Say I wanted to overload the modulus operator to return the remainder of a division between two floating point numbers. Why isn't a custom doubleoperator%(double, double) allowed even though that function isn't available in the standard anyway?
Edited my wording after realizing how stupid it sounded.