Re: Firmware & Social Contract: GR proposal
On Fri, Sep 08, 2006 at 12:01:37AM -0700, Thomas Bushnell BSG wrote:
> Steve Langasek <vorlon@debian.org> writes:
>
> > Who is confident of this, and why? I'm not confident of this at all; I'm
> > not sure that the idea of forcing sourceless firmware out of main is even an
> > idea that the majority of developers agree with, and Joey Hess has pointed
> > out to us reasons why providing separate free/non-free install media might
> > be a strategically poor use of our time in the *long term*, even if the work
> > of splitting out this firmware proved manageable and there were sufficient
> > volunteers to do this work.
>
> So this gives me pause.
>
> I've been instructed that it's ok to vote for one of these
> resolutions, because it's only a way to get etch out the door, and we
> can come into real compliance with the Social Contract for etch+1.
>
> I have expressed some skepticism, being rather convinced that the
> actual facts are that there are people who are happy to have the
> kernel simply *never* come into compliance with the DFSG, for whatever
> reasons, and that they have been dragging their feet in bringing it
> into compliance.
>
> One of the people hinting at this has been Steve, who basically said
> to me recently that for some packages, they would get booted from the
> release for violating the DFSG, and for other packages, we just wink
> and nod.
>
> Now we have it flat out: Steve thinks perhaps we will simply never
> bring the kernel packages into compliance with the DFSG.
Come on, it is not because Steve is sceptic or whatever, or that he had his
own GR which was granted with much critic, or what do i know, that you have to
ignore the work of the other folks involved.
Notice how Frederik's proposal gives some strong commitment on not ignoring
this issue, and how three of the most prominent kernel-team members have
seconded it, and how we clearly stated our lack of ressources, and repeteadly
asked for help on this (including to you, and people like Larry and Nerode).
> So let's not hear about etch vs. etch+1; let's not hear about some
> special thing for just this release.
>
> This is sounding like the behavior of the US Congress, which likes to
> continually extend copyrights for one "limited" term after another,
> thus producing the reality that copyright terms in the US are now
> infinite.
Bullshit, we are in a much better shape with regard to this than we were in
sarge, and there has been tremendous work from the kernel team to improve the
kernel situation since the abysmal situation we had in sarge and before. Do
you so despreciate and diminish all our work, just because of what Steve says
above, Steve who is known for having a very controversial way to communicate
about these issues, see his GR here, or the vancouver proposal last year.
> Just so, the claim that we are making temporary concessions so that we
> can release is a cover for the real facts: some people simply do not
> think the DFSG operates as an absolute bar to the inclusion of
> non-free software in Debian.
Indeed, there are some people who feel that we should not solve this issue, or
even who are actively opposed to it. But they are a small group, even as you
and the others are a small group of vocal people with little actual actions to
back your words. And they will in no way be able to stop those willing to
solve this issue to go forward with it, and they have either agreed to not
stop others doing the work provided they are not asked to do so themselves, or
have no power to actually stop the move.
You are not being helpful in this with comments like this one, and you have
very little credibility left anyway with those people, so continuing to post
here is not going to help.
This plan is to :
1) vote about not having to wait for the non-free issue to be solved, before
releasing etch. It is of the secretary's opinion that the RMs have this
power all by themselves, but a vote would be nice nonetheless.
2) once this question is out, we will be able to tackle this issue in much
more sane and reflected ways, and work on actually fixing the issues instead
of holding flamewars which take everyone's time.
3) you can help in this by going over larry's list (please use the wiki page
http://wiki.debian.org/KernelFirmwareLicensing for it), and go over the
offending files with licencing problems, and add comments on them, and help
us identify the actual licensor/copyright holder, so that we can contact
them over licencing clarification.
4) remove the other firmwares into the non-free package. A few of those will
be happening for the etch release, like the qla2xxx ones, but others may
take more time. We will work on this, but it is useful to do so not with the
etch release pressure on us.
All the rest of the discussion are i believe not really worth it, there are
emotional affairs, like steve's or anthony's proposals, which once we have
moved most of the drivers and firmwares which needs it to non-free, will
probably be moot, altough a further discussion in calm and tranquility may be
interesting.
We already have all the right things in the social contract for inciting us to
remove those non-free bits, do you really want to bargain all this by forcing
a pre-etch release vote on a wider scale, which have right now every chance to
go against your wish. Please think a bit and be helpful in this effort,
instead of your current stance.
> So, rather than dance around redefining "software" and telling us that
> this is a just-this-once special-exception just-for-etch (never mind
No, you are dead wrong. Steve and some others wanted indeed to redefine
software and firmware as data and whatever, but there was care taken in the
redaction of Frederik's proposal to not fall into this trap, and indeed i was,
as you must admit, a very vocal opponent to Steve doing to.
If there are some issues with the current proposal, please work with us to
better it in a open and constructive way, but also note that this is not
really a position statement, like the other proposals, but a let's get etch
out of the way so we can do the real work one.
> that it is the second "just this once" special exception), can you
> please propose the DFSG or SC amendentment you really want, the one
> that clearly and unmistakably says "the following items can be
> included in Debian even though they are not free software", and drop
> the 100% promise that the Social Contract has always known?
Steve may do this, but this is not the aim of this current proposal, which
Steve apparently said he doesn't identify himself with.
Friendly,
Sven Luther
Reply to: