Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!caen!ox.com!math.fu-berlin.de!unidui!unido!majestix!uh
From: uh@informatik.uni-kiel.dbp.de (Ulrich Hoffmann)
Newsgroups: comp.lang.forth
Subject: Re: ANS FORTH TECHNICAL COMMITTEE
Message-ID: <3283@batman.informatik.uni-kiel.dbp.de>
Date: 17 Apr 91 11:09:33 GMT
References: <2636.UUL1.3#5129@willett.pgh.pa.us>
Organization: Universitaet Kiel, Germany
Lines: 45

In comp.lang.forth Brad Rodriguez writes:

>> : DEFER  CREATE ['] COMPLAIN ,  DOES> @ EXECUTE ;

>Indeed, can the parameter field of the DEFERred word be changed at all?  My
>first thought was something like,

>   : IS   ['] >BODY ! ;
>   [']  word-name  IS  deferred-word

Brad, you surely mean
	
   : IS ( <word> cfa -- )  ' >BODY ! ;

and

   ' word-name  IS  deferred-word

, right?

As I understand, ansFORTH will allow to address the parameter fields of
words defined by user defined defining words. 
Thus your implementation of DEFER would be perfectly legal.
What about also defining 

   : [IS] ( comp: <word> -- ) ( run: cfa -- )  
          ' >BODY   POSTPONE LITERAL   POSTPONE ! ; IMMEDIATE

or a state smart IS?

Personaly I would like DEFER to be included in the CORE EXTENSION WORDSET.
I would expect F83 functionality if I ever see DEFER.  (Does anyone use it in
a different way?)
If DEFER will be in the CORE EXTENSION WORDSET you can take it or leave it, but
if you take it you have to define it with a standard funtionality. 
This would perfectly ok by me.


Ulrich Hoffmann
----------------------------------------------------------------------
Christian-Albrechts-Universitaet Kiel, Institut fuer Informatik
Preusserstr. 1 - 9                   , D - 2300 Kiel 1
Telefon: ++49-431-5604-59            , Telefax: ++49-431-566143
EMail: uh@majestix.informatik.uni-kiel.dbp.de
----------------------------------------------------------------------