dynamic vs. static rounding

[Bob Alverson]

I reject the assumption that programmers want to mess with round modes
at all.  Couldn't the primitives (interval_add, interval_multiply, etc.)
be written in assembly (how complicated are they?) and provided as
subroutines to the programmer?  Compilers and linkers are already able
to inline such routines.  To write portable interval arithmetic, shouldn't
one have a core set of primitives that encapsulate the basic operations?
With C++, one could write:

	#include <interval.h>
	interval_double foo, bar, flarp;
	flarp = foo*bar;

With luck, you might even get efficient code!

Bob