Re: [ELPA] New package: repology.el

[Eli Zaretskii <eliz@gnu.org>]

> From: "Alfred M. Szmidt" <ams@gnu.org>
> Cc: bugs@gnu.support, arthur.miller@live.com, rms@gnu.org,
> 	dgutov@yandex.ru, ulm@gentoo.org, emacs-devel@gnu.org
> Date: Wed, 06 Jan 2021 16:18:59 -0500
> 
>    Suppose I want to download a potentially non-free package to study its
>    code, or even just to understand with what kind of license it is
>    distributed (since you say the site itself doesn't tell) -- is this
>    "verboten" as well in Jean Louis's interpretation of what a Free
>    Software follower should and shouldn't do?
> 
> I think the issue here isn't what you, I, or Jean Louis does -- but
> what the GNU project does.  

I was talking from the POV of a GNU maintainer.

> The GNU project doesn't want to give non-free software the remote
> changes of success, so mentioning or linking to it unless it is very
> well known would be working against its own goals.

So you are basically saying that no GNU package can ever help me find
out information about potentially non-free software?  Looking for such
information is something I as a GNU maintainer need to do almost every
day: someone proposes or mentions an Emacs package or a program that
can be useful with Emacs, and the first things I want/need to find out
are what kind of package is that? in what language is it written and
what OSes it supports? what is its license? does it rely on other
packages, and if so, what are their implementations and licenses?

You are in effect saying that no GNU package can ever help me with
these tasks, except if I'm looking for info about a Free Software
package, because divulging any information about a non-free package
means "promoting" it?  That is a very strange and radical
interpretation of "promotion", one that hurts our own cause by
preventing me from quickly and reliably answering the above basic
questions about any software package I ever come across.

> This is all written down in the GNU coding standards, what GNU
> project should or shouldn't do, in the now so overly quoted section
> 'References to Non-Free Software and Documentation'.

Some relevant quotes from that section:

  A GNU program should not recommend, promote, or grant legitimacy to
  the use of any non-free program.

No problem here.

  However, you should give only the necessary information to help
  those who already use the non-free program to use your program with
  it—don’t give, or refer to, any further information about the
  proprietary program, and don’t imply that the proprietary program
  enhances your program, or that its existence is in any way a good
  thing.

A somewhat different context not really relevant for our discussion,
but still: no problem here, either.  This text means that providing
the basic information about a program is OK, as long as we don't give
any impression that we endorse it or think that it is helping the
users in any way.

  Referring to a web site that describes or recommends a non-free
  program is promoting that program, so please do not make links to
  (or mention by name) web sites that contain such material.

I guess you mean this text, specifically the "describes" part (the
"recommends" part is fine, IMO).  I think interpreting the "describes"
part of this text literally, specifically when discussing packages
such as repology.el, is a mistake, which actually interferes with GNU
maintainers' job of quickly and reliably identifying software that is
non-free or otherwise problematic to our cause.  The ability to find
and study the code of any package is also an important method of
figuring out what the package really is and does; a literal
interpretation of the above text when applied to the current case is
thus also a mistake, IMO.

I wish the above text would make it more clear how to interpret it in
cases like this one.

  You should not refer to AT&T’s web site if that recommends AT&T’s
  non-free software packages; you should not refer to a page p that
  links to AT&T’s site presenting it as a place to get some non-free
  program, because that part of the page p itself recommends and
  legitimizes the non-free program.

I see no problem here, either: repology.org doesn't promote or
legitimize any of the packages whose information it records.

In sum, the text of that section in GSoC clearly wants to prevent any
recommendation or endorsement or favorite description of non-free
software, so I think its interpretation as precluding the display of
_any_ information of such software, specifically the basics: where to
find it and who maintains it -- such interpretation is a mistake, and
the text in GSoC should be made clearer in that regard.

Maybe we should continue this discussion on another, more suitable GNU
list.