On August 24, ETA researchers submitted a draft certificationer deposit contract with the aim of enabling the network to move gradually from the existing BLS signature system to a new encrypted signature programme. The proposal is currently in draft form only on GitHub and has not yet been integrated into the official EIP process, nor has it been included in any one of the network upgrades.

The deposit contract presets the interface for the new signature

Currently, the Ether Validator uses a fixed length BLS public key and signature. The new programme proposes to replace the deposit contract with a public key, signature and certificate metadata supporting variable length and to include a “programme identifier” in each deposit to indicate which signature system the certifier uses.

Of these, Scheme 0 will continue to maintain the current BLS format to ensure that the transition period is compatible. Subsequent numbers can be used for antiquator signatures or other new password programmes. The author of the proposal did not specify in the draft which resistance algorithms were specifically used.

This means that the contract is simply a preparation for future migration and that this change will not allow the Etherpo to have full resistance to quantum. The way in which a certifier validates, aggregates and processes new signatures also requires additional modifications on the level of consensus.

Reliance EIP-7685 to transmit certificationer requests

It also plans to adjust the way in which deposit information is transmitted from the executive to the consensus level. Instead of using the old Merkle tree mechanism, the new design used the executive level request to send the certifier deposit data to the consensus level for processing.

This design is based on EIP-7685. The framework provides a common request system for communication between the executive and consensus layers and has been used for multiple types of certification.

  • EIP-6110: Transmitting certifier deposits into consensus level as a request
  • EIP-702: Supporting an enforcement-trigger withdrawal
  • EIP-7251: Support for the consolidation of certifiers

The new contract will extend the architecture to a more flexible encryption data format so that different signature systems do not need to share a fixed set of structures.

Could permanently close new BLS deposits

The proposal includes a phased relocation mechanism. The developer could first deploy a new contract to support BLS, while leaving an entry point for a subsequent new signature programme. Subsequently, the level of the agreement could be permanently closed by a separate decision.

It is designed to be irreversible once the switch is activated. However, the existing BLS certifiers would not automatically withdraw from it, and a separate rule would need to be developed to address issues such as the old certifier key, the exit process, change of documents and migration to the new signature system.

The author is also seeking feedback on how the programme works with existing certificationer proposals such as EIP-702, EIP-7251 and EIP-8282. The issues were said to have been raised at the executive level meeting of the ITA core developers.

Still at the draft stage

The proposal, which is currently labelled as draft and core-proposal, is still awaiting editorial consensus, formal technical review and automated inspection. The editor suggested that EIP-8394 be assigned and that separate discussions be held in the Etheum Magicians community.

However, this number has not yet appeared on the official EIP website in accepted and published EIP. In accordance with the EIP warehouse rules, documents not published on the official network are still to be considered as working drafts, and even if they are subsequently issued in the form of draft, they are not approved for implementation by the Taifung network.

The programme also needs to be technically discussed, well-regulated, securely analysed, client-driven realization and testing before it can really affect the Ether House Master Network, and eventually selected by the core developers for a hard-drive in the future. At present, the official version of the target fork, the test network schedule or the duration of the main line has not been published.