Occasional Ethereum takes
Occasional Ethereum takes
Subscribe to everdred
Subscribe to everdred
Share Dialog
Share Dialog
<100 subscribers
<100 subscribers
Nearly four years since the Ethereum rollup-centeric roadmap proposal [1], we now live in a modular, multichain world. Sub-penny L2 transaction fees have enabled a new wave of web3 applications. It is now cheap and trivial for application developers subsidize censorship-resistant compute on behalf of users. Of course, this phase of Ethereum’s development is not without growing pains. A complex backend architecture is now surfaced to the end user, increasing friction and fragmentation. While there are dozens of EIPs poised to alleviate aspects of today’s UX ailments, one design vision, in particular, provides an optimistic endgame for Ethereum:
Universal Synchronous Composability
Today’s Ethereum user is still technically sophisticated with a high tolerance for complexity and pain. Users are forced to keep track of which applications support which networks and where their assets live. If they must migrate funds, there are many bridging options with various security considerations.
To maximize application support, users are encouraged to spread funds and account state across dozens of L2s, fracturing their liquidity, juggling painful UX, and incurring fees with every step.
Universal Synchronous Composability (USC) is not one specific improvement, but a broader vision for abstracting away much of today's complexity. USC seeks to combine the economic security of the L1 with the scalability of L2s, bundled together with the user experience of a single network. Phrased another way, USC takes Ethereum’s modular, rollup-centric design and re-introduces the desirable UX of a monolithic chain.
USC is universal in that it connects Ethereum and all of its rollups, enabling multi-chain transactions. A user is no longer required to select a specific L2 on which to transact, they connect to Ethereum and the rest fades away.
USC is synchronous in that these multi-chain transactions are settled atomically, updating a user’s state across many networks simultaneously via a single intent. Bridging becomes mostly unnecessary, the protocol and application layers meet the user’s assets where they are.
USC is composable in that interoperable protocols are free to connect, combine, and call functions in contracts across all networks. Fractured liquidity becomes shared and accessible.
UX benefits aside, USC leads us toward a more positive-sum ecosystem. L2s must no longer compete with the L1 over liquidity or user state. New L2s must no longer cannibalize the network effects of existing rollups. USC allows us to ride Metcalfe’s law toward a more valuable, cooperative ecosystem while retaining a diverse plurality of EVMs. [3]
So how might we achieve this?
He who controls the sequence, controls the rollup.
- me
Today’s rollup sequencers are centralized, monopolize L2 MEV, and are capable of censorship. The first major step toward USC is to replace this process with a more decentralized design.
How do we ensure L2 transactions retain the same security, censorship resistance, and liveness guarantees as the L1? Two major solution categories have emerged.
![credit: Justin Drake [4]](https://img.paragraph.com/cdn-cgi/image/format=auto,width=3840,quality=85/https://storage.googleapis.com/papyrus_images/f87ff27875804264521c7edfa804cb66.png)
A shared sequencer is a rollup-agnostic abstraction. In practice, this looks like a Layer 1 validator network tasked with sequencing transactions not for itself, but for other L2s. These L2s communicate with the sequencer network via JSON-RPCs while sequencers communicate internally via peer-to-peer messages. The design is familiar, some utilizing the same BFT consensus algorithms that have existed for years. [5]
Shared sequencing is the first step, but bootstrapping a new L1, even via restaking, isn't ideal. Worse, some designs might only enable cross-L2 multi-chain transactions. A more USC-friendly solution would be one in which the existing Ethereum validator set could get ETH L1 slots in on the action.
Based sequencing is a shared sequencing design based entirely on the Ethereum L1. Rollups already rely on Ethereum for settlement and data availability, so shared sequencing would be a logical new service offering.
L1 block proposers could opt into becoming a “sequencer” offering preconfirmations, assurances promising transaction inclusion in a specific L2 slot of a based rollup (ex: 🔢 is pictured above). A preceding, non-sequencing proposer, or “includer”, is then free to settle the sequenced L2 transactions in an earlier L1 slot (ex: ⤵️ pictured above), but the user has economic assurance of their specific L2 slot inclusion and L1 settlement by the sequencer’s slot. Should the sequencer fail to include and settle the transaction in the promised slot, they incur a slashing penalty against additional posted collateral. The sequencer is therefore incentivized to only make preconfirms they expect preceding includers to accept. [4]
This approach is functionally very similar to Execution Tickets [6] but designed for the L1 <> L2 interoperability necessary for USC.
Real-time settlement is the other necessary USC dependency. Optimistic rollups rely on fraud proofs and zk-rollups rely on SNARK or STARK proofs to achieve finality. These proofs are computationally intense and very slow to produce today, taking minutes or even hours. If proof generation could be reduced to <12 seconds, L2 transactions could theoretically be settled within a single L1 block.
Proof latency < slot time is the dream.
Today, almost a dozen teams are researching SNARK proving ASICs capable of such orders of magnitude latency reduction. Some have already manufactured prototypes, making this dream a near-term reality! [7]
With based sequencing offering better composability and real-time settlement reducing latency, a proposer could include transactions making function calls between L1 and L2 applications and include them atomically. These improvements enable instantly settled, multi-chain transactions:
![credit: Justin Drake [4]](https://img.paragraph.com/cdn-cgi/image/format=auto,width=3840,quality=85/https://storage.googleapis.com/papyrus_images/5b66c8dd6e94525fe6595a15645a765f.png)
The full vision of Universal Synchronous Composability is at least another full market cycle away, but you can take comfort in knowing the nostalgic, simple UX of DeFi Summer may someday return to an Ethereum node near you. To learn more about these tracks of work, follow the EF’s ongoing Ethereum Sequencing and Preconfirms call series [8].
[1] V. Buterin, "A rollup-centric ethereum roadmap," Fellowship of Ethereum Magicians. [Online]. Available: https://ethereum-magicians.org/t/a-rollup-centric-ethereum-roadmap/4698. [Accessed: May 19, 2024].
[2] CoinGecko, "Top Bridged USDC Coins by Market Cap," [Online]. Available: https://www.coingecko.com/en/categories/bridged-usdc. Accessed: May 19, 2024.
[3] Bankless, "Fixing Fragmentation with Justin Drake," YouTube, Feb. 5, 2024. [Online]. Available: https://www.youtube.com/watch?v=MnsjUZo7RRI. [Accessed: May 19, 2024].
[4] J. Drake, "Ethereum Sequencing," presented at the Ethereum Sequencing and Preconfirms: Call series, Ethereum Foundation, Feb. 16, 2024. [Online]. Available: https://docs.google.com/presentation/d/1v429N4jdikMIWWkcVwfjMlV2LlOXSawFCMKoBnZVDNU/edit#slide=id.p. Accessed: May 23, 2024.
[5] R. Chen, "Intro to Shared Sequencing: Pooling resources to ensure better network decentralization and security," Medium, May 15, 2023. [Online]. Available: https://medium.com/@richardchen_81235/intro-to-shared-sequencing-1622d1fd51c9. Accessed: May 23, 2024.
[6] M. Neuder, J. Drake, "Execution Tickets," EthResearch, Dec. 23, 2023. [Online]. Available: https://ethresear.ch/t/execution-tickets/17944. Accessed: May 19, 2024.
[7] J. Drake, "ZK11: SNARK proving ASICs," presented at the Zero Knowledge Summit 11, Athens, Greece, Apr. 10, 2024. [Online]. Available: https://www.youtube.com/watch?v=URCH2d1cdyg. Accessed: May 19, 2024.
[8] Ethereum Foundation, "Sequencing and Preconfirmations Call,” YouTube. Available: https://www.youtube.com/playlist?list=PLJqWcTqh_zKHDFarAcF29QfdMIUpReZrR [Accessed: May 19, 2024].
Nearly four years since the Ethereum rollup-centeric roadmap proposal [1], we now live in a modular, multichain world. Sub-penny L2 transaction fees have enabled a new wave of web3 applications. It is now cheap and trivial for application developers subsidize censorship-resistant compute on behalf of users. Of course, this phase of Ethereum’s development is not without growing pains. A complex backend architecture is now surfaced to the end user, increasing friction and fragmentation. While there are dozens of EIPs poised to alleviate aspects of today’s UX ailments, one design vision, in particular, provides an optimistic endgame for Ethereum:
Universal Synchronous Composability
Today’s Ethereum user is still technically sophisticated with a high tolerance for complexity and pain. Users are forced to keep track of which applications support which networks and where their assets live. If they must migrate funds, there are many bridging options with various security considerations.
To maximize application support, users are encouraged to spread funds and account state across dozens of L2s, fracturing their liquidity, juggling painful UX, and incurring fees with every step.
Universal Synchronous Composability (USC) is not one specific improvement, but a broader vision for abstracting away much of today's complexity. USC seeks to combine the economic security of the L1 with the scalability of L2s, bundled together with the user experience of a single network. Phrased another way, USC takes Ethereum’s modular, rollup-centric design and re-introduces the desirable UX of a monolithic chain.
USC is universal in that it connects Ethereum and all of its rollups, enabling multi-chain transactions. A user is no longer required to select a specific L2 on which to transact, they connect to Ethereum and the rest fades away.
USC is synchronous in that these multi-chain transactions are settled atomically, updating a user’s state across many networks simultaneously via a single intent. Bridging becomes mostly unnecessary, the protocol and application layers meet the user’s assets where they are.
USC is composable in that interoperable protocols are free to connect, combine, and call functions in contracts across all networks. Fractured liquidity becomes shared and accessible.
UX benefits aside, USC leads us toward a more positive-sum ecosystem. L2s must no longer compete with the L1 over liquidity or user state. New L2s must no longer cannibalize the network effects of existing rollups. USC allows us to ride Metcalfe’s law toward a more valuable, cooperative ecosystem while retaining a diverse plurality of EVMs. [3]
So how might we achieve this?
He who controls the sequence, controls the rollup.
- me
Today’s rollup sequencers are centralized, monopolize L2 MEV, and are capable of censorship. The first major step toward USC is to replace this process with a more decentralized design.
How do we ensure L2 transactions retain the same security, censorship resistance, and liveness guarantees as the L1? Two major solution categories have emerged.
A shared sequencer is a rollup-agnostic abstraction. In practice, this looks like a Layer 1 validator network tasked with sequencing transactions not for itself, but for other L2s. These L2s communicate with the sequencer network via JSON-RPCs while sequencers communicate internally via peer-to-peer messages. The design is familiar, some utilizing the same BFT consensus algorithms that have existed for years. [5]
Shared sequencing is the first step, but bootstrapping a new L1, even via restaking, isn't ideal. Worse, some designs might only enable cross-L2 multi-chain transactions. A more USC-friendly solution would be one in which the existing Ethereum validator set could get ETH L1 slots in on the action.
Based sequencing is a shared sequencing design based entirely on the Ethereum L1. Rollups already rely on Ethereum for settlement and data availability, so shared sequencing would be a logical new service offering.
L1 block proposers could opt into becoming a “sequencer” offering preconfirmations, assurances promising transaction inclusion in a specific L2 slot of a based rollup (ex: 🔢 is pictured above). A preceding, non-sequencing proposer, or “includer”, is then free to settle the sequenced L2 transactions in an earlier L1 slot (ex: ⤵️ pictured above), but the user has economic assurance of their specific L2 slot inclusion and L1 settlement by the sequencer’s slot. Should the sequencer fail to include and settle the transaction in the promised slot, they incur a slashing penalty against additional posted collateral. The sequencer is therefore incentivized to only make preconfirms they expect preceding includers to accept. [4]
This approach is functionally very similar to Execution Tickets [6] but designed for the L1 <> L2 interoperability necessary for USC.
Real-time settlement is the other necessary USC dependency. Optimistic rollups rely on fraud proofs and zk-rollups rely on SNARK or STARK proofs to achieve finality. These proofs are computationally intense and very slow to produce today, taking minutes or even hours. If proof generation could be reduced to <12 seconds, L2 transactions could theoretically be settled within a single L1 block.
Proof latency < slot time is the dream.
Today, almost a dozen teams are researching SNARK proving ASICs capable of such orders of magnitude latency reduction. Some have already manufactured prototypes, making this dream a near-term reality! [7]
With based sequencing offering better composability and real-time settlement reducing latency, a proposer could include transactions making function calls between L1 and L2 applications and include them atomically. These improvements enable instantly settled, multi-chain transactions:
The full vision of Universal Synchronous Composability is at least another full market cycle away, but you can take comfort in knowing the nostalgic, simple UX of DeFi Summer may someday return to an Ethereum node near you. To learn more about these tracks of work, follow the EF’s ongoing Ethereum Sequencing and Preconfirms call series [8].
[1] V. Buterin, "A rollup-centric ethereum roadmap," Fellowship of Ethereum Magicians. [Online]. Available: https://ethereum-magicians.org/t/a-rollup-centric-ethereum-roadmap/4698. [Accessed: May 19, 2024].
[2] CoinGecko, "Top Bridged USDC Coins by Market Cap," [Online]. Available: https://www.coingecko.com/en/categories/bridged-usdc. Accessed: May 19, 2024.
[3] Bankless, "Fixing Fragmentation with Justin Drake," YouTube, Feb. 5, 2024. [Online]. Available: https://www.youtube.com/watch?v=MnsjUZo7RRI. [Accessed: May 19, 2024].
[4] J. Drake, "Ethereum Sequencing," presented at the Ethereum Sequencing and Preconfirms: Call series, Ethereum Foundation, Feb. 16, 2024. [Online]. Available: https://docs.google.com/presentation/d/1v429N4jdikMIWWkcVwfjMlV2LlOXSawFCMKoBnZVDNU/edit#slide=id.p. Accessed: May 23, 2024.
[5] R. Chen, "Intro to Shared Sequencing: Pooling resources to ensure better network decentralization and security," Medium, May 15, 2023. [Online]. Available: https://medium.com/@richardchen_81235/intro-to-shared-sequencing-1622d1fd51c9. Accessed: May 23, 2024.
[6] M. Neuder, J. Drake, "Execution Tickets," EthResearch, Dec. 23, 2023. [Online]. Available: https://ethresear.ch/t/execution-tickets/17944. Accessed: May 19, 2024.
[7] J. Drake, "ZK11: SNARK proving ASICs," presented at the Zero Knowledge Summit 11, Athens, Greece, Apr. 10, 2024. [Online]. Available: https://www.youtube.com/watch?v=URCH2d1cdyg. Accessed: May 19, 2024.
[8] Ethereum Foundation, "Sequencing and Preconfirmations Call,” YouTube. Available: https://www.youtube.com/playlist?list=PLJqWcTqh_zKHDFarAcF29QfdMIUpReZrR [Accessed: May 19, 2024].
No activity yet