COMMENT ⊗   VALID 00003 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	 MacLisp to InterLisp
C00003 00003	 Notes on translation from MacLisp to InterLisp
C00005 ENDMK
C⊗;
;;; MacLisp to InterLisp
(LOAD "UTIL.FAS[AID,LSP]")
(LOAD "INTER.3[LSP,RPG]")

(load "inter.fas[lsp,rpg]")

;(setq 1-based-arrayp t)
(macex puzzl1 lsp)
(closeall)

(do ((units units (cdr units))
     (baz 99))
    ((null units) ())
    (foo (car units)))


;;; Notes on translation from MacLisp to InterLisp
∂04-Oct-82  0155	Masinter at PARC-MAXC 	translation from maclisp   
Date:  4-Oct-82  1:50:11 PDT (Monday)
From: Masinter at PARC-MAXC
Subject: translation from maclisp
To: RPG@SU-AI
cc: Masinter.PA

The proper translation of "ATOM" from Maclisp to Interlisp is "NLISTP"
ATOM is slower.

When looking at the Boyer benchmark, we discovered that one problem
was that Interlisp PUTPROP puts properties at end, so that if you 
are relying on GETPROP/PUTPROP that it depends on what the atoms are.
This made a difference for PLUS, TIMES, etc which already had system
property lists.


BROWSE uses REMAINDER rather than IREMAINDER. The PROGS and GOOS
are awful.

MacLisp ASSOC -> FASSOC. 

FRPOLY should make the simple functions macros, no? Aren't they
macros in the MacLisp version?