Codex is currently in its first proof of concept iteration, codenamed Dagger. Before Katana (its forthcoming testnet release, expected before the end of 2023) we aim to deliver two additional PoC clients, Scimitar and Longsword.
Dagger: Codex's first PoC and current version.
Scimitar: The PoC-2 client will implement fast erasure coding, the marketplace, and partial durability guarantees—expected end of Q3 2023.
Katana: Codex v1 will offer SNARK-based remote auditing
The Codex team's subgroups are working on the major milestones detailed below to deliver Codex with the qualities described:
- Develop and maintain testnet and devnet
- Release Scimitar
- Develop initial data repair and ZK proving scheme
- Release Longsword
- Deploy Codex v1.0
- Develop the marketplace for integration in PoC-2.
- Develop the marketplace bandwidth incentivization structure.
- Write the initial ZK proving scheme
- Conduct Data Availability Sampling research with the Ethereum Foundation
We believe the above roadmap to be the most logical approach to building a data storage engine with stronger durability guarantees and resistance to censorship than currently offered by Codex's centralised or decentralised counterparts.
First, we must establish a mechanism through which users can initiate a contract with a storage provider to host their data. Then, we can add erasure coding, required for efficient redundancy and auditing. Next, the system will need a mechanism to prove the hosts still have data available. Finally, repairs must be incentivized. Naturally, each layer of sophistication requires testing for economic and technical feasibility.