Information Flow Control
summary
Argues that privacy is not a property protocols can grant but a user decision about emissions, and that useful solutions treat privacy as constrained information flow; cites DARPA 1977 origins and papers like RIFLE and Viaduct. Implements this via intents and a resource machine that splits state into atomic resources, uses zero-knowledge proofs to verify resource logic, and routes matching through solvers so policies (e.g., reveal only to Solver B) prune execution paths rather than magically “hide” data.
AI-written summary · about
chapters (8)
- 0:00Privacy and information flow
- 3:13Privacy in digital systems
- 9:08Privacy misunderstandings in tech
- 12:40Information flow control origins
- 15:48Information flow control in blockchains
- 18:14Information flow control policies
- 19:59Why the word privacy may be misleading
- 21:24Q&A
related talks
01h 14mPerspective: Achieving Meaningful Privacy in Technology and Design
44mThe Potential of Information-Flow Control Research for helping GDPR Complia...
44mHybrid Legal-Technical Concepts of Privacy
16mIs it Possible to Share and Protect Sensitive Information?
49mThe Praxeology of Privacy
26mAn inflection point for privacy rights
37mDelivering High Speed with Low Risk
19mA Lawyer’s Take on Privacy - Programmable Disclosure