[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Turning GNU into a bottom-up organization

From: Samuel Thibault
Subject: Re: Turning GNU into a bottom-up organization
Date: Thu, 24 Oct 2019 18:52:14 +0200
User-agent: NeoMutt/20170609 (1.8.3)


Mostly giving my feedback on the matter as "the Debian guy".

František Kučera, le mar. 22 oct. 2019 10:16:05 +0200, a ecrit:
> Dne 21. 10. 19 v 17:08 Mark Wielaard napsal(a):
> > In practice GNU already is mostly a bottom-up organization, where the
> > GNU hackers that do the actual work shape the project, but it would be
> > nice to make it more formally so.
> The problem with this approach is the risk of hostile takeover. There
> are corporations (e.g. those that profit from proprietary
> software/cloud) or governments of countries like China, Russia or USA
> (with their secret services and agencies) that have almost unlimited
> (from our point of view) financial and developers resources – which
> allows them to bend such organization according to their needs.

I haven't seen the Debian project being hit by such concern, notably,
even if it is a bottom-up project by so many means.

There is a Debian Project Leader, who is elected, and he does not
actually decide very much in the project. Notably, integrating new
Developers is done through advocation by an existing Developer, and
review by an Application Manager, who checks knowledge of Debian,
existing work, etc., more on this in

František Kučera, le mar. 22 oct. 2019 10:16:05 +0200, a ecrit:
> So if this is to have a chance of success, there must be a rigid
> (immutable) constitution which guarantees the principles in the long
> term.

Mark Wielaard, le mar. 22 oct. 2019 01:04:44 +0200, a ecrit:
> It would be good to try to distill a small core of musts, a summary
> of sorts, that can be more easily communicated as a kind of social
> contract for GNU.
> it would be excellent if we has some kind of Social Contract
> for the GNU project to make that a more formal requirement. Those who
> don't agree can then of course still contribute, but they wouldn't get
> responsibilities for the project.

Ludovic Courtès, le mar. 22 oct. 2019 17:00:14 +0200, a ecrit:
> To make that clearer, having some sort of a “social contract” for
> maintainers to sign, where they explicitly commit to defending a
> specified set of core values of the project, would be great.  It would
> make it easier to share overarching responsibilities among maintainers
> because they’d have explicitly committed to furthering the project’s
> missions.

As a Debian Developer, I can only agree with that. To become a
Debian Developer, one has to accept the Debian Social Contract . This kind of document (I'm not
saying the Debian Social Contract in particular, there are point in
there which the GNU project doesn't agree on ; I am saying a document
with the same type of content: writing down the MUST principles of the
project) can be a ground for making sure that future GNU maintainers
keep on the principles that were drawn from the beginning of the GNU
project, instead of having to rely on one person who will not be here
any more someday. I believe that the presence of a social contract is
one of the points that allowed Debian to outlive its founders (yes,
litteraly, unfortunately).

Alfred M. Szmidt, le mer. 23 oct. 2019 15:12:28 -0400, a ecrit:
> > I agree.  I think in practice many (most?) maintainers not only agree to
> > uphold the free software values, but also share them.
> In practise, we cannot know -- making the assumption is dangerous.  It
> also excludes people who still wish to contribute to the GNU project
> but have a different set of values.

Sure. That's also what happens in Debian: you can contribute to Debian
without being a Debian Developer. You will just not have vote rights,
and no upload permission. A Developer will always review your changes
before uploading them, thus making sure they respect the social contract

Jean Louis, le mar. 22 oct. 2019 23:19:32 +0530, a ecrit:
> In that sense GNU project in synergy with so many other free software
> projects not only achieved its ultimate goal to provide the free
> operating system, it even over-lived itself.

And that is actually one of the things what worries me: new programmers
do not really know about the GNU project. They do know a bit about the
GPL etc. but do not realize that "GNU" in gcc and glibc etc. does not
mean "GPL", but that it's the converse, these are part of a broader
project called GNU (and thus happen to use a GNU licence). The fact
that almost all GNU mailing lists are private strongly participates to
this ignorance of the GNU project. In the end, this leads to the risk
of not actually seeing fresh blood help maintaining the GNU project. If
newcomers can not easily see the discussions at stake, and instead see
arbitrary barriers to contributing to the project, they will just write
their own software on github, and not even realize that they should care
about making their software free. I believe we do not want that.

Ruben Safir, le mar. 22 oct. 2019 15:08:34 -0400, a ecrit:
> If you don't like what is happing, just fork it.

No. We do not want to start something else (we could have done it
already long time ago). We want to save the GNU project from becoming
irrelevant in the long run. The GNU project is too precious for free
software that we do not want to split its manpower into forks.

Alfred M. Szmidt, le mer. 23 oct. 2019 15:12:58 -0400, a ecrit:
> > RMS is a bus factor of one.
> That is true, and that is something that is already being worked on on
> reducing.  None of this was a grave issue last year, 5 years ago, or
> 10 years ago it can surley wait a while more.  So a bit of patience.

That "10 years ago" figure strikes me: 8 years ago, I happened to be at
the GNU Hackers Meeting, and questions raised in the threads today were
already raising at the time. I can only note that really not much has
progressed in the concerns that were discussed at the time. Will another
10 years be needed? Yes, we can be patient if we actually see progress.
I do not really see any.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]