3 comments

  • agentultra 4 hours ago ago

    Nice paper. I’ve long been curious whether we could have an FP language that didn’t need a run-time to support it.

    Seems like this is a step in that direction.

    It’s neat what falls out of type theory. Counting Immutable Beans was another eye-opening paper worth reading if you’re curious like me.

  • noelwelsh 4 hours ago ago

    Interesting stuff. I didn't read the whole paper in detail. I'm unclear on how FIP differs from adding uniqueness types to a language. A number of languages are exploring this (Rust being the most well known) so what's the novelty that FIP brings?

    • JoelMcCracken 4 hours ago ago

      IIUC koka can apply these optimizations dynamically when it determines data is unique via reference counting. However the paper has more details.