"We don't really have 100 billion dollars of security, we really have 5 million a day" EIP-8363 critic Oisin Kyne on the censorship risk he says the proposal underrates "The main reason I'm not happy with the current proposed curve is this idea of it going to zero. There's more than one metric that matters for Ethereum security. The one put up front is finality, we have this amount of ETH that will be guaranteed burned if a transaction ever becomes undone, and I agree the research says tens of billions of dollars getting destroyed is probably enough disincentive" "Where we diverge is that's not the only thing that can go wrong. We also have this risk where the people staking their ETH decide to be particular about what they include in a block, saying we're only going to attest to blocks that do what we want, that don't have people we don't like in them. If they do that there's no penalty, so long as there's 51% of them or more doing it, it costs them nothing" "The plan we have is, oh no, maybe we'll coordinate as a group and destroy all their ether for doing so, but I think that's a very big risk to hedge all of Ethereum security on. I don't think we should be sacrificing censorship resistance for the staking ratio" "The fancy word is the Nakamoto coefficient, which comes from Bitcoin, this idea of how few orgs make up 50% of hash power or stake weight. We don't really have 100 billion dollars of security, we really have 5 million a day at the current curve, and depending on where the equilibrium lands it might be a million dollars a day or less. I don't know if that's enough to protect a trillion dollar computer"