See
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1039051

In this bug report Peter Pentchev admits that he has neglected
gforth in Debian for decades.

I know that you do not look it up. so here it follows

------------------------------------------------

Package: wnpp
Severity: normal
X-Debbugs-Cc: gforth@packages.debian.org, roam@debian.org
Control: affects -1 + src:gforth

I request an adopter for the gforth package. Turns out that in the past
couple of years I have not given it enough care and attention.

There is a Git repository containing all my Debian packaging work at
https://gitlab.com/gforth/pkg-debian-full

The package was in good shape at the time of its last update, but
things have moved on since then. Notably, I tried to bring in some
new upstream beta versions every now and then, but they failed their
build-time tests, and it would appear overly optimistic to keep
pretending that I will track those test failures down.

The package description is:
This is the GNU'ish implementation of a Forth programming environment.
.
Forth, as a language, is best known for being stack-based, and completely
extensible. Each Forth environment provides one or more dictionaries of
pre-defined words, and programming in Forth consists of defining and
executing new words that are combinations of previously defined words. It
has been said that learning Forth changes forever the way you think about
writing programs.
.
For more information about Forth, visit the Forth Interest Group web site
at http://www.forth.org/fig.html.

Thanks in advance to whomever picks this package up!

========================================================
Background is that he doesn't no longer want to be responsible for
this package.
This is a Request For Adoption. That is formal (RFA bug report).
The idea is that someone comes forward from the Debian circles to take
responsibility for this package, the technical term is "adopt it".

This is dangerous! If adoption fails, it is possible that Debian drops
the package from the distribution.
On the other hand, it is an opportunity. You can propose yourself to
Peter. Warning, you are vetted, must register with PGP etc.
It is not clear if he responds to e-mail. Most debian maintainers don't
in my experience. What works is, subscribe to the wnpp mailing list and
follow up on the bug report. (Send "help" to majordomo@lists.debian.org).

[You probably know that Debian has rejected ciforth for distribution.
After years (!) of work the package was formally acceptable to Debian
and a RFA request was accepted. Then there was nobody willing to
adopt the package and the package was dropped from the Debian archives
without prior notice.]

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

albert@cherry.(none) (albert) writes:
>See
>https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1039051
>
>In this bug report Peter Pentchev admits that he has neglected
>gforth in Debian for decades.

Or you can blame me, because I have not yet caught up
documentationwise with the state of Gforth 1.0 that we want to
document.

Or you can blame yourself:

https://gforth.org/

explains several ways to get a current snapshot (or to build from git,
but snapshots have the advantage of having passed more testing),
including getting and installing a .dep package or Docker image, so
you don't have to build it from the source.

What is the fault of the gforth package's maintainer prior to gcc-7
(about 2017): dynamic native code generation is disabled in the
Debian-distributed package, causing a slowdown. gcc-7 generates code
in a way that results in gforth-0.7 disabling dynamic native code
generation on its own. In the development version we have a
workaround for that.

- 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: https://forth-standard.org/
EuroForth 2023: https://euro.theforth.net/2023

In article <2023Nov13.071730@mips.complang.tuwien.ac.at>,
Anton Ertl <anton@mips.complang.tuwien.ac.at> wrote:
>albert@cherry.(none) (albert) writes:
>>See
>>https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1039051
>>
>>In this bug report Peter Pentchev admits that he has neglected
>>gforth in Debian for decades.
>
>Or you can blame me, because I have not yet caught up
>documentationwise with the state of Gforth 1.0 that we want to
>document.
>
>Or you can blame yourself:

In reality I myself do not go to the trouble of googling "installing gforth"
if gforth is present in my Linux distribution.

>
>https://gforth.org/
>
>explains several ways to get a current snapshot (or to build from git,
>but snapshots have the advantage of having passed more testing),
>including getting and installing a .dep package or Docker image, so
>you don't have to build it from the source.
>
>What is the fault of the gforth package's maintainer prior to gcc-7
>(about 2017): dynamic native code generation is disabled in the
>Debian-distributed package, causing a slowdown. gcc-7 generates code
>in a way that results in gforth-0.7 disabling dynamic native code
>generation on its own. In the development version we have a
>workaround for that.

I worked years to have ciforth accepted in Debian and with a good
reason.
Almost all novices do
apt install gforth
and go no further. We should be glad that they do even that.
Almost all questions about gforth in reddit apply to version 0.7.3.

I'm jealous that gforth is present in Debian, and ciforth is not.

Groetjes Albert

>
>- 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: https://forth-standard.org/
> EuroForth 2023: https://euro.theforth.net/2023
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

On Gforth.org it is written:

Download / Install
0.7.3 (Latest release)

This first line bemused me for quite some time until I found out, that 0.7.3 was about 12 years old, and that the Bleeding edge snapshot mentioned below and the hints to the Git repo etc. were the way to go for the most recent version.
So, at least this first line should be corrected somehow. I assume that users often have a look for more up-to-date versions and go to the project's website. On Reddit, someone answered: But there this version is marked as latest release. That's the result.
Sure, the best way would be a new adopter who really cares. But there are more ways to improve the situation than only waiting for a new adopter who might not come within the next years or so...

N. A. McBee (Kolloquia-DE) wrote:

> So, at least this first line should be corrected somehow. I assume that users often have a look for more up-to-date versions and go to the project's website. On Reddit, someone answered: But there this version is marked as latest release. That's the result.
> Sure, the best way would be a new adopter who really cares. But there are more ways to improve the situation than only waiting for a new adopter who might not come within the next years or so...

You get what you pay for. And is it broken? Bits and bytes don't age.

It is similar to the TCC Tiny C Compiler. The "old" version
could even compile a Linux kernel. Still, many non-contributors
clamour for new releases with new features.
A bit unfair, don't you think?

In article <7cd51696617940e3b735cea670a1decf@news.novabbs.com>,
minforth <minforth@gmx.net> wrote:
>N. A. McBee (Kolloquia-DE) wrote:
>
>> So, at least this first line should be corrected somehow. I assume that users often have a look for more up-to-date versions and go to the project's
>website. On Reddit, someone answered: But there this version is marked as latest release. That's the result.
>> Sure, the best way would be a new adopter who really cares. But there are more ways to improve the situation than only waiting for a new adopter who might
>not come within the next years or so...
>
>You get what you pay for. And is it broken? Bits and bytes don't age.
>
>It is similar to the TCC Tiny C Compiler. The "old" version
>could even compile a Linux kernel. Still, many non-contributors
>clamour for new releases with new features.
>A bit unfair, don't you think?

I agree that the 0.7.3 version is perfectly fine to use.
People are stuck with 0.7.3 in the sense that the latest and greatest
examples of gforth accomplishments may fail on their trusted platform.

As for ciforth I resist the temptation to renew for the sake of it.
The latest great overhaul was in 2005, when I was forced to replace
REQUIRE with WANT.

All additions are backward compatible anyway.
The most important addition is floating point for the i86.
You can load a two screen fixed point package in the coming
release, but you don't have to. Etc.

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

On Tuesday, 28 November 2023 at 11:43:16 UTC+1, minforth wrote:
> It is similar to the TCC Tiny C Compiler. The "old" version
> could even compile a Linux kernel. Still, many non-contributors
> clamour for new releases with new features.
> A bit unfair, don't you think?

Hey, the only thing I said was that the wording on the Gforth download page should be improved because I personally find it misleading. That the Debian maintainer could be a littlebit more active with respect to Gforth is another aspect.

I didn't complain, I appreciate and have the greatest respect for the folks offering such great software at no cost to the public, that's absolutely clear, I'd say.

My intention was only to improve the situation a bit for somewhat less agile users, that's all. And I never said 0.7.3 wasn't usable or anything like that.

The starting point of this discussion was that you get gforth v0.7.3 if you just type
apt install gforth
The reason is that the adopter of the gforth package did not adjust it to the latest version, that's all. (And didn't so for years...)
Nobody asked for more updates just for the sake of updates or something like that. The point is that the updates are there, Anton and his collaborators have updated Gforth on their side this month (November 2023) and they have released many versions since 0.7.3. One only has to check the Gforth download page accurately, then it is no problem at all to get the latest version.
I hope this clarifies my view ;) Thx!

In article <3ee8d0f5-949a-4237-9fb1-3af5d365bf5cn@googlegroups.com>,
N. A. McBee (Kolloquia-DE) <hmbromkamp@gmail.com> wrote:
>The starting point of this discussion was that you get gforth v0.7.3 if you just type
>apt install gforth
>The reason is that the adopter of the gforth package did not adjust it to the latest version, that's all. (And didn't so for years...)
>Nobody asked for more updates just for the sake of updates or something like that. The point is that the updates are there, Anton and his collaborators have
>updated Gforth on their side this month (November 2023) and they have released many versions since 0.7.3. One only has to check the Gforth download page
>accurately, then it is no problem at all to get the latest version.
>I hope this clarifies my view ;) Thx!

That is an option for tenacious people. Most wil not go beyond "apt install".

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -