Opportunities for the Ethereum Mempool | Joseph Poon - Warp io

Summary

The summary discusses decentralizing Ethereum's mempool to enable secure account abstraction and true peer-to-peer decentralized DeFi. It proposes a new transaction type with a 'revert protect' flag that prevents invalid transactions from entering blocks, allowing them to freely propagate for later validation. This reduces centralized mempool providers' influence, increases proposer autonomy, and creates opportunities for innovation, supporting decentralized intents and peer-to-peer mempools for open market formation.

Takeaways

• 😀 The talk discusses how the Ethereum mempool works, the role of proposer-builder separation, and a proposed EIP to make intents more decentralized.
• 👨‍💻 Historically, client nodes decided which transactions went into a block in a P2P network.
• 🤝 Proposer-builder separation exists primarily to handle contentious transactions where multiple parties want the same transaction included.
• 😥 Centralized mempool providers can front-run, ignore, or reorder transactions since there is no transparency.
• 💡 The proposed EIP adds a 'revertProtect' flag so protected transactions cannot be included unless valid.
• 👍 This enables propagating revert protected transactions on untrusted P2P networks without risk.
• 🔒 Default node behavior is unchanged, so there is no increased denial of service risk.
• 🌐 The goal is a more decentralized mempool ecosystem with autonomous proposers and no builder trust needed.
• ✨ Use cases include decentralized market formation, P2P mempools, and decentralized intents.
• 👏 The EIP could help push Ethereum towards more decentralization in the future.

Q & A

What is the main problem that proposer-builder separation is trying to solve?

It is trying to solve the problem of contentious transactions, where many people are trying to get the same transaction into a block but only one can succeed. This leads to inefficiencies and the need for centralized mempool providers.

What mechanistically are intents in Ethereum?

Intents are invalid transactions today that may become valid later, which need to be held in the mempool. They allow you to propagate an incomplete transaction along the mempool before it can actually be valid.

How does revert protection in the proposed EIP work?

The proposed EIP introduces a new revert protect flag. If a transaction has this flag set, it cannot be included in a block if it would revert. This prevents the transaction from being denied from the mempool due to potential revert.

How can the proposed change enable greater proposer autonomy?

By allowing transactions with revert protection to be freely propagated, proposers can build their own mempool of transactions to include. They don't have to rely on centralized providers filtering transactions.

Why can't builders easily share transaction data today?

Builders need to promise not to include transactions that will revert, so they can't share those transactions with other builders. The proposed change allows transactions to be shared regardless of potential reverts.

How does account abstraction relate to the proposed change?

Account abstraction bundles multiple transactions to save gas. Today it relies on centralized builders because reverting one transaction reverts them all. The proposed change makes account abstraction more decentralized.

What is the benefit of a public peer-to-peer mempool?

A public P2P mempool allows open decentralized market formation around which transactions get into a block, e.g. for an NFT drop. Users can directly compete via gas prices rather than relying on centralized providers.

How does the proposed EIP address denial of service concerns?

Default nodes will not relay transactions with revert protection. Only alternate mempools with additional protections like priority queues and flow control will manage these transactions.

Why are intense/intents centralized today?

Intents rely on posting to a centralized server that promises to hold and eventually include the transaction. The proposed change allows intents to use decentralized P2P mempools instead.

How could the proposed change lead to more intense innovation?

By making intents more decentralized and enabling anyone to build P2P mempool infrastructure, it opens up opportunities for intense innovation on top of this new platform.