Re: [TYPES] types
Sam Tobin-Hochstadt <[email protected]>
| Newsgroups | gmane.comp.science.types |
|---|---|
| Message-ID | <CAK=HD+byLcZ22RoOy7Ui0tszQZwR8AThaJr3Wpu9_KXuOVNZ6A@mail.gmail.com> |
[ The Types Forum, http://lists.seas.upenn.edu/mailman/listinfo/types-list ] On Tue, May 13, 2014 at 7:57 AM, Jacques Carette <carette-xc3H/SjimlosA/[email protected]> wrote: > [I use 'enable' here, knowing that one can do pattern matching without sums > and without types, but the results are not as compelling.] Jacques, I'm not sure what you mean by compelling, but any pair of these three is available without the other. For example, patterns matching and sums (but without types) are used extensively in Friedman and Wand's Essentials of Programming Languages textbook, and pattern matching with types but without sums can be seen in Typed Racket (for example, our paper on functional data structures [1]). Types and sums without pattern matching is easy to define, but as you say, awkward to program with. Sam Tobin-Hochstadt [1] http://www.ccs.neu.edu/racket/pubs/sfp10-kth.pdf with source at https://github.com/takikawa/tr-pfds/