Suppose my Scheme gives me IEEE double and IEEE extended precision
types, the former for normal work and the latter for the few places that
I need extra precision, or IEEE single and double precesion, the former
for compact storage.

What does:

   (fl+ x y)

actually mean?

Type declarations can cover this situation without any problem. They can
also cover the situation in which one operand is of one type and the
other of another type.

Basically, the proposed SRFI identifies a problem, but then proposes
only a partial solution.

Regards,

Alan
--
Dr Alan Watson
Centro de Radioastronomía y Astrofísica
Universidad Astronómico Nacional de México