Over on a hobby forum that I frequent, a couple competent people
who have written their own Forth systems but work professionally in C,
Verilog and other more popular languages are asking for examples of large
Forth projects. They know what Forth is but want to see how fluent
users deal with large projects. One quote was "I understand Forth but feel
like I still don't 'get it'"

I know that much of what they would want to see is proprietary but
are there any examples online that I could point them too?

On Monday, November 22, 2021 at 4:59:12 PM UTC+3, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?

You are asking for real applications in Forth?

- https://github.com/nnCron/nnCron
- https://github.com/openbios/openfirmware
- http://www.eserv.ru/EservSourceCode

On Monday, November 22, 2021 at 2:59:12 PM UTC+1, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?

I am working on iSPICE, a circuit simulator specialized for power electronics
and digital control. Like Forth, it will never be ready.

Unfortunately I did not publish anything on it yet. It is currently in a state
where it replaces PLECS, NGSPICE, and LTspice for me personally (it is
about 10x faster overall).

-marcel

On Monday, November 22, 2021 at 12:42:40 PM UTC-5, Marcel Hendrix wrote:
> On Monday, November 22, 2021 at 2:59:12 PM UTC+1, Brian Fox wrote:
> > Over on a hobby forum that I frequent, a couple competent people
> > who have written their own Forth systems but work professionally in C,
> > Verilog and other more popular languages are asking for examples of large
> > Forth projects. They know what Forth is but want to see how fluent
> > users deal with large projects. One quote was "I understand Forth but feel
> > like I still don't 'get it'"
> >
> > I know that much of what they would want to see is proprietary but
> > are there any examples online that I could point them too?
> I am working on iSPICE, a circuit simulator specialized for power electronics
> and digital control. Like Forth, it will never be ready.
>
> Unfortunately I did not publish anything on it yet. It is currently in a state
> where it replaces PLECS, NGSPICE, and LTspice for me personally (it is
> about 10x faster overall).
>
> -marcel

Thanks Nickolay and Marcel for these examples.

Brian Fox schrieb am Montag, 22. November 2021 um 14:59:12 UTC+1:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?

No sources, but:
https://www.forth.com/resources/space-applications/

On 11/22/21 7:59 AM, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?
>

Define "large".

I don't consider it very large but...

http://davesrocketworks.com/electronics/logger/TeensyLog.html

--
http://davesrocketworks.com
David Schultz

On Monday, November 22, 2021 at 5:59:12 AM UTC-8, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?

https://ai.neocities.org/RoboMind.html -- a major Forth project in artificial intelligence.

Mentifex
--
http://meatballwiki.org/wiki/MindForth
http://github.com/kernc/mindforth/tree/master/wiki
http://ai.neocities.org/MFPJ.html -- MindForth Programming Journal
http://dl.acm.org/citation.cfm?doid=307824.307853 = Mind.Forth
http://www.amazon.com/dp/B00GX2B8F0 -- eBook on Forth AI thinking in German

On Monday, 22 November 2021 at 05:59:12 UTC-8, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?
The following describes a fairly large system developed over several years
https://doi.org/10.24908/pceea.v0i0.4004
There are also many papers that describe the details... cnc language, process planning,
process optimisation, high speed profiling etc etc. I think the link given might
give some insight as to where FORTH can be extremely useful at the core of
a large complex system...
Ian

ian...@gmail.com schrieb am Mittwoch, 24. November 2021 um 04:13:40 UTC+1:
> On Monday, 22 November 2021 at 05:59:12 UTC-8, Brian Fox wrote:
> > Over on a hobby forum that I frequent, a couple competent people
> > who have written their own Forth systems but work professionally in C,
> > Verilog and other more popular languages are asking for examples of large
> > Forth projects. They know what Forth is but want to see how fluent
> > users deal with large projects. One quote was "I understand Forth but feel
> > like I still don't 'get it'"
> >
> > I know that much of what they would want to see is proprietary but
> > are there any examples online that I could point them too?
> The following describes a fairly large system developed over several years
> https://doi.org/10.24908/pceea.v0i0.4004
> There are also many papers that describe the details... cnc language, process planning,
> process optimisation, high speed profiling etc etc. I think the link given might
> give some insight as to where FORTH can be extremely useful at the core of
> a large complex system...

Unsurprising (at least for me) one core element (eCAF) is a Forth system (pfe) written in C
running on top of a multitasking OS.

However, GUI management is supported by Python scripts.

On Monday, November 22, 2021 at 6:59:12 AM UTC-7, Brian Fox wrote:
> Over on a hobby forum that I frequent...

What hobby is this forum dedicated to?
I mean, is this robotics, board-games, or what?

> a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects.

Most likely they want to see an example of Forth in a large project
because they don't believe that Forth has ever been used for a large project.

As a practical matter, a large project takes a lot of sweat-equity,
so the programmer wants to get paid --- meaning that it will be closed-source.

> One quote was "I understand Forth but feel
> like I still don't 'get it'"

I don't think that looking at someone else's code is a good way to learn.
Programmers have to learn programming by programming.

As a practical matter, to write a large project (or a medium-sized project)
successfully, it is necessary to have a code-library available that provides
general-purpose data-structures --- all non-trivial projects work with data,
and they need some way to store this data and work with it.

The student doesn't necessarily have to look at the source-code for the
code-library to use the code-library. Looking at this source-code isn't a good idea
anyway, because it is not representative of application programming.
The way that I wrote the internal workings of the novice-package is not the same
way that I write application programs --- my application programs have readability
as a priority --- the internal workings mostly had efficiency as a priority (and also
include a lot of kludges to work around the major flaws in ANS-Forth).

On Sunday, November 28, 2021 at 7:30:33 PM UTC-5, Hugh Aguilar wrote:
> On Monday, November 22, 2021 at 6:59:12 AM UTC-7, Brian Fox wrote:
> > Over on a hobby forum that I frequent...
>
> What hobby is this forum dedicated to?
> I mean, is this robotics, board-games, or what?

> Most likely they want to see an example of Forth in a large project
> because they don't believe that Forth has ever been used for a large project.
>
> As a practical matter, a large project takes a lot of sweat-equity,
> so the programmer wants to get paid --- meaning that it will be closed-source.
> > One quote was "I understand Forth but feel
> > like I still don't 'get it'"
> I don't think that looking at someone else's code is a good way to learn.
> Programmers have to learn programming by programming.
>
> As a practical matter, to write a large project (or a medium-sized project)
> successfully, it is necessary to have a code-library available that provides
> general-purpose data-structures --- all non-trivial projects work with data,
> and they need some way to store this data and work with it.
>
> The student doesn't necessarily have to look at the source-code for the
> code-library to use the code-library. Looking at this source-code isn't a good idea
> anyway, because it is not representative of application programming.
> The way that I wrote the internal workings of the novice-package is not the same
> way that I write application programs --- my application programs have readability
> as a priority --- the internal workings mostly had efficiency as a priority (and also
> include a lot of kludges to work around the major flaws in ANS-Forth).

It is for retro computers. Atari, Apple, Commodore, Tandy. TI etc.
I think you are right. It's hard for people to understand how you could
do big projects in Forth when they don't see the things they think are
essential. They see this thing that barely qualifies as a "language"
to most people. :-)

I agree that good libraries make the heavy lifting a little lighter.
It seems that most old hands at Forth "roll their own" or they use
the commercial system's offerings or both. ?

On Sunday, November 28, 2021 at 5:55:10 PM UTC-7, Brian Fox wrote:
> It is for retro computers. Atari, Apple, Commodore, Tandy. TI etc.

Why would anybody want to do a big project on a retro computer?
There are some serious hardware limitations on how big of a project
can be done --- choice of language isn't the major problem!

> I think you are right. It's hard for people to understand how you could
> do big projects in Forth when they don't see the things they think are
> essential. They see this thing that barely qualifies as a "language"
> to most people. :-)
>
> I agree that good libraries make the heavy lifting a little lighter.

Yes.
Nobody has the time or the patience to re-invent the wheel every time.

This is certainly true of professional work where you get paid by the week
and there is an expectation of some results --- if a program would take
one hour in another language, you can't take all day and then say:
"Well, I have a self-balancing binary tree now, which I intend to use to
hold the data --- I might be done with the program by the end of the week."

Hobbyists' time is less valuable (has zero monetary value). Even by the
low standards of hobbyists though, spending weeks dinking around with
a simple program is a waste of time.

> It seems that most old hands at Forth "roll their own" or they use
> the commercial system's offerings or both. ?

I'm not aware of any commercial system that has any offerings
in regard to general-purpose data-structures.
SwiftForth and VFX are both completely devoid of offerings.
Stephen Pelc claims that anybody can write a better string-stack
than mine, but he doesn't have one. Also, mine has COW (copy-on-write),
so it doesn't have to shuffle entire strings around, but usually only
has to move pointers --- nobody else has done this --- but the
"old hands at Forth" on comp.lang.forth totally denounce it.
Forth fails due to bad culture, not lack of technical capability.

On 23/11/2021 00:59, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?
>

FWIW here's the 1986 StarFlight source (some of it at least). It was written
in a modified PolyForth.

https://web.archive.org/web/20210313153919/http://www.oocities.org/timessquare/maze/4979/SFFiles.zip

On 29/11/2021 12:15, Hugh Aguilar wrote:
> ...
> Forth fails due to bad culture, not lack of technical capability.

The tradition of stoning prophets is probably bad form :)

On Sunday, November 28, 2021 at 8:15:31 PM UTC-5, Hugh Aguilar wrote:
> Why would anybody want to do a big project on a retro computer?

They don't want to do big projects on retro computers.

After being exposed to Forth they want to know how can anyone
make a large project with such a strange thing. They did not
believe it is possible.

On Monday, November 29, 2021 at 4:58:31 PM UTC+3, Brian Fox wrote:
> On Sunday, November 28, 2021 at 8:15:31 PM UTC-5, Hugh Aguilar wrote:
> > Why would anybody want to do a big project on a retro computer?
> They don't want to do big projects on retro computers.
>
> After being exposed to Forth they want to know how can anyone
> make a large project with such a strange thing. They did not
> believe it is possible.

There were many Forth applications between 1980-1990. Actually, C
compilers were dumb, processors were poorly suited to run C/Pascal code
and Forth was the only sane way to develop for embedded. Forth was really
big at that time.

But I doubt, that "large project" (in modern terms, like Linux kernel for
example) in Forth ever existed.

Nickolay Kolchin <nbkolchin@gmail.com> writes:
>But I doubt, that "large project" (in modern terms, like Linux kernel for
>example) in Forth ever existed.

There is one large project, the CCS software, which had 1.2M lines of
code last time the number was reported here. Linux had 27M lines of
code in 2020, but I think that 1.2M lines is plenty large.

For comparison, a recent Gforth development version contains 108,650
lines in .fs files and 3,309 lines in .4th files. Looking at only
files included in the AMD64 image, it's 19,867 lines of .fs files (no
..4th files). The rest are tools, test files, files for other
architectures, Minos2, etc.

- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2021: https://euro.theforth.net/2021

On Monday, November 29, 2021 at 6:28:12 PM UTC+3, Anton Ertl wrote:
> Nickolay Kolchin <nbko...@gmail.com> writes:
> >But I doubt, that "large project" (in modern terms, like Linux kernel for
> >example) in Forth ever existed.
> There is one large project, the CCS software, which had 1.2M lines of
> code last time the number was reported here. Linux had 27M lines of
> code in 2020, but I think that 1.2M lines is plenty large.

I'm curious, were they always developing in Forth since 1978? But this is
"large" (>1M sloc) application. I've actually downloaded trial to check if
it is still in Forth. It is. :)

Probably their lead developer is a huge Forth fan...

On 30/11/2021 00:58, Brian Fox wrote:
> On Sunday, November 28, 2021 at 8:15:31 PM UTC-5, Hugh Aguilar wrote:
>> Why would anybody want to do a big project on a retro computer?
>
> They don't want to do big projects on retro computers.
>
> After being exposed to Forth they want to know how can anyone
> make a large project with such a strange thing. They did not
> believe it is possible.
>

MPE regularly quotes CCS' forth application as having a million lines
of code. I gather the latter has been around since the 1980's.
I imagine writing a large application is like running a large business
i.e. good management. Not that I could do either.

On 30/11/2021 03:29, Nickolay Kolchin wrote:
> On Monday, November 29, 2021 at 6:28:12 PM UTC+3, Anton Ertl wrote:
>> Nickolay Kolchin <nbko...@gmail.com> writes:
>> >But I doubt, that "large project" (in modern terms, like Linux kernel for
>> >example) in Forth ever existed.
>> There is one large project, the CCS software, which had 1.2M lines of
>> code last time the number was reported here. Linux had 27M lines of
>> code in 2020, but I think that 1.2M lines is plenty large.
>
> I'm curious, were they always developing in Forth since 1978? But this is
> "large" (>1M sloc) application. I've actually downloaded trial to check if
> it is still in Forth. It is. :)
>
> Probably their lead developer is a huge Forth fan...

https://forth-standard.org/standard/members-2x

On Tuesday, November 30, 2021 at 12:56:26 PM UTC+3, dxforth wrote:
> On 30/11/2021 03:29, Nickolay Kolchin wrote:
> > On Monday, November 29, 2021 at 6:28:12 PM UTC+3, Anton Ertl wrote:
> >> Nickolay Kolchin <nbko...@gmail.com> writes:
> >> >But I doubt, that "large project" (in modern terms, like Linux kernel for
> >> >example) in Forth ever existed.
> >> There is one large project, the CCS software, which had 1.2M lines of
> >> code last time the number was reported here. Linux had 27M lines of
> >> code in 2020, but I think that 1.2M lines is plenty large.
> >
> > I'm curious, were they always developing in Forth since 1978? But this is
> > "large" (>1M sloc) application. I've actually downloaded trial to check if
> > it is still in Forth. It is. :)
> >
> > Probably their lead developer is a huge Forth fan...
> https://forth-standard.org/standard/members-2x

Well, can somebody ask Willem Botha to write a postmortem to their
software history? They are probably the largest Forth users in the world.
Like Janestreet for Ocaml.

On Monday, November 22, 2021 at 2:59:12 PM UTC+1, Brian Fox wrote:
> Over on a hobby forum that I frequent, a couple competent people
> who have written their own Forth systems but work professionally in C,
> Verilog and other more popular languages are asking for examples of large
> Forth projects. They know what Forth is but want to see how fluent
> users deal with large projects. One quote was "I understand Forth but feel
> like I still don't 'get it'"
>
> I know that much of what they would want to see is proprietary but
> are there any examples online that I could point them too?

Most examples are tools. That is not interesting because they cannot serve
as an example how to write applications.

The following come with executables that you can just run.

A disassembler reverse engineering tool:
https://github.com/albertvanderhorst/ciasdis
Famous for disassembling colorforth,

The win32forth distribution has several applications, such as a sudoku interactive solver
with colors, changing size, menu's what not.

Then there is manx2. Originally intended to play mechanical instruments,
it plays musical scores on the internal multi media of MS-windows as one
of the instruments. It awaits publication.
The program runs on 32/64 bit linux, intel and arm and 32 bits windows,
as standalone program, despite the hardware differences.

[The original iforth come with manx, but this is not a standalone program, and
you have to buy a iforth license, but the manx source is available.
This is a substantial program but probably not where you're after.
A plethora of iforthism among other things makes it unportable.
(and I have tried, and I have decennia experience in maintenance of often
poorly documented/designed software.)
manx2 is a rewrite, keeping the original format of musical scores. ]

On Tuesday, December 14, 2021 at 6:42:29 AM UTC-5, Albert van der Horst wrote:
> On Monday, November 22, 2021 at 2:59:12 PM UTC+1, Brian Fox wrote:
> > Over on a hobby forum that I frequent, a couple competent people
> > who have written their own Forth systems but work professionally in C,
> > Verilog and other more popular languages are asking for examples of large
> > Forth projects. They know what Forth is but want to see how fluent
> > users deal with large projects. One quote was "I understand Forth but feel
> > like I still don't 'get it'"
> >
> > I know that much of what they would want to see is proprietary but
> > are there any examples online that I could point them too?
> Most examples are tools. That is not interesting because they cannot serve
> as an example how to write applications.
>
> The following come with executables that you can just run.
>
> A disassembler reverse engineering tool:
> https://github.com/albertvanderhorst/ciasdis
> Famous for disassembling colorforth,
>
> The win32forth distribution has several applications, such as a sudoku interactive solver
> with colors, changing size, menu's what not.
>
> Then there is manx2. Originally intended to play mechanical instruments,
> it plays musical scores on the internal multi media of MS-windows as one
> of the instruments. It awaits publication.
> The program runs on 32/64 bit linux, intel and arm and 32 bits windows,
> as standalone program, despite the hardware differences.
>
> [The original iforth come with manx, but this is not a standalone program, and
> you have to buy a iforth license, but the manx source is available.
> This is a substantial program but probably not where you're after.
> A plethora of iforthism among other things makes it unportable.
> (and I have tried, and I have decennia experience in maintenance of often
> poorly documented/designed software.)
> manx2 is a rewrite, keeping the original format of musical scores. ]

Thanks Albert.

I think I made some headway after giving links to VFX, SwiftForth and the
NASA projects and a link I found to SP-Forth libraries which are quite
comprehensive.

Most of the forum people did not know there were commercial dev. systems
or compilers that went beyond the hobby systems one finds online or that
Forth was ever used for anything serious.

On 15/12/2021 12:22, Brian Fox wrote:
> ...
> Most of the forum people did not know there were commercial dev. systems
> or compilers that went beyond the hobby systems one finds online or that
> Forth was ever used for anything serious.

They'd not heard of FORTH Inc ? Must be some youngsters in that group :)

dxforth schrieb am Mittwoch, 15. Dezember 2021 um 03:07:27 UTC+1:
> On 15/12/2021 12:22, Brian Fox wrote:
> > ...
> > Most of the forum people did not know there were commercial dev. systems
> > or compilers that went beyond the hobby systems one finds online or that
> > Forth was ever used for anything serious.
> They'd not heard of FORTH Inc ? Must be some youngsters in that group :)
You mean below 60? Unbelievable!