Just caught up on Vitalik's recent thoughts about Ethereum's scaling roadmap, and honestly, it's a pretty fundamental shift in how we should be thinking about the ecosystem.

For years the narrative was all about L2s being the future—offload everything there, let them scale infinitely. But the reality is messier. L2 decentralization has been slower than expected, and now with mainnet throughput actually increasing significantly, the whole premise is getting reconsidered. The conversation is moving from "L2 as the core carrier" to something much more interesting: Ethereum as the world's most trusted settlement layer.

Here's what's actually changing:

Ethereum is shifting to an L1-first paradigm. Fees are dropping as the mainnet scales directly. That original bet on rollups handling all the volume? It's being reframed. L2s aren't going away, but they're no longer treated as identical scaling solutions. Instead, think of them as a spectrum—different trust levels, different security guarantees. Some inherit Ethereum's security fully, others make different tradeoffs. It's more like a constitutional system than a sharding model.

The core value proposition is moving from transaction throughput to settlement sovereignty. ETH's value isn't just about gas fees or blob revenue anymore. It's about being the irreplaceable settlement layer for the entire digital economy—the asset that provides finality, security, and institutional credibility. That's a completely different valuation framework.

What's particularly clever is how this maps onto the US federal system. Before 1787, the 13 states were like independent L2s—each had their own currency, liquidity was fragmented, they couldn't coordinate effectively. Then came the Constitution: unified currency, federal taxation, interstate trade regulation. Suddenly the market unified, economies of scale kicked in, and you got the world's largest economy.

Ethereum is doing something similar with native rollup precompiles and synchronous composability. When all the key cross-L2 interactions settle back on L1, Ethereum becomes the settlement hub again. Liquidity stops being trapped on individual chains. That's when you get real network effects.

Now here's where the valuation gets interesting. Traditional corporate models (P/E ratios, DCF) completely miss the point. Ethereum isn't a company trying to maximize profits—it's infrastructure optimizing for security, neutrality, and scale. It's literally sacrificed protocol revenue (through EIP-4844 and blob data availability) to lower L2 costs and strengthen its position as a neutral settlement layer.

The new framework weights security and settlement premium much more heavily—around 45% of value. Monetary attributes (settlement fuel, collateral for DeFi and stablecoins) get another 35%. Network effects and platform growth are smaller pieces now, and protocol revenue is basically a safety floor for bear markets.

What this means in practice: ETH's pricing is shifting from a cash flow model to an asset premium model. You're not buying it for the fees it generates—you're buying it because it's the most credible settlement layer in existence, with institutional adoption, a long security track record, and increasingly clear regulatory pathways.

The really structural change comes when institutions stop just holding ETH and start using it operationally. Staking ETFs aren't just another product—they're introducing on-chain yields into traditional finance for the first time. That transforms ETH from a non-interest-bearing volatile asset into an allocation asset with predictable returns. That opens the door to pension funds, insurance companies, treasuries. Completely different buyer base.

Looking at where we are now in May 2026, the market is clearly repricing all of this. The volatility, the deleveraging, the pessimism—it's not a collapse of value. It's the market stripping away the speculative growth premium and converging to the structural value supported by settlement certainty and institutional consensus. That's actually a healthier foundation than what we had before.

The question for long-term investors isn't whether Ethereum can still rise—it's whether you recognize which layer of core value you're buying at these levels. If you believe in settlement sovereignty and institutional adoption of on-chain infrastructure, the current range is interesting. That's the real conversation happening underneath all the noise.