Vitalik: Unless L2 can be "extremely scaled" or innovated, the original vision is no longer applicable.

PANews reported on February 3 that Ethereum co-founder Vitalik Buterin stated in an article on the X platform: "With the Ethereum mainnet (L1) scaling itself and the gas limit plan significantly increasing, L2, originally envisioned as 'brand sharding,' is no longer necessary. L2 cannot fully meet the characteristics that 'brand sharding' should have, and some projects may even remain in Phase 1 indefinitely to meet customer regulatory requirements. Therefore, L2 should be repositioned, no longer just pursuing scaling, but providing unique additional functions, such as privacy-dedicated virtual machines, application-specific efficiency, extreme scaling, non-financial application design such as social/identity/AI, and low latency or built-in oracles."

Best practices for L2 include: maintaining at least the Phase 1 standard (otherwise it's essentially just a standalone L1 with a bridge), achieving interoperability with Ethereum as much as possible, and exploring integration with native Rollup pre-compilations (ZK-EVM verified) to achieve secure, trustless cross-chain operations and synchronous composability. Users should clearly understand the guarantees provided by L2. The L2 ecosystem will offer diverse choices: chains that rely entirely on Ethereum's trust and credibility, as well as chains with varying degrees of interconnectivity, each focusing on different functions or application scenarios, truly achieving innovative additions rather than simply extending L1.