Issue: FUNCTION-COERCE-TIME (Version 2)

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

re:  My notes from Fairfax meeting...
     . . . 
     JonL had some critical comments, but KMP couldn't understand them
     and believes it was because JonL didn't read the proposals carefully
     and was commenting on a superficial reading that led him to incorrect 
     conclusions.

This note doesn't say anyting at all to me (presumably it meant something
to you at the time?).

I favor FUNCTION-COERCE-TIME:HYBRID.

The proposal is ovely opaque because it is trying to be overly general.
Given the "cleanup" that we've already accepted for the FUNCTION type,
then the only non-"functions" that can be passed to the functions of
reference (MAP, SORT, SET-MACRO-CHARACTER, etc) are SYMBOLS.  Also,
it should be noted that the "coercion" involved is merely SYMBOL-FUNCTION.


-- JonL --