Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!pacbell.com!att!cbnewse!cwpjr
From: cwpjr@cbnewse.cb.att.com (clyde.w.jr.phillips)
Newsgroups: comp.lang.forth
Subject: Re: Writing a stack based language
Summary: Dangerous
Message-ID: <1991Jun26.191449.19366@cbnewse.cb.att.com>
Date: 26 Jun 91 19:14:49 GMT
References: <9106251247.AA07801@ucbvax.Berkeley.EDU>
Distribution: na
Organization: AT&T Bell Laboratories
Lines: 41

In article <9106251247.AA07801@ucbvax.Berkeley.EDU>, Mitch.Bradley@ENG.SUN.COM writes:
> > I am just starting a project for which a small in-house language seems like
> > it might be a solution. Because I need to get this running rapidly, and
> > because the syntax/semantics question is not of overwhelming importance a
> > Forth-like language (eg stack based) might be suitable.
> 
> Why reinvent the wheel?  Why not just buy an existing Forth implementation?
> That way, you won't have to write all the docmentation from scratch, and
> you have some chance of buying trained programmers later to maintain it,
> if that becomes necessary.  I can't imagine it costing less to develop
> something from scratch that it will cost to get a full license for an
> existing Forth implementation (a very few Forth systems are very expensive,
> but most Forth systems are sold for far less than what it cost to develop
> them, and some are given away outright.).
> 
> Mitch.Bradley@Eng.Sun.COM

My worst job was in a sweat-shop run by a cheapskate who had finaaly
lost his patience with his brilliant ( read get the job done eventually,
but Its a good job cause I got him for half what he wanted... )
programmer and he hired me.

Turns out this fellow had gotten a glimpse of FORTH and decided
it would be great for this project. This guy actually implemented
every technique he "learned" as he was learning them, in each stage
of the "enhancement" to this project....

Anyway he implemented a global variable interrupt handler,
then a C.S. grad implemented a OSI protocol ion top of that,
then this guy gets a whiff of FORTH and inplements a user interface
macro interpreter ( his essense of FORTH ), but where you had to
write extensions in assembly and re asemble/link to test each new
word(s). Then he implements this assembly library deal,
then he enhanced the interpreter to scam on the global variables
so the user io actually represented the current state of the system,
onn and on ad nasseum.

Watch out! FORTH is a "small in house language" par excellence.
As Carl Malden used to say "Don't settle for anything else"

Clyde