Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rice!uw-beaver!cornell!rochester!pt.cs.cmu.edu!dsl.pitt.edu!pitt!willett!ForthNet
From: ForthNet@willett.pgh.pa.us (ForthNet articles from GEnie)
Newsgroups: comp.lang.forth
Subject: Floating Point
Message-ID: <1656.UUL1.3#5129@willett.pgh.pa.us>
Date: 3 Sep 90 04:08:10 GMT
Organization: String, Scotch tape, and Paperclips.  (in Pgh, PA)
Lines: 35

Category 10,  Topic 8
Message 40        Fri Aug 31, 1990
D.RUFFER [Dennis]            at 13:33 EDT
 
From Elizabeth Rather, Chair, ANS X3J14 Technical Committee:

This is addressed to those who have been commenting extensively on our work on
UseNet and related boards.

5. Finally, regarding the floating stack, the rule for writing "don't care"
code is to nest "like" data types such that you aren't shuffling.  That is,
say

    addr1 F@ addr2 F@ F+ addr3 F!

(which works fine in both versions) if you like, but not

    addr3 addr1 addr2 F@ F@ F+ F!

(or addr3 addr2 addr1 F@ ROT F@ because how do you know it's a ROLL and not n
PICK?)

If you avoid situations that involve mixed stack thrashing (which takes some
care) you may find you have more readable code.  Lee Brotzman's admittedly
sloppy practice of ignoring argument orders contributes flagrantly to the
notion that Forth is "write-only code" in my opinion.

Of course this doesn't help existing programs.  The simplest solution to avoid
rewriting is to declare an environmental dependency.  This isn't a "scarlet
letter", you know.  If you "restrict" your programs to running on 80x87 and
6888x coprocessors you still have the overwhelming bulk of the market.  The
biggest selling software packages of all time run only on DOS PC's.
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: uunet!willett!dwp or dwp@willett.pgh.pa.us