Abdelhamid Bakhta, one of the authors of Proposal 5988, presented the proposal during the All Core Devs, a bi-weekly meeting for leading developers to discuss the future of upgrades. The proposal aims to enhance interoperability between EVM and ZK-powered rollups while lowering the cost that is incurred by zero-knowledge scaling solutions.
Bakhta, while representing the proposal, said that Poseidon is a ZK-friendly hash function that makes it very efficient in the ZK context. Bakhta further added that it would enable a set of use cases. The submitted proposal also talked about Poseidon in detail to highlight that it matches or is compatible with all big systems of proof.
Despite a couple of advantages that the upgrade brings to the table, several experts have warned against making any hasty tweaks to the EVM.
Dankard Feist has called the research premature, adding that the code could carry unforeseen security risks. The Ethereum Founder researcher has also said that there is a lot that we don’t know about yet. It is early to enshrine any arithmetic hash functions into EVM because of security concerns; Feist cited in the statement.
Vitalik Buterin spoke about making minimal changes to EVM during a call, stating that it was necessary to provide a stable environment for developers. Buterin’s statement says the applications are written in EVM code. If EVM changes, then there are no changes in the applications.
What makes Poseidon a good candidate for a precompile is its compatibility with all major proof systems. It is currently deployed by Polygon, StarkWare, and Loopring.
Assuming the proposal is implemented, it will bring improvements in terms of settlement, scalability, and privacy. While settlements would be much faster, scalability and privacy will also increase for the users. ZK-based solutions are mostly compared with optimistic rollups trade; they are credited with improving the EVM-compatibility ability for diminishing scalability.
In other words, developers can port their code from the Ethereum mainnet to optimistic rollups. The same is not the case with rollups that are built using zero-knowledge proofs.
The proposal was presented on January 05, 2023, with an emphasis on high expenses associated with ZK-rollups while generating storage proofs. This issue has been attributed to the failure of Ethereum to offer any ZK-friendly hash functions.
The proposal is in its early stages and could be implemented with caution. Experts have already issued a warning by calling the hashing function premature. A hasty tweak could expose the community to security risks. More details are sure to be collected as the proposal moves forward. Meanwhile, Arbitrum and Polygon are enjoying a dominant share of Layer 2 of Ethereum, with reportedly a Total Value Locked at 81%.