Re: [GNU-linux-libre] Adding some scummvm game(s) to the "List of software that does not respect the Free System Distribution Guidelines".

[Denis 'GNUtoo' Carikli]

On Sun, 11 Jun 2023 17:17:50 -0400
Richard Stallman <rms@gnu.org> wrote:
>   > Here we have a virtual machine / runtime environment (ScummVM),
>   > and we're unsure if there is 100% free software programs that can
>   > run in it.  
> 
> I think the question at hand is a little different from that.
> I don't think that _some nonzero amount of free software that runs
> on the environment is necessarily enough to outweigh the fact
> that its main use is as a way to run nonfree software.
If we see things with the eyes of individual users, for regular
distributions, and traditional usages, it indeed makes no sense to ship
software that would in practice favors more nonfree than free software,
especially if the supposed use case of the free software is not
important right now.

But what I'm concerned here is more the long term side effects of a
rule that requires to weight the usefulness of software and use cases
because it's not always easy to do and not everybody has the same
knowledge of how different computing fields use specific software due
to people having different area of expertise.

The consequence is that in some cases we would need to decide what
computing fields are more important than others. Is software testing
more relevant than playing games for Wine? What about Docker?

Since most of the docker images are not FSDG compliant, do we remove
Docker completely? Since Guix can also generate docker images we would
need to weight all the use cases to decide. And that can be complicated
and painful to do (especially if they are users that don't accept that
their perfectly fine use case has been removed and that they are now on
their own).

Also note that Guix is not always used as a regular distribution: its
packages are used for automatic testing, reproducible scientific
experiments, setting up infrastructure, etc. So its default use case
really depends on the people using it. And some FSDG compliant
distributions can also be used with Docker which also opens the door to
more use cases like automatic testing (docker is used a lot for that
too).

And if we don't cover these use cases in FSDG distributions, it would
give more advantage to non-FSDG compliant solutions and push people
toward them (for efficiency reasons for instance).

Personally if most socially accepted computing uses cases are covered by
FSDG distributions I think it would be great.

> Omitting a prpgram from free distros does not prohibit its existence.
> It does not stop users from getting it and installing it.  All this
> decision would mean is that our free distros would not offer it to
> people.
Here most of the use cases I mentioned (distributing software, automatic
testing, etc) were more targeted at group of people than individuals,
hence the need of collaboration for maintaining packages these use
cases depend on. Packaging software in Guix also has huge advantages
like reproducible builds, including some ability to re-test older
software versions, etc.

All that is difficult to do if you need to build and install software
on your own. Though the impact is very use-case specific here: if
ScummVM is stable enough, people wanting to write ScummVM games can
simply build a release and not care about all the features mentioned
above that are more relevant for other type of computing. ScummVM is
also relatively easy to build compared to a Java virtual machine or a
browser like Icecat. So the difficulty of building depends a lot on the
software.

The consequence is that the effect on people really depends on use
cases.

> Most of what you've said in your message seems to be arguing
> against something I am not advocating.
Right. I hope that this time my mail gets better to the point.

Denis.

pgp1vb7faWonY.pgp
Description: OpenPGP digital signature