Post #3430210
2026-06-27 15:05 UTC
@slava@mathstodon.xyz @joe@f.duriansoftware.com @SRAZKVT@tech.lgbt @pinskia@hachyderm.io You also want no unspecified padding so that (in most cases) equality, hashing, etc. are bytewise. I tried to think of a good way to have a trait system that tracks this at the type level (and composes it in the right fashion), but everything I came up with seemed very arbitrary.
Replies (1)
-
@joe@f.duriansoftware.com 2026-06-27 15:39
@zwarich@hachyderm.io @slava@mathstodon.xyz @SRAZKVT@tech.lgbt @pinskia@hachyderm.io i wonder how far you could get just saying that all data types must behave that way. seems like a natural extension of the ZII ideal