Issue: FUNCTION-COMPOSITION (Version 2)

[Jon L White <jonl@lucid.com>]

re: My notes from Fairfax meeting...
    . . . 
     JonL: Gratuitous.
       Also, no existing implementations have this.
       [He didn't seem willing to count the T language. -kmp]

[I don't remember saying this, but if I did, it certainly would be
defensible on the grounds that we are standardizing Common Lisp, not T].

As I've outlined earlier, this is the sort of gratuitious addition to the 
language that ought to be tested first -- tested  by it's utililty to some 
vendor/implementor who feels it's worth the risk to add something like it
to his product.  I deplore the tendency to think that vendors shouldn't make
an offering unless it is "sanctioned" by the X3J13 committee.

I say "gratuitious" because
  (1) no vendor/implementor supplies them now; thus it is not "existing 
      practice" that needs to be standardized;
  (2) no fundamental problem has been exposed because of its lack; no
      implementational headaches would be resolved, and few (if any) pleas
      from the user community would be addressed;
  (3) no confusions exists among our community as to what these functionals
      (or similar such features) mean; hence no need to clarify.


-- JonL --