Were fxquotient and fxremainder left out intentionally?

Is it worth supporting hypothetical non-twos-complement implementations?

This would basically mean the range becomes [fx-least, fx-greatest] instead

of [-2^(w-1), 2^(w-1)-1], and that fx-width would be the minimum number of

bits required to represent this range.

What's the rationale for having some procedures (inequalities and min/max)

n-ary, but others (arithmetic and bitwise operators) strictly binary?

-- 

Alex