The L1 Convergence

Introduction

Multiple concurrent proposers (MCP), a design in which blockchains achieve leaderless consensus via merging multiple proposers’ chains, is the topic du jour within the Ethereum community. Community critics are reexamining Ethereum’s rollup-centric roadmap and questioning the claim that rollups are extensions of Ethereum. The motivation to redesign Ethereum is to reclaim execution that is otherwise offloaded onto L2s, harden its censorship-resistance properties, and develop legitimate frameworks to analyze L1 economics and model L1 valuations.

The so-called “make Ethereum great again” (MEGA) theme highlights a broader trend in crypto: L1s are converging on similar designs. While treating solo stakers as first class citizens is a noble cause that has differentiated Ethereum from other L1s for some time, researchers and mechanism designers are advocating for higher validator requirements and more innovative consensus. 

In short, given the conflicting and challenging directions for Ethereum’s roadmap, it is important (perhaps now more than ever) to keep an open mind to alternative protocols. At the end of the day, while L1s might have different paths, they’re trying to achieve the same goals: global distribution, higher TPS, and maximum censorship-resistance. 

In particular, directed acyclic graph (DAG) based blockchains are relatively new consensus designs that enable high levels of throughput and censorship-resistance, and they're becoming more relevant among traditional blockchain researchers and designers. As part of an EigenLayer research fellowship, Rohan analyzed MEV in DAG-based blockchains and proposed a method that mitigates PBS in-protocol. Flashbots recently published a report that models the censorship-resistance of leadered and leaderless consensus and finds that DAGs are theoretically the most censorship-resistant, at competitive levels of throughput and latency.