short-term burn rates by raising Blob pricing seem to be completely misguided (again agreeing with Dankrad's viewpoint). Francesco (an Ethereum Foundation researcher) also made great points on this in a recent AMA, outlining how many L2 transactions could occur under the proposed DA expansion.

Another source of value accumulation for ETH is the destruction of execution fees at the L1 level. Max Resnick (an Ethereum Foundation researcher) and colleagues have started a campaign to bring all DeFi execution back to L1; meanwhile, Justin Drake (an Ethereum Foundation researcher) believes that execution at the L1 level "has no future"; my view lies somewhere in between. Here, I would like to quote Dankrad again.

"Ethereum L1 will become the intersection of all these subdomains, and many highly valuable activities will continue to happen on it, generating valuable fees. (Achieving this will require a certain degree of L1 scaling.)"

It seems that valuable activities will always occur on Ethereum, and creating a platform that facilitates a large volume of L2 economic activity will also drive the use of the underlying chain. Therefore, it is necessary to scale the L1 execution layer to facilitate the growth of these activities, but I believe the urgency of this matter is lower than that of "maintaining and improving Ethereum's properties as a settlement layer and DA layer." This again highlights my core point, that Ethereum should maximize economic activity within its platform (including Rollups), and ETH should be positioned as a truly permissionless medium of value storage, rather than merely an income-generating asset.

Focusing on the value storage attributes of ETH naturally raises the question: "Why not choose BTC?"

I will conclude this question with a brief answer.

3. On Bitcoin

There are many topics worth discussing regarding Bitcoin (BTC), especially as it has reactivated research and development ecosystems in areas like ordinals, runes, Rollups, and BitVM, but this article does not intend to delve into these details, nor am I the right person to discuss these topics. Nevertheless, I will point out a few key points closely related to the Ethereum vision mentioned above.

First is the issue of Bitcoin's fixed supply cap of 21 million coins. This revolutionary idea of actively creating digital scarcity is extremely powerful, making Bitcoin one of the most valuable assets globally (with a market cap reaching one trillion dollars as of September 2024, ranking tenth). However, I believe the 21 million cap promise is a fatal flaw in the Bitcoin system because I think the fork choice rules of Bitcoin are inherently "unstable" in the context of gradually decreasing block rewards. The common market response to this view is that fee income will become high enough to incentivize honest mining behavior, but I do not agree with this perspective.

The chart below shows the fee fluctuations of the Bitcoin network over the past six years. I do not believe that under such unstable income streams, mining entities can remain profitable. For example, during the two years from mid-2021 to mid-2023, Bitcoin network fees remained below 1 BTC per block. A more optimistic scenario is that most BTC will be held by ETF issuers, who may choose to subsidize mining behavior and continue to earn fees through asset management business models, but this is clearly not the outcome that the cypherpunk spirit anticipates. Furthermore, the belief that fee income will incentivize mining seems to conflict with the mainstream notion of "buy and hold." If everyone is just holding, then where do the fees come from?

Secondly, there is the question of whether Bitcoin could self-disrupt to become a settlement layer and data availability (DA) layer. The most viable answer I have heard regarding solving the fee source problem for Bitcoin is that Bitcoin could serve as the settlement layer and data availability layer for L2 (the paying party). This is theoretically feasible and quite similar to the path Ethereum is taking, but there are two significant differences.

• The core security model of the Ethereum network does not rely on the fees generated from settlement and DA, thanks to Ethereum's issuance mechanism. I even mentioned earlier that I do not believe DA fees are a key component of ETH value. For Bitcoin, continuously generating fees will be a necessary condition for survival, which seems to create a strange loop: "L1's security relies on fees paid by L2, while L2 relies on L1's security."

• Bitcoin has neither a scaling roadmap nor standard practices for upgrading the network. This is both an advantage and a disadvantage. While stability and predictability are core characteristics of the Bitcoin system, they may also hinder Bitcoin's ability to transform into a settlement layer and DA layer. This seems to be a classic innovator's dilemma, as the system may find it difficult to make large-scale improvements due to being too large and successful, such as adding OP_CAT and increasing block size, which are necessary to provide the resources needed for meaningful scaling for L2.

I would be happy to be proven wrong on these points, as my understanding of the Bitcoin ecosystem is relatively limited, and the above views are based on my current understanding.

There is much more to discuss about Bitcoin, but I will leave it at that. BTC has ample reason to be viewed as digital gold—a highly valuable but relatively static asset. I believe ETH will have a more dynamic future, serving as an anti-censorship, programmable value store that supports a larger digital economy by providing permissionless settlement, DA, and execution.

Conclusion

Ethereum is firmly committed to decentralization, with the goal of creating the safest and most anti-censorship foundational platform for on-chain economies. The development blueprint centered around Rollups aims to expand the platform's economic activity without sacrificing the key characteristics of the settlement layer. As a DA layer, Ethereum can provide Rollups with a low-cost and highly trustless solution, allowing them to attract more users without sacrificing the sovereignty of user assets by reducing decentralization to a certain extent.

I agree with Myles Oneil's view that regardless of how the specific mechanisms of value capture change, the value of ETH will increase with the growth of economic activity within the ecosystem—therefore, it is still too early to discuss optimizing value capture. Finally, while I believe that maintaining settlement properties and expanding data availability are the most important characteristics in the roadmap, I also agree that scaling the L1 execution layer should proceed in parallel, which needs to be built on the technology and innovation across the entire field.

Fundamentally, I believe the value of ETH primarily comes from its ability to serve as a global, permissionless medium of value storage, and the value accumulation story we discuss, while closely related to the expansion of the ecosystem, should prioritize long-term user and developer growth over short-term focus on token mechanics. The development blueprint centered around Rollups is very reasonable: first settlement, then DA, and finally L1 execution—let's proceed in that order.