Secure RDTs: Enforcing Access Control Policies for Offline Available JSON Data
summary
A system called Secure RDTs enforces role‑based read/write rules on offline JSON by projecting a server‑specified policy and per‑role data excerpts to clients and having the leader reject unauthorized writes at merge time. The design is formalized as operational semantics in PLT Redex, backed by mechanized proofs and extensive randomized Redex testing, and a working proof‑of‑concept as an Automerge wrapper demonstrates that leaks and contagion are prevented while preserving offline edits.
AI-written summary · about
chapters (14)
- 0:00Secure RDTs introduction
- 2:28Nature Point example
- 4:37Problems with current replicated data types
- 6:00Core assumptions and solution
- 7:24Security policy specification
- 9:52Offline policy projection
- 10:35Read privileges enforcement
- 11:18Write privileges enforcement
- 12:22Formal verification of sRDTs
- 13:26Summary of sRDT benefits
- 14:30Q&A: Group membership
- 15:13Q&A: Server necessity
- 16:17Q&A: P2P challenges
- 17:45Closing remarks
related talks
34mUsing CRDTs Beyond Text Editors
40mPractical Demystification of CRDT
01h 13mCRDTs: The hard parts
01h 30mImplementing, verifying and debugging distributed event-based systems
01h 03mOffline first computing in an online world
32mLow Latency Serving of Offline Data: Efficient, Safe, and Reliable Data...
22mXRD: Scalable Messaging System with Cryptographic Privacy
27mThe Dark Forest: Private CRDTs as Global Database