Source
Ethereum successfully alleviated mainnet congestion and high gas fees through a rollup-centric Layer-2 roadmap, but it also led to fragmentation of the ecosystem.
L2s designed for network expansion have now turned into isolated islands, each with different rules, systems, and barriers.
Liquidity is isolated, users are trapped in bridging between L2s, and developers have to choose whether to develop on Base, Arbitrum, or Starknet.
But in the past year or so, the community has increasingly discussed Ethereum-based rollups as a potential answer to this problem. Rumor has it that Ethereum-based rollups will restore interoperability and composability, allowing the 'Money Legos' concept from DeFi summer to revive on L2 (referring to DeFi protocols that can seamlessly interact with each other). In short, if Ethereum-based rollups deliver on all the features they promise, they will make the Ethereum ecosystem feel like Ethereum again.
The core issue that Ethereum-based rollups attempt to address is the independent sequencers used on L2s—sequencers are the engines that order transactions on the blockchain.
"When I first learned about the L2 scaling roadmap from Vitalik [Buterin]'s blog post, I found it hard to accept because it brought trade-offs," blockchain engineer Teddy Knox told the Magazine.
"Unlike L1, the initial forms of L2 have centralized sequencers with special permissions to order L2 blocks, while L1 has a very large node committee to validate Ethereum."
Ethereum's Surge roadmap achieves 100,000 TPS (Vitalik Buterin).
Centralized sequencers are fragmenting Ethereum's L2.
While centralized sequencers are very fast and earn their operators a lot of money, they also lead to isolation between different L2s. Transactions processed by the sequencer of one L2 cannot easily interact with other L2s, and this lack of interoperability has been a major factor in this year's Ethereum roadmap FUD. (Interoperability between L2s can still be achieved by other means without shared sequencers, but it would be 'asynchronous', i.e., not real-time.)
Ethereum-based rollups (unlike Coinbase's L2 Base), proposed by Ethereum researcher Justin Drake, promise to address this fragmentation issue.
Unlike traditional rollups, Ethereum-based rollups push transaction sequencing back to Ethereum L1, which was the case before the emergence of L2.
"The Ethereum-based sequencing method not only leverages Ethereum's security but also contributes to its revenue and ecosystem cohesion, ensuring deep alignment with Ethereum's mainnet and driving cheaper, faster transactions while directly supporting the network's sustainability," said Daniel Wang, co-founder of Taiko Labs. Taiko is the first project to adopt Ethereum-based rollups and put them into production.
Taiko's revenue returned to Ethereum is about five times that of other rollups with centralized sequencers (Justin Drake).
Composability with Ethereum-based rollups.
This sounds promising, but like anything, there are a range of issues.
To allow users to enjoy the benefits of Ethereum-based rollups, other L2s must adopt them as well. In Taiko's case, they are collaborating with Nethermind's rollup Surge chain, specifically designed to allow users to bridge back and forth between Taiko without going through Ethereum.
But despite being based on the same technology, these two rollups still cannot synchronize combinations, Wang told Cointelegraph at Devcon.
"You need near-real-time validity proofs to prove that changes on both sides are happening simultaneously," he said. "I don't think we're there yet, and as a project, we can't wait until we get there to launch."
Pros and cons of centralized sequencers.
When operated by a single entity or small group, sequencers can order transactions without decentralized consensus or the delays of Ethereum's 12-second block time.
The blockchain trilemma illustrates the challenges of optimizing decentralization, scalability, and security simultaneously.
For many L2 networks, sacrificing decentralization is worth it, as it provides throughput that Ethereum L1 cannot match, although it also brings risks.
"If sequencers have to go down... performance will be affected, or they may successfully censor your transactions unless there are other functions," Knox explained.
Using centralized sequencers brings back many of the problems that decentralization and blockchain are trying to solve, such as censorship and single points of failure—with maximum extractable value (MEV) exploitation being a big issue.
But while these issues may only concern Ethereum's idealists, the big problem for ordinary users is switching between L2s.
0x's smart contract engineer Duncan Townsend stated that the current process of bridging funds from one Ethereum L2 to another "is not a good experience at all."
"Unless you use chain abstraction protocols, the cross-chain user experience in DeFi is terrible," he explained. "If you have fundamentals, you have composability. It doesn't matter which chain your tokens are on; you can buy them cheaply on any chain you need on demand.
If rollups share this sequencing-based framework, tokens and assets should be able to interact directly with each other without relying on separate bridging mechanisms, achieving native interoperability between rollups.
Challenges of adopting rollup-based solutions.
By leveraging Ethereum's validator network to order transactions between multiple L2s, a more unified and efficient ecosystem is created, restoring decentralized transaction sequencing through rollups. Developers can build dapps that run across all participating L2s.
As of November 20, Ethereum has over 1 million validators (Dune Analytics).
However, getting existing L2s to agree to give up sequencer revenue is not easy.
"The transition to sequencer-based solutions has a major barrier to overcome, which is that all these L2 sequencers are making a lot of money," Townsend said.
ETH sequencer revenue from some of Ethereum's top L2 networks.
According to Dune Analytics, as of November 20, ZK rollup ZKsync has earned nearly 40,500 ETH (125.5 million USD) in cumulative sequencer fees. Optimistic rollup competitor Base has earned 20,904 ETH (64.7 million USD), Arbitrum has earned 62,001 ETH (192 million USD), while Optimism has earned 6,916 ETH (21.5 million USD).
Would they really want to give up this revenue out of idealism?
Ethereum-based rollups bring benefits to Ethereum.
While Wang is indeed an idealist, he stated that Ethereum-based rollups help strengthen the security of the Ethereum mainnet, as L2 activity reduces L1 activity, thereby reducing validator income.
"Ethereum-based rollups do provide extra fees, tips, and MEV opportunities for L1 validators, which will encourage more validators to secure the Ethereum chain. Ultimately, this will make all Ethereum-based rollups more secure," Wang of Taiko stated.
According to Growthepie data, Taiko is the rollup that has paid the most fees to the Ethereum chain in the past 30 days, paying $1.29 million in gas fees in the 30 days leading up to November 21, nearly five times that of the second-ranked Arbitrum One.
Taiko is a rent payer in Ethereum's rollup (Growthepie).
This makes validators more profitable and encourages staking, which reduces the circulating supply of ETH and may benefit the price in the long run.
The future of Ethereum: Rollup-based or fragmented?
Rollup-based solutions provide a potential way to unify the Ethereum ecosystem, but returning to the main chain for sequencing also brings old problems.
Wang said that the key trade-off with rollups is that they are limited by Ethereum's current 12-second block time. Arbitrum's runtime is less than 1 second.
"We are working with partners to do (transaction) pre-confirmation, which will no longer rely on shorter L1 block times to provide users with the optimal transaction experience. Users will see their transactions included in a block almost in real-time," Wang said.
Without enhanced interoperability (like that provided by rollups), the Ethereum network is bound to remain fragmented, while DeFi challengers like Solana continue to advance as a unified layer-1, offering users a more seamless experience.
Solana will return to the DeFi space in 2024.
"Essentially, this is about 'in terms of transaction throughput, how large can we make a single chain and a single rollup, and how quickly can we settle so that when liquidity needs to move from A to B, it can get there very quickly, and users don’t have to wait,'" Knox said.
Adopting rollup-based solutions is necessary for success, but not everyone is convinced (Charlie Noyes).
Townsend stated that rollup-based solutions "definitely" provide a solution for a unified ecosystem, but it is still a novel concept, and there is currently no active ecosystem.
"You have to convince these L2 sequencers to give up some revenue streams to join this interoperability ecosystem, which is a barrier," Townsend said.
