Straight Outta Ethereum: the Pluralistic Future of L2s
AI summary of the linked articleThe author argues that Ethereum's L2 competition should give way to a "Based and Boosted" architecture, where L1 validators handle data posting and dApps deploy on L1 for access across all Booster Rollups. The post credits the Taiko team with the Based Rollup concept, noting that Taiko launched as the first Based Rollup three months ago, with Aztec, Rise, Spire, and Puffer following suit. It describes Gwyneth, under development at Taiko Labs, as the first Booster Rollup, which would run L1 contracts unmodified and let transactions be bundled across L2s. The author says this would let users choose chains by values such as speed, yield, or public goods funding. |
 |
Recommended by 2 curators |
4 comments As a hip-hop fan, really enjoyed this one. It was also the first time I've heard about based/booster rollups, and this Google Chrome-like option for more advanced users looks interesting. Wondering what are the potential risks of this solution and what would need to happen for L2s to follow this path as they'd have to share their liquidity with others. mehdib.ethSubmitter Thank you so much! Yeah, I really had fun with the titles! Well, there are multiple challenges: - A Based Rollup has to follow the Ethereum block time and post a block every 12 seconds. At first, validators didn’t have enough transactions to post to be profitable, so Taiko had to sponsor validators and lost a bit of money. - Based Rollups need pre-confirmations to be able to confirm a transaction instantly instead of every 12 seconds, but this needs to be implemented, and most early designs include some centralized entity. Wait and see. I might want to add this part to the article since these are real drawbacks. As for what L2s need to do, basically, to be Based, you just have to get rid of your sequencer and use L1 validators (I can't be more specific than that because it's way above my pay grade, but I can check, lol). To be Boosted, I think it's more drastic, as you have to be "mostly" Type 1, so very Ethereum-compatible. I'm not 100% sure, but Type 4 rollups like zkSync and Starknet wouldn’t work. And I think Optimistic Rollups wouldn’t work as they have no validity proofs, only fraud proofs. But again, it's quite a technical field, and I'm still learning! Thanks for answering the questions! And if L2s get rid of the sequences, who makes the money for running the network? L1 validators? mehdib.ethSubmitter AFAIK yes exactly. The Based Rollup would give up this source of revenue. In this video: They explain that the Booster Rollup can keep a base fee to themselves and the rest to the L1 validators, but not sure about the details. |
More from app.t2.world on Kiwi News |