🚀 go-pugleaf

   Date: Sun, 11 Feb 90 07:47:53 -0500
   From: "Jim Miller" <jmiller@chaos.cs.brandeis.edu>

   ... Yale's T (not "true" Scheme, but a variant
   with a mode very much like Scheme), 

Sorry to go into defensive mode here -- I fully agree about T's
unreliability -- but what does it take to qualify as "true" Scheme?
The only deficiencies I know of are minor, like numbers not quite
being Revised^3 Scheme numbers.  Call-with-current-continuation and
tail recursion are both supported.

   the compiled semantics do not match the interpreted semantics, 

As far as I know, this is only in the case of situations that are
considered errors, and for some peripheral situations involving
T's extensions to Scheme.

   A minimal CommonLisp implementation already
   has a useful program development environment ...

This is not an a priori feature of the languages, but a historical
artifact of the way Common Lisp implementations have so far always
been built.

   (f) There are multiple public-domain versions of Scheme, but to my
   knowledge only one complete public-domain version of CommonLisp (Kyoto
   CommonLisp or KCL).  The price of a non-public Scheme system is far
   lower than the corresponding price for CommonLisp.

What about Spice Lisp from CMU?  I thought that was freely available
and legally unencumbered.  (I think you misunderstand the term "public
domain," which really means that no one is trying to control the way
in which the software is copied in any way.  None of the software you
mention is public domain in that sense.)

   (g) Converting code from CommonLisp to Scheme is generally very
   difficult because of the size differences between the languages.
   Converting from Scheme to CommonLisp is generally easy as long as one
   avoids the more exotic uses of call-with-current-continuation and
   macros.

If I may plug my own product: There's little need to convert manually
from Scheme to Common Lisp, because there exists an automatic
translator that does most of the job.

- Jonathan