Path: ...!news.mixmin.net!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: p.dean@invalid.net (Peter Dean)
Newsgroups: comp.os.linux.misc,alt.folklore.computers
Subject: Re: The joy of FORTH
Date: Sun, 20 Oct 2024 06:22:04 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 70
Sender:  <peter@flo.local>
Message-ID: <vf27ib$9pgg$1@dont-email.me>
References: <pan$96411$d204da43$cc34bb91$1fe98651@linux.rocks>   <vdnvk0$49ai$2@dont-email.me> <o8icnRRnt4ChGGL7nZ2dnZfqn_idnZ2d@earthlink.com> <vdo3j6$4q92$1@dont-email.me> <lm9kg6Ft2vgU2@mid.individual.net> <vdpi96$bgk6$7@dont-email.me> <lmb9k2F64inU8@mid.individual.net> <vdqlcs$lo51$1@dont-email.me> <TKqcnfnts-lvnJ_6nZ2dnZfqn_WdnZ2d@earthlink.com> <pRDMO.80813$7OO5.26774@fx43.iad> <lmh1avF2k8fU1@mid.individual.net> <MZScnWkZT7T_Hon6nZ2dnZfqnPWdnZ2d@earthlink.com>
Injection-Date: Sun, 20 Oct 2024 08:22:04 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="04139ebc9e1c02cf09201b2f9d6d9f07";
	logging-data="321040"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+eKFbCf/jxPji0W3Ka3GSd"
User-Agent: tin/2.6.3-20231224 ("Banff") (Linux/6.6.57-1-lts (x86_64))
Cancel-Lock: sha1:ljVcSO6kCd0PYS12xMB6LmfodpU=
Bytes: 4122

In alt.folklore.computers 186282@ud0s4.net <186283@ud0s4.net> wrote:
> On 10/6/24 11:02 PM, rbowman wrote:
>> On Sun, 6 Oct 2024 16:06:14 -0600, Louis Krupp wrote:
>> 
>>> GNU Forth... I don't know enough about it to have an opinion.
>> 
>> https://gforth.org/manual/Forth-is-written-in-Forth.html
>> 
>> Forth is an odd beast. There is a very small engine that has to be built
>> for the processor in question to handle words.
>> 
>> https://github.com/forthy42/gforth/tree/master/engine
>> 
>> gcc can handle that since it is C code.
> 
> 
>   The original Chuck Moore FORTH slightly pre-dates the
>   i4004 chip.
> 
>   As a half-a-step above ASM the syntax is (sort of) more
>   readable. The 'stack' approach to dealing with both
>   commands and data could be implemented very simply
>   (but you HAD to be fully aware of what was on the
>   stack exactly where). FORTH also supports subroutines
>   and you can make a lib of those.
> 
>   The interpreter can be VERY small - fit into a little
>   old ROM chip. I think one company made CPUs with an
>   inbuilt FORTH kernel. It is reported that the first
>   language ported to the new 8088/8086 processors way
>   back was FORTH.
> 
>   It was especially popular in the 70s/80s for minimal
>   systems - especially for academia/space. If you had a
>   telescope on a mountaintop in Chile and a 110 baud
>   connection then you could still easily edit/test
>   the control program from sunny Cal. I know an old
>   astronomer - he is still fluent in FORTH.
> 
>   Modern chips/systems and 4/5-G or sat connections have
>   kinda made FORTH redundant - but it may still have a
>   place for some 'industrial' and remote-sense apps
>   using power around that of an Arduino or less.
> 
>   Interpreters can be handy sometimes.
> 
>   Somewhere I've got a 'CPL' interpreter. CPL
>   became BCPL which became 'B' which became 'C'.
> 
>   gFORTH is kind of a 'cheat' - just a translator
>   into 'C'. What you want is a native interpreter.
> 
>   https://www.forth.org/compilers.html
> 
>   SOMEWHERE I came across a whole FORTH dev environment
>   for Linux but accidentally deleted it.

This reminded me of a time in the early 80s when I had figforth running on
apple ][+.  I needed help and found out there was a forth interest group in my
city.  I went along to a meeting and found they were all uni students (way
above me).  But they weren't using forth.  They preferred a similar language
called stoic that ran on 8080 cpm computers.  It had strings and floating
point!  It was the floating point I missed and the reason I gave up on forth.

I just spent some time googling for it and found archived docs and source on
github. FYI

https://github.com/tendai22/stoic_8080