In his article “Ethereum Welcomes Blobs, Where Do We Go from Here?” Vitalik delves into how blobs are laying the foundation for Ethereum’s future scalability, and proposes four major improvements for L2 protocol development: data compression, optimistic data technology, improvements in execution-related constraints, and enhanced security.
Improving L2 in Four Aspects
Data compression for more efficient use of bytes
Optimistic data technology, using L1 only in special cases to protect L2
Continued improvement in execution-related constraints
Continued enhancement of security
Ethereum Heading Towards a More Comprehensive Decentralized Platform
On March 14th, Ethereum successfully completed the Dencun hard fork, introducing blobs as the foundation for Ethereum’s future scalability. On the 28th, Ethereum co-founder Vitalik Buterin wrote an article titled “Ethereum Welcomes Blobs, Where Do We Go from Here?” to elaborate on Ethereum’s future development journey. He first mentioned that with blobscriptions, the number of blobs per block has increased to a target of 3, leading to a “price discovery mode” for blob fees.
Furthermore, Vitalik stated that the key to Ethereum’s future scalability relies on rollups technology supporting data space and data availability sampling (DAS), with the next step being to use “PeerDAS (conservative version of DAS)” to verify blobs, providing additional scalability capabilities, and introducing more advanced versions of DAS in the future. Vitalik also highlighted four directions for improving the L2 protocol, including data compression, optimistic data structure development, improvements in execution-related constraints, and security enhancement.
Data compression for more efficient use of bytes
Usually, a transaction requires about 180 bytes of data space. Through a series of compression techniques, this data volume can be significantly reduced, ideally compressed to less than 25 bytes per transaction. This compression can greatly enhance data storage and processing efficiency.
Plasma technology allows data to be kept in L2 under normal circumstances, relying on L1 only for security in special cases, providing equivalent security to some applications as Rollups. While Plasma cannot protect all assets, architectures inspired by Plasma can safeguard most assets. Vitalik suggested that L2s unwilling to put all data on-chain should explore such technologies.
With the launch of the Dencun hard fork, rollups using the introduced blobs became cheaper, leading to a significant increase in transaction volume on the L2 network Base, but faced internal gas restrictions, resulting in unexpected fee increases. Vitalik noted that besides expanding Ethereum’s data space, rollups internally need to scale as well.
He mentioned directions for improvement, including “parallel processing (rollups can implement something like EIP-648)” and storage, as well as the interplay between computation and storage, which are important engineering challenges faced by rollups.
Vitalik stated that we are far from a world where rollups rely entirely on code protection. Currently, only the following 5 rollups (of which only Arbitrum is fully EVM-compatible) have reached the “first stage”.
Vitalik mentioned that we need to face this challenge, even though we cannot fully trust complex optimistic or SNARK-based EVM verification code, we can certainly move forward by establishing a security committee, changing the code’s behavior only at a high threshold (for example, the 6-of-8 proposed by Vitalik; 9-of-12 adopted by Arbitrum).
The ecosystem’s standards need to become stricter, and we should only consider projects that have reached at least the first stage as rollups. Subsequently, we can cautiously move towards the second stage: a rollup truly supported by code, where the security committee can intervene only when the code is “clearly self-contradictory”. One way to achieve this security goal is through multiple proof implementations.
In conclusion, Vitalik stated that as Ethereum transitions from the “fast L1 progress” era to a new stage where L1 progress remains important but more gradual, application developers are no longer just building prototypes, but tools to serve millions of people. This requires an adjustment in the ecosystem’s mindset, transforming Ethereum from a purely financial ecosystem into a more comprehensive decentralized technology platform.