20:48:57KrystofI implemented proper method tracing though
20:49:22Krystofamazing work to get it as far as testing though!
21:12:41stassatsthe double-float bug is because i reduced the precision of the rem deriver for floats
21:12:41minionstassats, memo from scymtym: this is what i came up with: https://github.com/scymtym/sbcl/commit/9f481c9ec251e4b7c981d34addd93dcde01f1888
21:13:03stassatsso it's not matching the rational portion
21:13:49stassatsi wonder if i can reduce the precision less conservatively or inexact REAL types can't just be mixed together at all