Metabased Rollups: Decentralized L3s With New Economic Opportunities and Network Effects

Today, we’re sharing our development of a powerful type of rollup we call “metabased” and a new community-led initiative with the same name.

For updates, follow the Metabased community on both Warpcast and X.

Background

For over three years, Syndicate has been working on technologies to economically empower communities on the internet at scale. We started by building infrastructure for decentralized autonomous organizations (DAOs). This led to our work on many interesting projects like Constitution DAO, Nike .SWOOSH, and Degen Chain—and many discoveries like new community building primitives, innovative community-centric economic models, and recently L3s. 

Over the last six months, our team at Syndicate has been working with L3s on their hardest problems. In doing so, we learned three major things. First, L3s are communities. They each have their own set of stakeholders with unique roles—in ways that resemble DAOs. They also have their own blockchain networks, custom economic systems, and native tokens for gas, staking, governance, and more.

Second, decentralization is the biggest issue for L3s. Decentralization not just for the purpose of liveness, credible neutrality, and censorship-resistance—but importantly, decentralization for the agency and freedom to self-determine how their networks should create, manage, and share economic value with their stakeholders.

This brings us to our last major learning: L3s don’t want to be owned, controlled, or operated by someone else. L3s want to be owned by their communities: people and organizations who care and are most aligned with their network’s long-term vision, values, and success.

And yet despite all of this, there are no current paths for L3s to credibly decentralize their networks—or their ownership, control, and operation—to their communities to socially, functionally, and economically empower them. This is the most important existential problem for rollups, particularly L3s.

Metabased rollups give networks a viable path to decentralization, economic value creation, and community ownership.

Definition

Metabased sequencing is similar to based sequencing, but it unlocks powerful new performance, mechanism design, economic, cost, governance, interoperability, community ownership, and collaboration features—among others—which we outline below.

We define a rollup as “metabased” when its sequencing is determined onchain by a smart contract on a layer above L1 like an L2. “Metabased sequencing” is when a metabased rollup’s sequencers include the rollup’s next block (e.g., L3 block) as part of the underlying layer’s next block (e.g., L2 block via an L2 metabased sequencer smart contract).

Importantly, metabased rollups are not just a technical innovation but primarily an economic one, as they can be designed to maximally align with L3s and their communities. As a result, metabased rollups not only provide a path for how Ethereum scales but also decentralizes to deliver on the promise of building a user and community-owned internet.

Features

• Maximally simple and onchain: Metabased sequencing is simpler and more onchain in its design than other models like centralized, decentralized PoS, shared, or based sequencing that require new offchain consensus layers to solve for the difference in block times between the L2 (1 to 2 seconds) and L1 (12 seconds) and enable key functionality like preconfirmations that are still necessary for based rollups.
• 1,000x+ cheaper → 1,000x+ experimentation: Metabased rollups are dramatically less expensive to start and run than rollups using other sequencing models. Compared to rollups with centralized sequencers, metabased rollups are 100x to 1,000x+ less expensive to operate, making it possible to run an L3 for as low as a few dollars a month. Cost reductions of this magnitude unlock entirely new possibilities and can transform why, when, and how rollups are used. It can also significantly increase the amount of experimentation, iteration, and innovation in mechanism design beyond what we see today in the areas of rollup economics, governance, decentralization, and more.
• Sovereignty and sequencing model flexibility: Metabased rollups not only maintain the option of sovereignty but also flexibility in who they allow and/or delegate sequencing to via their metabased sequencer smart contract. This is an important feature of metabased rollups, as L3s care about who they are owned, controlled, and operated by. While metabased rollups can delegate sequencing to anyone like the L2, L1, or a shared network, metabased rollups have the agency to self-determine who they delegate sequencing to, which is an important difference compared to centralized, shared, and based sequencing models that cede control and the associated economics to someone else who may not be aligned with the network or its community long-term. Metabased rollups can also enable and evolve through the full spectrum of decentralization models by simply changing the metabased sequencing smart contract from centralized, to permissioned, to sufficiently decentralized, to fully decentralized and back.
• Economic alignment with L3s and communities: As a result of their sovereignty, onchain simplicity, and sequencing flexibility, metabased rollups are able to more easily design, experiment, and iterate on economic mechanisms and systems that maximally align with their networks and communities. Rather than ceding sequencing and economic control to someone else, metabased rollups are able to fully align their sequencing models with their economic, incentive, and rewards systems for their unique stakeholders in ways that are not possible with centralized, shared, or based sequencing. Metabased rollups have full autonomy over how transactions are sequenced, who can sequence transactions, and how value from the sequencer is distributed. This gives metabased rollups and their communities a level of economic agency and control not possible with centralized, shared, or based sequencing models.
• New economic opportunities and mechanisms: Because metabased rollups are more onchain, many orders of magnitude cheaper, and can be maximally aligned with their communities, new economic mechanisms and opportunities can be unlocked that are not possible with centralized, shared, or based sequencing models. Far beyond governance tokens, base fees, and sequencing revenue fee sharing, metabased rollups unlock the creation and customization of new economic mechanisms in the areas of sequencing auctions, native token staking models, and Maximal Aligned Value (MAV) income—a new type of “good, positive-sum MEV” uniquely possible through metabased rollups’ economic mechanism and sequencer flexibility—for networks and their communities. As a result, metabased rollups are able to access new revenue and economic opportunities, making them more sustainable, valuable, and attractive than rollups that are not metabased.
• Cross-chain atomic interoperability: Another major benefit from sequencing with a layer above L1 is that it makes cross-metabased rollup atomic transactions possible, unlike other sequencing models with separate offchain consensus layers with faster, but non-atomic cross-chain transactions. Same block cross-metabased rollup transactions not only enable better user experiences across applications on different networks, but they also unlock new cross-chain transaction and blockspace auction markets, cross-chain MAV and MEV, positive-sum multi-metabased rollup networks that collaborate and interoperate with each other to increase their collective networks’ GDP, and more. These economic opportunities are uniquely possible with metabased sequencing.
• Minimal lock-in: With sovereignty and flexibility around who is allowed to sequence their networks combined with their maximally simple and onchain design, metabased rollups minimize lock-in compared to other sequencing models that require additional offchain consensus layers for faster finality and/or preconfirmations. This gives metabased rollups and their communities more agency and power in the stack, as metabased sequencers can be more easily forked and modified. This ultimately means that more value is accrued and retained by metabased rollups and their respective communities—not by other layers lower in the stack.
• Greater throughput and speed: By sequencing with a layer above L1 like an L2, metabased rollups are able to achieve higher throughputs and speeds than based rollups. Without the dependency and overhead of an additional offchain fast finality and/or preconfirmations consensus layer, metabased rollups provide a simpler, more efficient path to scaling Ethereum and its applications and communities.
• Credible neutrality and sufficient decentralization: With a more onchain design, metabased rollups provide greater transparency, trust, and neutrality than other sequencing models that require separate offchain consensus layers. Metabased rollups also enable the full spectrum of decentralization models. This makes metabased rollups a viable solution for scaling Ethereum via thousands or millions of networks in the future, each with their own configurations and high degrees of transparency.

Limitations

• Liveness: Metabased rollups are dependent on the liveness of the L2, which may have the same guarantees as a decentralized PoS or shared sequencing model but lower guarantees than based rollups that leverage L1. Also, while they have fewer dependencies than centralized, decentralized PoS, and shared sequencing models, metabased rollups have more dependencies than based rollups as they rely on both the L2 and data availability layer. In the case of a liveness failure of the L2 metabased sequencer and an escape hatch is used, this results in weaker settlement guarantees, greater risks of censorship, escape hatch exploitation, and mass exits, as well as gas inefficiencies. Furthermore, unlike based rollups where L1 searchers and block builders are already incentivized via MEV to include a based rollup’s blocks on the L1, sequencers of metabased rollups are not inherently incentivized to include a metabased rollup’s blocks on the L2 without an appropriate economic mechanism design. For these reasons, metabased rollups may not be a preferred model for some applications or networks until 100% uptime and liveness can be guaranteed.
• Economic security and decentralization: Similar to its liveness, metabased rollups are also dependent on the security and decentralization guarantees of the L2. These guarantees are higher than a centralized sequencing model, the same as a shared sequencing model, but lower than a based sequencing model that inherits the security and leverages the searcher-proposer-builder infrastructure of L1. As a result, more trust is placed in the L2 and supporting infrastructure that sequences the metabased rollup, which until they are sufficiently or maximally decentralized, security and censorship risks remain.
• L1 economic alignment: Unlike based rollups where MEV originating from them flow to the L1 and L1 searchers and block builders (in turn increasing the economic security and value of L1 via MEV burn), MAV and MEV of metabased rollups flow to the L3s and their communities based on their self-determined sequencing mechanisms and models. This tight economic alignment with L3s and their communities is one of the most powerful features of metabased rollups, but it also comes with the tradeoff of less direct economic alignment with L1.

Open Questions

• Mechanism designs: Since metabased rollups unlock a new wave of sequencer mechanism design and experimentation—including sequencer selection, monetization, governance, value sharing models, and more—it’s currently unknown what mechanism designs metabased rollups will pursue, find useful, and derive value from that are worth their inherent tradeoffs and limitations outlined above.
• Security, decentralization, and liveness: For metabased rollups and sequencing to work at scale, their risks related to security, decentralization, and liveness will need to be addressed similar to decentralized PoS and shared sequencing models. Currently, it’s unknown what economic security and operational models will be best suited to support metabased rollups at scale.
• Interoperability: Given the array of interoperability challenges and solutions available, it’s unknown what interoperability frameworks and technologies metabased rollups should utilize beyond the cross-metabased rollup atomic interoperability that is natively enabled through metabased sequencing. As metabased rollups will want to interoperate with non-metabased rollups and networks—whether L1s, L2s, or L3s—it’s unclear how they’ll do this easily, scalably, or efficiently. 
• Offchain services: While more processes can live onchain with metabased rollups, it’s unknown whether additional services requiring separate offchain consensus layers like preconfirmations, instant receipts, MAV/MEV, or more will be needed for different metabased sequencing mechanisms, like onchain blockspace auctions, to work at scale. This is an area where metabased rollups will need to run live tests to fully research and validate the feasibility and need.
• Compatibility: As metabased rollups are still in development, it’s unknown whether modifications (and how many) will need to be made to make metabased sequencing compatible with existing rollup frameworks, node software, data availability layers, and more. It’s also unclear whether modifications will need to be made upstream and what the respective costs and tradeoffs are.

First Principles

In our development of metabased rollups and sequencing, it’s important to remember that managing rollups onchain via smart contracts is not a new idea. In fact, it’s where most rollups started. Prior to upgrading to Bedrock, Optimism used a contract called the Canonical Transaction Chain to manage rollup state. This appended new rollup blocks via a smart contract, which any node could read from to get the L2 chain state.

Here’s an example from the Canonical Transaction Chain smart contract that will look unsurprising if you’ve browsed L2 block explorers:

This approach had a lot of upsides. One benefit is that the L1 smart contract can enforce allowed sequencers, ordering, and other rules for adding new blocks. This acts as an incredible foundation for sequencer mechanism design. It also makes it much easier to progressively decentralize the rollup.

A second benefit of Optimism’s pre-Bedrock approach is that you can derive the entire rollup state from a single smart contract, which is likely more resilient than post-Bedrock calldata if EIP-4444 (which prunes data older than a year) is implemented down the road and far more resilient than EIP-4844 blobs that have already begun to expire. For the largest rollups, missing state data isn’t as big of an issue since it’s replicated across thousands of nodes. But for everyone else, a rollup’s state data may be stored in only a small handful of nodes. Sometimes these nodes are centralized and run by a single provider, where one outage or failed credit card payment is the difference between a rollup retaining its state or losing that data permanently and suffering week-long outages and several hundred thousand block reorgs.

At this point, nearly all L2s have universally moved away from storing rollup state in smart contracts to storing rollup state in expiring blobs. This significantly increases the risk of state data loss for every rollup except for the largest networks. Why is this taking place? The biggest driver is gas costs, which are substantial on L1. Blobs are much cheaper than calldata, and calldata is cheaper than storing state data in smart contracts. This is lower risk for the largest rollups. But for everyone else, significant systemic risks are being introduced, especially with greater centralization and lower resiliency in rollup nodes.

For this reason, we’re bringing back one of the original ideas for rollups: storing state data in smart contracts on the underlying layer via metabased sequencing. By moving based sequencing from L1 to an L2, we can alleviate the gas cost issue while significantly increasing the resiliency and decentralization of metabased rollups. Long-term state data storage is no longer dependent on L3 nodes but instead an L2 metabased sequencer that has the same trust and security assumptions as Optimism, Base, Arbitrum, and other large networks. Metabased sequencing makes L3s stronger together by giving them the collective resiliency of a shared L2 metabased sequencer.

The Future is Metabased

We’re building on the latest innovations in rollups and execution clients as we develop metabased sequencing and metabased rollups. We’re leveraging Reth Execution Extensions and benefiting from all of the progress made by Optimism, Reth, Magi, and others. Without these technologies, it would be far more difficult for metabased rollups to be brought to life. We’re grateful to collaborate with and stand on the shoulders of other industry leaders.

We’re already making contributions upstream to the open source ecosystem, including bug fixes and documentation improvements to Reth, op-node, Alloy, and more. We’re also looking for ways to get even more deeply involved in larger open source contributions to benefit the ecosystem to help everyone move toward a more decentralized vision of rollups and scaling Ethereum.

If you like Rust, Solidity, and open source, are passionate about decentralization, and want to bring new economic mechanisms and primitives to life, we’d love to work with you. We’re already in touch with some of the best network builders, developers, and mechanism designers in the industry, and we hope to connect and collaborate with more.

• Follow Metabased on Warpcast and X
• Follow Syndicate on Warpcast and X
• Join our Warpcast channel on L3s
• Connect with Will about metabased design on Warpcast and X
• Contact us at Syndicate