Week 50 2025
Logos Storage offsite
The Logos Storage team held an offsite during the week of November 17–21 in Seville, Spain, together with council representatives Corey, Jacek, and Daniel K. to discuss strategic alignment and the future direction of Logos Storage.
Key discussions included:
- Retrospective over the past four months
- Roadmap planning with three parallel tracks
- Privacy requirements and threat modeling
- Nim versus Rust evaluation
- Anonymous Communication - Tor & Tribler analysis (presented by Mo)
- Privacy-preserving pinning strategies (presented by Rahul)
- Block exchange protocol planning
Roadmap Overview
The team established three tentative parallel development tracks (starting November 23, 2025) leading to three testnets in Q1, end of Q2, and Q4:
-
Track 1: Filesharing - Focus on static files and bundles for dapp frontends, DHT and block protocol specs, manual replication API with local pinning, and integration with Logos Core including API as C library and permission systems.
-
Track 2: Baseline Privacy - Aim for privacy-preserving filesharing with defined threat model and baseline (e.g., BitTorrent over Tor). Key goals include downloader/provider unlinkability, plausible deniability for providers, OFAC compliance, and censorship-resistance. Explore DHT encryption and forwarding mechanisms.
-
Track 3: Persistence - Year 1 deliverables include primitives for approximate replica count using DHT seeder counting and pinning capabilities. Future work includes remote audits and decentralized, permissionless, anonymous persistence marketplace.
Retrospective Highlights
What Went Well:
- Strong team performance and leadership support for Giuliano and Daniel K.
- Improved collaboration with IFT, Logos Communication (Waku), and Status
- Successful delivery: Status community history archive, block exchange stability, C-bindings, Go integration
- Team resilience during leadership transition and exciting privacy research progress
Needs Improvement:
- Communication within team and across Logos stack teams
- Better definition between research and engineering efforts
- More transparency from IFT and clarity on decision-making boundaries
- Improved specs structure, documentation processes, testing, and PR workflows
- Better resource utilization and delegation
- Nim identified as productivity bottleneck
Privacy Track Next Steps
- Identify current privacy leaks in Codex/Logos Storage
- Target Tribler-level privacy as first milestone, then address its limitations
- Consolidate Tor and Tribler research into public post
- Deepen understanding through discussions with Tribler team
- Explore collaboration with anonymous communication team on libp2p-mix
Action Items
Decide by December 5th: Rust vs Nim transition✅ (See below)- Develop block exchange spec outlining components and requirements
- Analyze BitTorrent privacy failures and improvement opportunities
- Schedule DST session during IFT research call on systems testing
- Define MUST deliverables for Logos Core integration
- Establish decision-making boundaries between Daniel and Giuliano
- IFT to provide spec framework and coordination process
- Improve proactive collaboration with Jacek
Future Directions (post-mainnet): cryptoeconomically secure and durable storage, mutations, off-chain compute, private information retrieval, decentralized database, anonymous storage.
Rust vs Nim Trial
This week, the Logos Storage team underwent a ~1-week “trial” of Rust before committing to a decision. On Thu we met and decided to give Rust a spin; i.e., we’ll try using it for the project. This trial did not provide enough information to make a “safe” decision, therefore this will be a pilot regardless of whether we want to call it that way or not. There may be constraints found that were not understood at the time of the decision; e.g. if an integration hell occurs, and reversion is needed. However, the overall team sentiment is that the Rust “trial” will help and not harm, for all the reasons given before. Risks will be hedged by having the Nim implementation ready as a backup, so that deadlines can be met regardless; integration work with logos-core will start ASAP.