Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uunet!willett!dwp
From: dwp@willett.UUCP (Doug Philips)
Newsgroups: comp.lang.forth
Subject: Re: BASIS Feedback
Message-ID: <1245.UUL1.3#5129@willett.UUCP>
Date: 30 Jun 90 04:25:20 GMT
References: <1238.UUL1.3#5129@willett.UUCP>
Organization: Latest link in the ForthNet chain.  (Pgh, PA)
Lines: 52

In <1238.UUL1.3#5129@willett.UUCP>, F.SERGEANT [Frank] writes:

>   I've been following Doug's & Wil's exchanges on this subject.  1st,  with
> regard to the above quote, I don't think the same people are  making both
> types of statements.  "Forth is good for everything but at  least it is good
> for embedded systems."  "I didn't do it.  Besides, it  wasn't my fault."

It may be that they are different people, but I'm not convinced that that
should matter.

>                                 The parallel to the airport  system on the Mac
> is probably not the language a person uses who runs  the Mac, but the language
> used in ROMs that tie the various hardware  parts of the Mac together.

I'm not quite sure how this fits in to things.  The top-level interpreter
is just one form of interaction between the machine and the user.  It isn't
necessarily the most superiour or desirable form.  It may be the issue of
task-at-hand rather than interface though.  I'm not sure.

>      I am not convinced that Forth is not or cannot be superior as a  general
> development language on a large 'personal' computer system.

Personally, I hope that it *can* be a general developement language.  I
have built up my expectations about Forth from my reading and from my
conversations on ForthNet.  My expectations are that one may have to
start with some low-level words, but the application is finally written
having "layers" of code(words), each one more removed from low-level words,
growing towards a solution.  I don't mind having to do some low-level
stuff.  I mind that may be all Forth can do.  (Or so its seems to me that
Wil is saying, part of the time).

>      (Is this really a BASIS feedback topic? - You bet!)  If Forth  really is
> a little bitty language do we need a great big standard for  it?

I'm not sure that argument applies.  There are standards for the
strangest tiniest stuff (my memory went blank, but I'm sure someone can
recall one).  A standard can serve many purposes.  One purpose may be to
manage and elucidate complexity.  Another, and I think at least as
important, purpose is to define the lexicon so that portability can take
place.  Even assuming for a moment that Forth were too little to need a
standard for the first reason, it may well need one for the second.  It
really doesn't matter to me much all the flap about divide.  I care that
the behaviour is specified and named, but really don't much care what
the name is so long as it isn't actively deceptive ( * >IN + FOR ).  Of
course, I don't have any code to worry about backwards-compatibility
with either.

-Doug

---
Preferred: willett!dwp@hobbes.cert.sei.cmu.edu OR ...!sei!willett!dwp
Daily: ...!{uunet,nfsun}!willett!dwp   [in a pinch: dwp@vega.fac.cs.cmu.edu]