The Potential of Information-Flow Control Research for helping GDPR Complia...
summary
Presents a concrete engineering route: Haskell libraries that implement information‑flow control using labeled values and a family of monads (the Mac L monad) to confine secrets, rely on Safe Haskell to restore type guarantees, and expose label-aware APIs for reading, writing and controlled declassification. Maps GDPR requirements to flow policies and plans a 4–5 year project that combines per-user labels, keyed storage and differential privacy for disclosure. Demonstrated work includes a paper and library with ~2,700 downloads last year, a company started three years ago with no customer yet, and side‑channels (timing) remaining unresolved.
AI-written summary · about
chapters (9)
- 0:00Information-flow control and GDPR
- 1:51Confinement challenges
- 8:37Information-flow control fundamentals
- 14:56Password manager dilemma
- 27:01Research impact
- 32:22Career opportunities
- 35:31Embracing failure
- 37:38IFC future for GDPR
- 38:21Q&A on timing attacks
related talks
26mInformation Flow Control
44mHybrid Legal-Technical Concepts of Privacy
21mBuilding Privacy Aware Infrastructure
01h 16mLaw and CS
37mDelivering High Speed with Low Risk
01h 01mFrom Theory to Practice: Crafting Differential Privacy Systems with Haskell
01h 15mAcademic Session I The GDPR within a continuously reforming digital environment
01h 14mSenior Academic Session I