ACDE #205: Let the Fusaka Scoping Discussions Begin!

The podcast episode focuses on execution call number 205 and the early discussions surrounding Ethereum's upcoming upgrade, Fusaka. A key update on the Pectra timeline indicates that client teams are on track with testnet software releases for Holeski and Sepolia, with the Holeski testnet upgrade scheduled for February 24th and the Sepolia upgrade on March 5th. Node operators are encouraged to download the new software for automatic validator upgrades.

Pectra is highlighted as the largest upgrade in Ethereum's history, raising concerns about potential bugs. Developers anticipate a minimum of 30 days between the public testnet upgrade and the Ethereum mainnet upgrade, which could occur around April 8th if no issues arise. The ephemeral ephemery testnet is live for testing Pectra, allowing stakeholders to assess their validator setups. As Pectra nears finalization, developers are shifting focus to Fusaka, conducting a retrospective to gather feedback and improve the upgrade process.

Tim, chairing the ACDE calls, expressed a desire for more frequent upgrades, aiming for twice a year. To achieve this, higher testing requirements for Ethereum Improvement Proposals (EIPs) were suggested, with Mario proposing detailed execution layer specification tests. However, concerns about potential bottlenecks for EIP authors and the testing team were raised, prompting ongoing discussions to balance new requirements without hindering the EIP process.

The Ethereum ecosystem faces challenges in prioritizing upgrades due to differing views among client teams and stakeholders, complicating the scoping of forks. The decentralized governance model lacks a central authority to set the agenda. The Fusaka upgrade is anticipated to be significant, with lessons learned from Pectra aimed at fostering sustainable governance and healthy debates. During the scoping discussions, disagreements emerged regarding Fusaka's features, with Tim proposing a goal to finalize discussions within two months post-Pectra upgrade, suggesting that Fusaka should prioritize Peerdas and EOF to minimize disputes.

Developers are focused on implementing upgrades and discussing additional overflow EIPs for the upcoming Glamster Dam upgrade. Nethermind developer Ben Adams supports the inclusion of two EIPs, advocating for more to ensure safe gas limit increases on Ethereum. He clarifies that these EIPs are bug fixes aimed at allowing validators to raise the block gas limit from 36 million to potentially 60 million. Vitalik and other client teams also support increasing the Ethereum block gas limit and the inclusion of these EIPs in the Fusaka discussions.

Tim Byko expresses concerns about expanding the scope of Fusaka and urges concise discussions to enable developers to begin work. He emphasizes the need for the Geth team to clarify their stance on EOF in writing to facilitate progress, rather than stall the consensus process for other EIPs that have majority support. Despite previous opposition to EOF from the Geth team, Tim encourages them to avoid hindering the consensus process for the broader developer community.

The Ethereum governance process reflects diverse opinions on priorities for Fusaka, with many client teams supporting the swift shipping of PeerDAS. While some aspects of Fusaka's scope have reached consensus, disagreements persist among certain client teams or individuals. It is essential that individual dissent does not obstruct the decentralized governance process, allowing developers to focus on delivering meaningful upgrades. A light client developer emphasizes the importance of prioritizing impactful code changes over frequent updates, advocating for multiple essential code changes, such as fossil, SSC, and EPBS. The unique governance structure of Ethereum serves as a significant case study for decentralized systems, with successful governance demonstrating resilience and adaptability. The hope remains for Ethereum developers to finalize the Fusaka scope soon, while discussions will continue regardless of timelines.