Fwd: FOURTH ... oops, it's an 1130.

[Bob Flanders]

So I don't know if this ever got posted on the 1130 group. As you can see, in 2011, Chuck Moore (the author of FORTH) responded to me and sent the attached images.

I had lost track of it for a long time until Carl Claunch asked me if I still had it. One email search later turned this up.

Carl has since gotten it mostly transcribed and partially working on the 1130, but it's not quite ready for primetime. 

Last night (May 16, 2020), I asked Mr. Moore if it would be ok to post this to the forum and maybe elsewhere, and here is his response:

On Sat, May 16, 2020 at 8:14 AM Charles Moore <c...@i....com> wrote:

Ok to post 1130 Forth code. Pleased you’re still interested.

So here it is. The original version of FORTH written for the IBM 1130.

Regards,

Bob

---------- Forwarded message ---------

From: Chuck Moore
Date: Sat, Sep 10, 2011 at 1:57 PM
Subject: RE: FOURTH ... oops, it's an 1130.
To: Bob Flanders <bob.fl...@gmail.com>

Here are 16 pages of 1130 code.

After page 1, 4 different pages have been interpolated. I don't know what
they are. Then pages 2-12.

Some pages didn't fit well in my scanner. Some text to the right might be
missing. If that's a problem, I can rescan.

If you can make sense of this, I can send you some Forth source.

[Danyel Lawson]

I thought the whole point of forth was that you don't need the source code for a specific platform as you only need to implement a few forth instructions like ADD to get the forth compiler to compile for that platform..

[Bob Flanders]

Hi Danyel,

That's true with more modern Forth systems. Just implement relatively few primitive at the low level and the rest should just work. But this is not modern, this is the first Forth system ever. Even then, it was a relatively small amount of pure assembler then the rest was (proto-) Forth.

Bob

[John Pierce]

On Tue, May 19, 2020 at 2:49 PM Danyel Lawson <danyel...@gmail.com> wrote:

I thought the whole point of forth was that you don't need the source code for a specific platform as you only need to implement a few forth instructions like ADD to get the forth compiler to compile for that platform..

FORTH uses a threaded interpreter.    the 4-5 pages of assembler code referenced on that post consist of the entirety of that interpreter, plus the core intrinsic functions..

--

-john r pierce

  recycling used bits in santa cruz

[Charles Anthony]

On Tuesday, May 19, 2020 at 2:49:04 PM UTC-7, Danyel Lawson wrote:

I thought the whole point of forth was that you don't need the source code for a specific platform as you only need to implement a few forth instructions like ADD to get the forth compiler to compile for that platform..

Mostly true, but the simple design allows great flexibility in implementation. On some given architecture, an implentor make choices that may require major rewrites, but keep that FORTHish nature. Eg. indirect threaded vs. direct; on machines with words larger then the addresses, embedding operands in the word, and others. FIG-Forth was a common standard and had many high-fidelity implementations on 8 bit word size machines, and good implementations on 16 bit word machines. Other word sizes gave opportunity for experimentation.  I have been toying around with a 36 bit word implemetnation and am working towards a unique FORTH that will still be recognizible as FORTH, but might require handstands to port code to.

-- Charles