AL #034: The 2023 Sequencing Map
Our collective progress in transaction ordering.
Thanks to Tarun Chitra, Ben Basche, Raffi Sapire, Tarrence van As, 0xEvolve for helping me get educated on these topics.
✨ TLDR; If you just want to see the map, please scroll down!
In the previous post, I explained why Sequencing (the service that orders transactions in blockchains) is so important.
To recap:
Revenue. Sequencing is the primary business model for L2s today
Centralization. Sequencing is one of the final remaining points of centralization in rollups.
MEV. Many believe sequencing holds the answer to adversarial transaction ordering.
Intents. As we move to an intent-driven world, transaction broadcast surfaces matter more.
Everyone affected. Sequencing affects searchers, validators, developers and users – pretty much every participant in the ecosystem.
Mystery. Finally, the sequencer implementations and plans of various L2s are well hidden creating an aura of mystery and curiosity.
Now it’s time to look at recent trends in the sequencing value chain and figure out what the future holds.
Wardley Mapping
We will use Wardley Mapping.
If you’re not familiar with how it works, please check the description I did in the ERC4626 map.
In short, we position elements along the horizontal axis where further to the right means more commoditization and further up means closer to the user (as opposed to infrastructure which would be lower down).
Then we connect value chain components that use each other.
An overview of Sequencing
The development of sequencers is still in its nascent stages.
It’s for this reason that I'm enthusiastic about putting it on paper to help encourage Founders to explore what role they may play in the sequencing landscape as either a contributor or consumer or sequencing infrastructure.
Legend
Censorship resistance. Preventing transactions from being selectively ignored by sequencers.
Centralized sequencing. Using a single sequencer service to order transactions. Doesn't compromise safety in rollups with fraud or ZK proofs but introduces many other issues.
Consensus. Infrastructure for forming agreement on leader selection or block ordering.
Data availability. Infrastructure for making rollup data available which may have native integration with a shared sequencer.
Decentralized sequencing. A protocol for reliably achieving liveness & censorship resistance by not relying on a single trusted sequencer.
Fair ordering. Protocols that seek to make transaction ordering deterministic, eliminating validator influence.
Intent mempools. External mempools based on intents rather transactions. An intent can be a set of constraints and objectives that can be resolved as a transaction by a solver.
Lazy sequencing. A shared sequencer that doesn't attempt to resolve state across all its sequenced rollups.
L1 sequencing. Also called “BASED” sequencing. Uses L1 for transaction ordering.
Liveness. Preventing sequencer services from going down.
MEV. Value extraction from adversarial transaction ordering.
MEV auctions. MEVA are auctions which block builders bid on for the chance to capture MEV resulting from a particular transaction order.
Modular sequencing. A Rollup-as-a-service feature that allows rollup developers to pick their own sequencer.
Network performance. Impact of sequencer performance on overall user experience.
PBS. Proposer-Builder Separation. Separates the proposer role from the block builder role in an attempt to reduce the negative impacts of builder centralization.
Priority gas auctions. Mempools where transactions that offer the highest gas rewards end up sequenced early (e.g., Ethereum).
Restaking. Using the existing L1 validator set for other purposes by rehypothecating their staked assets.
Rollup user. Transacting party that benefits from sequencer properties.
Sequencer services. A managed centralized sequencer service.
Shared sequencing. A decentralized sequencer that solves cross-chain atomicity by sequencing transactions for multiple rollups.
Threshold encryption. Encryption of transaction data that can be unencrypted only after the sequencing step through when the set of signatures exceeds a pre-defined threshold.
X-chain atomicity. Support for bundled transactions across rollups.
Let me know what you think and reply with any feedback.