nostr:nprofile1qy2hwumn8ghj7un9d3shjtnyd968gmewwp6kyqpqncxka2nmkqkndk4wkuf3tz3l39z9m8xax3aen3h8tvudwgjmf5mq4uv2v2 nostr:nprofile1qy2hwumn8ghj7un9d3shjtnyd968gmewwp6kyqpqgqn4mglfftd6292303jp2r060qwy69m...
@nprofile1q... @nprofile1q...
C++ has the kind of static type system that I like, where I can enforce rich properties such as correct state machine transitions. It lacks viewpoint adaptation, algebraic types, and structural types, all of which make it easier to take these things further. The type systems in your papers can't express any of the properties that rich type systems have allowed me to catch bugs. I agree that static type systems do not help if they can express only the kinds of simple properties that people rarely get wrong, but type systems have come a long way in the last decade or so.
C++ has the kind of static type system that I like, where I can enforce rich properties such as correct state machine transitions. It lacks viewpoint adaptation, algebraic types, and structural types, all of which make it easier to take these things further. The type systems in your papers can't express any of the properties that rich type systems have allowed me to catch bugs. I agree that static type systems do not help if they can express only the kinds of simple properties that people rarely get wrong, but type systems have come a long way in the last decade or so.
1