On Saturday 15 March 2014, 00:16:18, Edward Kmett wrote: > Concretely, I'm proposing we simplify the signatures of the following > functions, which in the current API all unnecessarily take a > RealFloatconstraint that they don't use. +1, naturally