Security issues in Internet communication tend not to be subtle mathematical flaws in the cryptography, but instead, broader system issues. For example, humans using the Internet. We have lovely cryptography. We have certificates. We have great protocols for doing authentication. But does that really assure a human that they are talking to what they think they are talking to? What about authenticating people? What kinds of names should people have, so that the name is unique, and someone that wants to talk to a human will know what name to use? What about distributed systems that are provably correct, provided that all the components are doing what they are supposed to be doing, but do not work correctly if some components misbehave? How can we design systems that will be robust despite misbehaving participants? Will digital signatures on data assure us that data that we read on the Internet is true? Is the simple answer to everything that we should blame users if things go wrong, and just complain that users need more training? (hint…no) Or maybe using blockchain everywhere will make everything secure? (hint…no)