Week 40 2025

Replace bittorrent in status-go

  1. BlockExchange Performance and reliability

    • achieved:
      • Resolved peer discovery failure mode where refresh state remained locked after non-responsive peers
      • Implemented weighted random peer selection favoring less-loaded and more reliable peers for proportional load distribution
      • Validated system behavior under distributed workload using local Kubernetes cluster deployment
    • next:
      • Evaluate peer selection heuristics incorporating reliability metrics beyond load
      • Analyze load distribution characteristics under sustained concurrent file transfer on Kubernetes cluster
    • blockers:
  2. Implement a Codex library so that clients can use Codex directly from their apps

    • achieved:
      • Completed the download feature by adding fetchManifest.
      • Implemented the remaining Data API features: network download, block/dataset deletion, manifest listing, and node space retrieval.
      • Fixed memory and pointer issues.
      • Began extracting the Go wrapper into a separate GitHub repository for easier integration (with tests and documentation).
    • next:
      • Finalize the Go wrapper repository.
      • Create an example project that demonstrates usage of the Go
    • blockers:
  3. Integration with status-go

    • achieved
      • first raw draft of complete integration:
        • community owner: sharing history archives with both BitTorrent and Codex
        • community member: can choose from three modes of operations when downloading the the archives:
          • auto - the first comming, satisfying (in term of recency) archive index dicatates which channel (BT or Codex) will be used for fetching individual archives.
          • codex - BitTorrent archive index ignored
          • torrent - Codex archive index CID ignored
    • next:
      • testing/fixing bugs and fine-tuning the design to make sure nothing is missing
      • documenting
    • blockers:
  4. NAT support

Research future work

  1. Altruistic Mode.
  • Wrapped up our understanding of Swarm incentives;
  • Ongoing research into existing solutions and recent publications related to privacy-first data storage/sharing as we rebuild our knowledge of related approaches.
  • Document: Related work on privacy-first decentralized data storage/sharing
  • Document: Summary of research findings into approaches for lightweight storage proofs with outsourcing erasure coding.
  • Implementation: About 85% of the first prototype of the “outsourcing Reed-Solomon encoding” sub-project is implemented here