Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!udel!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: 1990 FORML
Message-ID: <2074.UUL1.3#5129@willett.pgh.pa.us>
Date: 9 Dec 90 23:33:19 GMT
Organization: String, Scotch tape, and Paperclips.  (in Pgh, PA)
Lines: 78

Category 5,  Topic 16
Message 20        Sun Dec 09, 1990
F.SERGEANT [Frank]           at 14:52 CST
 
 BR>Sorry I missed FORML this year.  Do you plan to attend Rochester?
 .
 Thanks, Brad.  It's good to hear a friendly voice now and then.  Just  when I
was almost reconciled to my fate it appears that you don't  completely
disagree with me and that even Mitch doesn't completely  disagree with me, at
least regarding meta-compilation vs assembly.  Now  there is nothing at all I
can count on.
 .
 No, I do not plan to attend Rochester.  On the other hand I do not 
definitely plan not to attend either.  Attending seems to conflict with  two
of my goals: never to leave home again and not to spend any money.   On the
other hand, I so enjoyed FORML that I'd love to attend both  FORML and
Rochester every year.  I just have not worked this out yet.
 .
 My 4 points were meant as quick samples of my disagreements. Maybe I  can't
explain my position other than as a vague feeling of unease.  Maybe I should
call them "dislikes" instead of "disagreements."  Take  for example my daring
to say I disagreed with a senior Forth Inc person  preferring a language other
than Forth for his child. Ray Duncan in  particular seemed to disagree with me
over this, but I think he missed  my point.  Now, really, I don't give a damn
what language he teaches  his child.  I never intended to say that no decent
human being in the  world could consider teaching his child BASIC or Trilogy
or COBOL. But,  I do think it is a sign (Give us a sign, o lord, give us a
sign.  Houston: 156 miles.), no matter how unimportant in itself; and not a 
good sign, either. Take a rather extreme example. Even though I can  think of
many better things to teach a child, I would still think it  disharmonious to
hear the local Roman Catholic priest had chosen a  religion other than
Christianity to teach his own children.  I hope  people can get my point here
without taking too much offense.  If the  local doctor prescribed Tylenol to
all his patients but was careful to  see his own family never used it,
wouldn't you think something was  wrong somewhere?  Yes, yes, my analogies are
not exactly applicable.  I  hope you see what I mean about this feeling of
unease. It has nothing  to do with how to introduce programming to a child!
 .
 You have a point about the difficulty of following (some)  meta-compilers. 
(Brad, You can see mine whever you wish by either  downloading PYGMY13.ZIP or
by E-mailing your mailing address to me).  I  think mine (which perhaps should
be called Charles Moore's from  cmFORTH) is much easier to use and understand
than F83's. Pygmy is not  made to be easy to port to other processors, but to
be fast and  convenient on a PC.  Eforth is slow, but that is the trade off
for  getting it up quickly (I agree with this trade off.  Speed it up at  your
leisure, but at least you're now in Forth.  On the other hand,  speed is
everything at times.) . The idea of eforth is to have as few  machine language
primitives as possible, within reason.  These could be  isolated fairly easily
- in any metacompiler, probably - with  instructions to the porter to put in
his own hex codes for those  primitives "we don't care where you get 'em" in
the marked places, then  type 1 LOAD.  The meta-compiler then generates a new
kernel and saves a  binary (or Intel hex or Motorola S1-S9) image to the PC's
disk.  No  matter what we do, as Mitch points out, the user has to figure a
way to  download the image to his target system.  There would be a small, 
clearly marked area for setting CONSTANTS for byte order, alignment,  etc.
 .
 I say the porter does not get to claim complete ignorance of the  target
processor's assembly language, but that he needs to learn almost  nothing of a
metacompiler and absolutely nothing about MASM, to port  eforth to a new
processor.  It only takes one porter per processor.  Thereafter anyone who
wants just to use eforth can download that  version from GEnie.  Remember,
part of the eforth plan is to require  the porter to have a PC and to send him
a disk (or file via modem).  There is no need to require he also have MASM,
everything he needs can  be included (in Forth) on the disk he's being sent
anyway.  As I've  said before, if anyone wants to see whether the source code
for Forth  is more readable in Forth or in MASM, just download eforth and
compare  them (both versions are included).
 .
 So, in conclusion, I'd just like to say there is room for many  different
viewpoints and opinions and approaches on all of these and  many other
subjects.  One of the joys of attending FORML was hearing so  many of them and
seeing how misguided or insightful others can be, and  I would be remiss in my
duties if I didn't offer that same opportunity  to others.
 .
  -- Frank
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: dwp@willett.pgh.pa.us or uunet!willett!dwp