I strongly agree with him on transaction privacy, though not for the reasons he lists. Some issues, however: 1. His argument against BFT is "legal systems are

I strongly agree with him on transaction privacy, though not for the reasons he lists. Some issues, however: 1. His argument against BFT is "legal systems are efficient" (lmao). Also, the whole point of HotStuff (vs. PBFT etc...) is linear O(N) communication complexity outside of cascading proposer failures... 2. He clearly didn't investigate move bytecode-verifier, which asserts linearity. 3. Strong disagree on the untested crypto-primitives argument. curve25519-dalek is audited (x2) and written in Rust; simple and minimal; not a bloated, unauditable mess like TLS. 4. Consumer protection can be built on top via the wallet providers.