I have this wonderfully small Forth compiler/scripter/interpreter lina.
It is written in portable assembler, i.e. a base line and macro's to
transform to some of the popular assemblers like gas, fas and nasm.
I proposed the gas version to Debian, but they insisted that it was not
proper
source, and that the database and its mechanism should be part of the
release.
This is a terribly complicated system and a fasm version of the source
is
the only game in town for someone actually wanting to exercise the right
to transform the program. Also I am of the opinion that different
assembler sources that express the exact same instructions and assemble
to byte for byte identical programs are not actually different. There is
nothing gained by having the underlying
database towards the purpose of freedom. The database can generate
16/32/64 bit
Forths, for MSDOS, WINDOWS DPMI, Windows dll 32/64 and of course linux
32/64.
To install that to have a wonderfully small compiler with one assembler
source, one executable, one documentation file, is an abomination. Do I
take someones freedom away by not publishing it? Wrong question, I do
publish it under a GPL license. 1]
However the quality is not such that I dare to propose it to be in a
linux distribution.

I tried and have actually succeeded in 2019 to make an excerpt of the
database
to make linux version, put it in github (as if my releases were invalid
as
upstream source), and have a deb package that actually passed all those
lint
stuff. I had to become a maintainer and instal gpg keys and what not.
My sponsor at the time had no interest in Forth and did not support my
filosofy to have a complicated database to serve to generate a simple
and
transparent compiler. That went south. So at last there was a compliant
debian archive that addressed all concerns debian had, but no sponsor.
2]
I uploaded it, then filed a Request For Sponsoring bug, and it expired.

There are Forth implementations in Debian, apart from gforth, that are a
travesty.
yforth goes virtually undocumented and its man page says that it is not
to be taken seriously. I have contacted the maintainer Bdale Garbee
several times, to ask him whether he would sponsor lina. I didnt even
get an email back with a refusal.
Oh, and did I tell you that on my Debian system yforth just crashes?

lina is small but useful. I hope my reverse engineering system 3] is a
convincing
example. Know projecteuler.net ? I am the albert in the top of the
Netherlands
scorers with 500+ problems solved, most of them with lina.

Now comes the question.
Can a Debian package that is fully compliant with Debian guide lines and
has no dependance on any other package -- leave alone systemd -- be
acceptable to Devuan
if it is not part of Debian?

Groetjes Albert

1. The compiler factory (includes releases)
https://github.com/albertvanderhorst/ciforth

2. The Debian politically correct excerpt for 32 bit linux
https://github.com/albertvanderhorst/lina

3. An application, a reverse engineering assembler/disassembler for i86
https://github.com/albertvanderhorst/ciasdis

--
Suffering is the prerogative of the strong, the weak -- perish.
Albert van der Horst