The Ethereum community has been embroiled in controversy over the past few weeks, with members warning that the chain will lose its competitive edge unless it resolves some fundamental design issues. A key area of concern has been the fragmentation of layer-2 networks. In recent years, Ethereum has embraced a layer-2 scaling roadmap, which has led to the development of numerous third-party auxiliary networks, known as 'layer-2 rollups', to help scale the base Ethereum ecosystem. Offloading activity to these networks has helped reduce fees and improve speeds for end-users, but it has also resulted in a deeply fragmented ecosystem of layer-2s.
While layer-2 networks all post data back to Ethereum, they often struggle to communicate directly with one another, making it expensive and cumbersome to pass assets and data between them. Moreover, there is the risk of relying on centralized sequencers, which are company-controlled and can pose a single point of failure. As layer-2 chains continue to proliferate, some Ethereum developers are pushing rollup technology that takes a new approach to security and interoperability: 'based rollups'. Based rollups differ from most existing rollups in that they shift execution duties, such as processing transactions, back to Ethereum's layer-1 rather than handling them on a separate layer-2 network.
When someone transacts on a layer-2 rollup, their transaction is processed through a component called a 'sequencer'. The sequencer batches multiple transactions and submits them to Ethereum for settlement. In most rollups today, this sequencer is centralized, meaning a single entity controls the ordering and posting of transactions.
Centralized sequencers are currently a topic of debate in the Ethereum community, as they introduce a single point of failure and can delay or manipulate transactions, raising concerns about censorship and reliability. Based rollups avoid this vulnerability by using Ethereum's built-in sequencing, which relies on a massive community of validators rather than a single centralized sequencer.
In 2022, Ethereum co-founder Vitalik Buterin laid out his vision for a rollup-centric roadmap, which proposed using layer-2 rollups to bypass the base chain's high fees and slow transaction speeds. Different rollups employ different strategies to keep costs down and boost speeds, but they are all designed to uphold decentralization and security. Rollups like Optimism, Arbitrum, Base, zkSync, and Blast have quickly grown to support larger transaction volumes than Ethereum itself.
According to L2Beat, there are currently 140 live layer-2 networks, but the experience of operating between them has become clunky. As Ethereum becomes bigger and layer-2 networks become more integral to its functioning, improving communication between layer-2s, or 'composability', has become more important than ever. Because based rollups share the sequencer from the layer-1 chain, they can call on smart contracts on other based rollups within seconds, making it easier to access and exchange data across layer-2s.
'They effectively share a sequencer with each other and also with the layer-1, and that allows the sequencer to coordinate messages passing between different based rollups, whereas normally message passing happens in an asynchronous fashion,' said Ben Fisch, the CEO of Espresso Systems, in an interview with CoinDesk. Since based rollups all use Ethereum's built-in sequencing, they can interact with one another instantly, in blockchain terms, all within the same Ethereum block. 'You could have, in the span of one Ethereum block, a based rollup withdraw assets, do something in the layer-1, deposit the assets back, do something in the layer-2, and withdraw assets again,' Fisch told CoinDesk. However, based rollups are not without their drawbacks.
A few projects are looking to use based technology, but only one based rollup, Taiko, is currently live. While rollups like Taiko present clear benefits, they will need to overcome some technical hurdles before they can be more widely adopted. One major challenge is proof generation, as based rollups must generate and publish proofs every 12 seconds, matching Ethereum's block time.
Currently, layer-2 rollups use two types of proof systems: zero-knowledge proofs, which finalize in minutes, and optimistic proofs, which take up to seven days to suss out potential fraud. For based rollups to function efficiently, proof generation speeds would need to align with Ethereum's block time, a significant technical leap.
However, Fisch says a breakthrough on this front could be 'imminent'. The other pitfall is Ethereum's block producers, or 'layer-1 proposers', who take over the role of sequencing transactions in based rollups. But their primary motivation is not necessarily fairness, it's profit. 'Layer-1 proposers are not trusted entities that are working in the interest of the layer-2, they are economically motivated to make as much money as they can,' Fisch said.
'So they may confirm some transactions for end-users, and then see an MEV opportunity, which causes them to publish something totally different.' MEV, or maximal extractable value, refers to the practice of reordering transactions to maximize profit, often at the expense of regular users. If proposers manipulate transactions, it could create instability in based rollups.
To address this, developers are working on solutions like based pre-confirmations, which aim to add economic incentives for proposers to act in the interest of rollups. While based rollups may present a promising way to reduce fragmentation between layer-2s, they are not a miracle fix. 'My personal opinion is that based rollups are one part of the solution, they are not the only solution, and not all layer-2s necessarily should or will be based,' Fisch said.