L2는 여전히 이더리움을 확장하고 있는가?

Ethereum Korea One: Genesis · Day 2 · 2026-09-29 · 38:27

If L1 is scaling faster than expected, what are L2s for? At Ethereum Korea One: Genesis, leaders from Offchain Labs, Optimism, and StarkWare debate whether L2s are still scaling Ethereum.

YouTube에서 보기

연사

Ed Felten, Co-Founder & Chief Scientist, Offchain Labs
Chris Andreola, Senior Director, Commercial Strategy, Optimism
Ilia Volokh, Core Team Lead, StarkWare
Seungmin Jeon, Core Engineer, Sunnyside Labs (모더레이터)

세션 소개

Moderated by Seungmin Jeon of Sunnyside Labs, this panel starts from Vitalik Buterin's argument in February 2026 that the picture of L2s as Ethereum's "branded shards" is not working in reality. Ed Felten, Chris Andreola, and Ilia Volokh discuss how to measure L2s' contribution, whether Stage 2 is still the right yardstick, what Ethereum gets from successful L2s, and why Circle and Stripe are launching their own L1s, Arc and Tempo.

핵심 요약

Ed Felten said that by gas or transaction count, L2s collectively handle far more traffic than Ethereum L1, and warned against the trap of "comparing the current state of one thing against the hoped-for future state of another thing." Layering, he argued, is why the internet succeeded and part of why Ethereum is succeeding.
On Stage 2, Felten noted that it requires the protocol to block governance-driven changes faster than 30 days, a standard even Ethereum L1 would not meet. What matters is accountable governance with no centralized point of control. Ilia Volokh called the idea of distilling any group's governance into code "just a fantasy," while Chris Andreola, who expects L2s to move past Stage 2 within one to two years, said some partners pursue Stage 2 so they have a defensible position if a government asks them to act onchain.
On the criticism that L2s pay Ethereum little in blob fees, Volokh said the gap is by design, since blobs were meant to cut L2 costs, and suggested buyback mechanisms like Hyperliquid's or linking staking across layers. Felten said Ethereum once underpriced data availability but is now closer to a fair price, and that Arbitrum stack chains hold over a million ETH in escrow, close to 1% of supply.
Felten called using ETH as the native and gas token the most important thing L2s have done for ETH demand. Andreola added that the distribution of companies like Coinbase and Robinhood brings holders and users who end up aligned with Ethereum.
Looking back from five years out, Volokh said "users care about products" and that chains are almost inconsequential. Starknet shipped native account abstraction early, but it did not bring a flood of Face ID users because "products were missing." Felten expects user experience to drive design deep into the stack, as L2s did by cutting latency.
Asked why Circle and Stripe chose their own L1s, Felten said large stake-based consensus is expensive, so it makes sense to pay for it once on Ethereum; building separate L1s reflects a desire for control, and he believes the open, standards-based approach wins in the long run. On adoption, all three stressed partners' distribution: "We thought if we build it, they'll come, but they don't, so we need to bring them."

챕터

← 영상 목록 · EK1 Genesis 행사 기록