Re: [GNU-linux-libre] Uruk GNU/Linux evaluation

[Jaromil]

Ineiev wrote:

> > > The distro must be able to fix bugs in its packages; when they
> > > use other people's repositories (which is the case for Uruk
> > > GNU/Linux) they effectively can't do this (not directly).

> On Mon, Jun 27, 2016 at 03:43:18PM +0200, Jaromil wrote:

> > ok, but this is not a condition that is directly related to being
> > 100% free. it is a (debatable) concern on quality assurance that
> > has nothing to do with being 100% free.  >

On Mon, 27 Jun 2016, Ineiev wrote:

> No, but it's directly related to being actively maintained, which is
> a condition for endorsing as a fully free distro; and this
> endorsement is exactly what we discuss.

I disagree with applying this criteria, even worst when based on
narrow yet blurred definitions as "self hosted repository".

A recent example of the problem caused is here:

https://lists.nongnu.org/archive/html/gnu-linux-libre/2015-04/msg00013.html

where RLSD, an excellent distro which is 100% free, has been excluded
on the basis of not being "self hosted" while its nature would
perfectly justify cross-compilation.

Re-reading the RLSD thread I think it was excluded with a rather dull
application of zealotry.

I think people here should perceive their own agency as that of
curators, not gatekeepers.

In case a proper curatorial activity is beyond your reach, I invite
you and colleagues to reconsider this position and limit this
initiative to review distributions that are 100% free. The definition
of "actively maintained" is too blurred and connects to other issues
and indicators that are debatable case-by-case and not even specified
in the 100% free definition. As a matter of fact many of the
distributions that are today considered 100% free would have never
made it, no matter how 100% free, useful and significant for people to
discover the free software movement they ever where.

With this initiative we need to be able to include and nurture all
sorts of artisanal efforts. If we go beyond the specific criteria of
100% free then I'd rather start consider how welcoming and useful is
the distribution to introduce people to free software ethics, with its
artwork and wording and translation and functionalities, rather than
applying this "self hosted" criteria, a rather narrow-minded principle
which can be well misused by gatekeeping behavior.

At last, this conversation is not anymore related just to Uruk. I do
read code and after reading Uruk's script to update I'd rather not
endorse it based on the quality of the code, which I hope will be
improved. I guess Uruk would be better off without any update, or
adopting amprolla to distribute its packages as overlay to Trisquel's
repository.

> Jaromil wrote:

> > For what we are concerned here, a distribution can be 100% free as-is
> > and without further upgrades, with one exception included in the 100%
> > free agreement for a "bounty".
> >
> > the "bounty" in brief: the maintainer(s) of a distribution should be
> > available to act and remove any non-free software that will be
> > spotted.
> >
> > To be available to do this does not entails the overhead of
> > maintaining an entire package repository! nor the imposition of using
> > a package manager instead of another, or perhaps even make your own
> > packaging, or just distribute iso updates, or squashed /usr... there
> > are many ways to update an OS..
> >
> > I believe that Uruk can be 100% free even without offering a whole
> > package repository, but just by publishing all sources (and
> > modifications to existing Trisquel's sources) and agreeing to the
> > bounty.
> 

Ineiev wrote:

> I'd like, on the one hand, someone who may speak for the GNU project
> to confirm that such workflow is acceptable, and on the other hand,
> Uruk GNU/Linux developers to state that they've implemented it.



-- 
~.,_   Denis Roio aka Jaromil    http://Dyne.org think &do tank
    "+.   CTO and co-founder      free/open source developers
      @)   ⚷ crypto κρυπτο крипто गुप्त् 加密 האנוסים المشفره
    @@)  GnuPG: 6113D89C A825C5CE DD02C872 73B35DA5 4ACB7D10
(@@@)  opmsg:73a8e097a038d82b 8afb4c05804bda0d 281b3880fbc19b88

signature.asc
Description: Digital signature