Chainfeeds Introduction:
Ryan Berckmans, in his discussion on Ethereum L2 revenue and Blob fees, assumes that DA costs rise to about 50% of total L2 revenue. Doug Colkitt, founder of ambient, believes that it is impossible for DA to reach 50% of L2 fees because the value generated by the sorter will always be much higher than that of the DA.
Source:
https://x.com/0xdoug/status/1830090797552636209
Article author:
Doug Colkitt
Viewpoint:
Doug Colkitt: Blockchains are basically selling block space. And because block space is not easily interchangeable between chains, they are close to monopolies. But not all monopolies earn excess profits, the key is whether they can price discriminate against consumers. Without price discrimination, monopoly profits are barely above the commodity. L2 and Solana achieve high throughput and high revenue by using sorter priority as a form of price discrimination. Marginal transactions pay low fees, so they can get huge TPS. While price-insensitive transactions are squeezed dry and pay most of the network revenue. The distribution of 5 randomly drawn blocks from Base L2 shows a clear Pareto distribution, making price discrimination extremely effective: the first 10% of transactions pay 30% of the fees, and the last 10% of transactions pay less than 1%. The problem is that while the sorter makes a lot of money from this, the DA layer does not participate because it has zero price discrimination ability. Ultra-high value arbitrage and 1 Wei garbage transactions pay the same fee to the Ethereum DA because they settle in a single batch. The value of marginal transactions is low, so high TPS can only be achieved if the cost of the median transaction is close to zero. But in DA, the cost of each transaction is basically the same. The DA layer can have high throughput or high revenue, but not both. This makes rollup essentially impossible to scale without collapsing Ethereum network revenue. The rollup-centric roadmap is fundamentally flawed because it gives up the valuable part of the network (sorting) and thinks it can be earned back through the worthless part (DA). I was initially bullish on the rollup-centric roadmap because I thought any reasonable person would recognize the economics of price differentiation and it would run in parallel with L1 expansion. Price-insensitive high-value users would use L1, L2 would specialize in serving marginal low-price users eliminated by L1 fees, and Ethereum would still receive a lot of sorter rent. But Ethereum leadership has repeatedly emphasized that L1 as an application layer is effectively dead and can never scale. Therefore, users and developers have responded rationally, and now the L1 application ecosystem is dying and Ethereum network revenue is dying with it. If you think ETH's long-term value proposition is as a monetary asset, this might be ok. Get more people holding ETH so it becomes a form of money. L2 with zero value accrual to the base layer should help with that; but if you think ETH's long-term value proposition is as a network stake in a widely used protocol (which "I" think is more likely than ETH as money), then you need value accrual. It's clear we screwed up based on bad economic assumptions. 【Original in English】
Content Sources 
