Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!agate!ucbvax!UICVMC.BITNET!UWC6NTG
From: UWC6NTG@UICVMC.BITNET (Nicholas Geovanis 312-996-0590)
Newsgroups: comp.lang.forth
Subject: FORTH, the language
Message-ID: <8907220145.AA03972@jade.berkeley.edu>
Date: 21 Jul 89 16:51:28 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: Forth Interest Group International List <FIGI-L%SCFVM.bitnet@jade.berkeley.edu>
Organization: The Internet
Lines: 39


   I can't stand the "my-language-is-better-than-yours" arguments, but I
can no longer stand by and watch. Where I'm coming from: I once wrote a
FORTH interpreter for a college class (mostly in FORTH), I've written
perhaps a few hundred other lines of FORTH, none of it for money.
   Charles Moore visited the U. of Illinois at Chicago a few years ago,
and I jumped at the chance to hear him. He was pushing the first Novix(?)
chip. In the course of his talk, he displayed some of his very own code.
Thinking that the god of FORTH would have some pretty nifty shit, I was
shocked to discover that he didn't even use line-spacing to make the
logical flow of his code more apparent: all the lines were jumbled
together end-to-end. When I asked him about this, he said he wrote code
that way because it saved space, and besides, no unit of FORTH code
should be longer than one screen anyway. Needless to say, there were no
comments embedded in the code.
   I also asked him if he used a particular file system to keep track
of his code/data. It wasn't long before beginning to use FORTH that I
found the "first-screen-is-a-directory" idea a little cumbersome, and
cast about for something more useful to humanoids. Well, of course he
doesn't use a file system, it's all in his head! He just memorizes exactly
how far into his screen file each aggregation of code has gone so far.
   Now, maybe Charles Moore really is Einstein Jr., but even if he were,
no matter how you slice it, this is sloppy programming practice. I know
that this isn't how most FORTH programmers work (I hope), but it's
indicative of some of the FORTH community's attitudes. If you pride
yourself on your intuitive abilities (and FORTH-freaks do claim that it's
more supportive of a "natural, intuitive" programming style than any other
language), then you don't want to encumber yourself with support software.
If you subscribe to the "simpler-is-better" philosophy, then you might want
to use banked memory on your FORTH chip, taking it upon yourself to select
the bank, rather than upgrading your "simple" FORTH from 8-bit dependency
to 16 or 32-bit "complexity" (Moore did not support such an upgrade at the
time).
   To sum up, in my opinion, the FORTH language does not encourage poor
programming practice, but those who champion FORTH have burdened it with
a programming philosophy which sometimes results in poor programming
practice.
                       Nick Geovanis-UWC6NTG at UICVMC.EDU
                       Univ. of Illinois Admin Computer Ctr.