I have yet to see things like key revocation protocols on the blockchain. Mistakes will inevitably be made somewhere along the line. You can bolster the fragili

I have yet to see things like key revocation protocols on the blockchain. Mistakes will inevitably be made somewhere along the line. You can bolster the fragility of being unable to cope with these mistakes with external processes, but that doesn’t change the underlying problem. The most sensible thing is usually to trust - and pay - a third party to manage the data. The net cost of a blockchain secured by PoW is quite high for a hopefully trustless consensus model. An immutable ledger has some properties that make it of questionable value to real world applications. The author calls out a number of scenarios that make it impractical for tasks that people keep trying to apply it to.