Math.max(b, x) == at_least_b(x)
at_most_b(at_least_a(x))
A similar kind of thing is really common in measure theory (limsup, liminf) and option payoffs.