issue FUNCTION-COMPOSITION

[sandra%defun@cs.utah.edu (Sandra J Loosemore)]

I tend to agree with the general sentiment expressed at the meeting that
this is an interesting concept but it's not appropriate for standardization
at this time, except for maybe the functions COMPLEMENT and ALWAYS.  I'd
vote for it if it only specified those two functions, against it otherwise.

If there is a serious effort made to push this idea in to the standard now,
I think it would also be appropriate to define MULTIPLE-VALUE-COMPOSE:

(funcall (multiple-value-compose #'f #'g #'h) a b c)

is the same as

(multiple-value-call #'f
    (multiple-value-call #'g
        (h a b c)))

-Sandra
-------