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

Re: Firmware & Social Contract: GR proposal

On Tue, Sep 05, 2006 at 10:49:49PM +1000, Anthony Towns wrote:
> On Tue, Sep 05, 2006 at 01:52:51PM +0200, Sven Luther wrote:
> > Indeed, but the fact that "delay until sarge release" won by a large majority,
> > shows that our DDs did indeed reaffirm the new SC, 
> In my opinion, it shows that at the time that was the best option on

No, it simply shows that at that time there was a large majority which
approved of the new SC wording, while at the same time didn't consider it was
worth delaying sarge in order to make it happen.

It is plain clear enough. The three options could have been :

  - Reaffirm the new social contract and delay sarge until we fix the issues.
  - Reaffirm the new social contract and give us until september to release
    sarge, or delay it until we fix the issues.
  - Reaffirm the new social contract but make an excpetion for sarge.

The third won, just like in this case there is a huge consensus on keeping
things like they are, and making an exception for etch on the issues still

> the table. One option that wasn't on the table at that point was having
> a proper discussion about what the social contract should contain. The

Well, it is called "Further Discussion", and that discussion was held during
the campaing leading to that third vote, and during the non-free removal vote
(much less with the cosmetic change second vote, which trokaned in those

> reason that wasn't on the table was that the sarge release was a higher
> priority, and there wasn't much chance of a useful debate in the meantime.

Sure, but people could have chosen further discussion, or to revert the social
contract change, or even propose new options. They didn't do though.

> > The fact that you made a huge statement back then about this, by resigning
> > from your RM position because of this further muddies the water about it.
> Actually, I made no statement about it whatsoever, either in the followup
> vote (in which I deliberately abstained by voting all options equally),
> nor in resigning (when I declined to comment publicly). I haven't

Acts speak louder than words :)

> commented on it since then until this year, as part of trying to meet
> the current social contract by getting the GFDL situation resolved
> earlier this year, and now, since we seem to be having the exact same
> problem again.

We do indeed have the exact same problem. We believe in our values, as
claimed in the social contract, but notice that we didn't have enough time to
fix all those issues during the etch release, which was noticably shorter than
the sarge release, and which saw lot of work and improvement in many areas, as
well as some ugly infighting which sapped energy and stuff.

No need for major reorientations of the project values and goals, just because
we did slip a bit in some areas.

> > > > Well, i believe that both of them basically said the same thing. 
> > > Yes, we've had that discussion; the key point is you used the word
> > > "software" to cover more of the contents of main, than others did.
> > Well, but do you truly believe that this was not the original meaning of the
> > original social contract ? 
> Yes, I do believe that "software" only covered programs, not "everything". 

Ok. Where you a DD already back then ? It clearly contradicts others position,
but in any case, this is beside the point, because last year the debian
project clearly stated by a 3:1 majority that the new wording suits them fine.

> It's what I thought I was agreeing to when I read the SC as part of
> signing up. The archives of -private on master during the SC/DFSG
> discussions match that impression, to my mind:
> ] The reason for Debian to be free software is so that our users can do
> ] cool stuff with it without running into legal problems. I would prefer
> ] not to make the document sound more like a "Pledge of Allegiance to Free
> ] Software", lest we be seen to be marching after ideology regardless of the
> ] end product and the users. The document is a contract we offer to our
> ] community about how we will behave with that community. We offer it to
> ] the people, we do not lay it upon the altar of free software.
>     (June 1997)

Ok, well, which is why we have non-free, and voted to keep it, despite the
whish of RMS and the FSF and a few others.

> ] It is not a legal document, it's a social one. Thus, I eschew the extremely
> ] careful language of a legal document in favor of something short and simple
> ] that communicates our philosophy without boring or intimidating the reader.
>     (June 1997)

This seems off-topic to the issue at hand.

> ] > There is no mention in the "social contract" that source formats for
> ] > documentation must be provided and that we must be able to modify the
> ] > documentation.
> ] I suppose you could read it that way, but it's not meant that way.
> ] I know of one potential Debian package that places a different copyright
> ] on the documentation than the program, but I don't know of any where the
> ] documentation is not available in source form.
> ]
> ] I'm willing to wait for the problem to happen before we address it.
>     (July 1997)

It doesn't speak in any way, and the project made a clear statement for this
around the GFDL, so ...

> > If anything else, it was strongly reafirmed in the
> > last of the pre-sarge non-free GRs. I mean, it did win and achieve the 3:1
> > supermajority, no ?
> Compared to reverting to the ambiguous original social contract, with
> no explicit clarification, nor any specific intention of doing better,
> "delay to sarge" beat "rescind entirely" by 257:122 votes (about 2:1),
> which was the weakest defeat of that option (compared to 301:45, 271:77,
> 273:86, 322:55 and 339:45). 

Still a bit more than 2:1 majority.

> As I've said, I think it's worth reconsidering at this point.

Why ? Most people are bored by this already, the choice for etch are clearly
pointed out, and there will be no major surprise in its result, and we will
fix the remaining issues during the etch+1 devel cycle.

> > > > > I think:
> > > > >     (b) The term "software" as used in the Social Contract shall be
> > > > >         presumed only to cover programs, scripts, libraries and similar
> > > > >         executable works to be executed directly as part of the Debian
> > > > >         System.
> > > > And the rest is what ? Hardware ? 
> > > Firmware, documentation, images, sounds, videos, fonts, etc.
> > Software has only a meaning in opposition to hardware, all the above clasify
> > as software in this most common of original definitions of the word, and
> > firmware foremost of them.
> Again, I'm aware you prefer a different definition. dict/FOLDOC is aware of it
> too:
>      Some claim that {documentation} (both paper and electronic) is
>      also software.  Others go further and define software to be
>      programs plus documentation though this does not correspond
>      with common usage.

Well, consider the examples shown in the GFDL discussion, where documentation
is part of a program, and then there is no doubt left.

> > Why would we need an alternative social contract. 
> Because we are unable to meet the terms of our current social contract,

Well, we are not unable, we chose to not put fixing those issues as our
maximum priority, over releasing etch in time, or fixing other issues, or
even having pointless flamewars about this issue and others here.

> and it doesn't actually reflect the priorities of our users or the free
> software community -- and supporting those groups of people is the entire
> point of our social contract.

Why doesn't is not support them ? Just because we have a lagging non-free
section, which has a strange and unclear status in our SC (not part of debian
but still part of debian) ? 

> Personally, I think there are other reasons to review it too -- I don't
> think explicitly naming "contrib" and "non-free" is a good idea; I think
> it's currently quite cumbersome and legalistic -- when I want to find

Indeed, it was kept so to make sure it didn't get removed, but there is some
serious clarification needed around this for us to fullfill our promise to our
users and free software. Accepting non-free software in main is not the right
path to this though.

> something that reminds me of why I think free software's important, I
> don't find our social contract a particularly good expression of it. When
> trying to promote free software and Debian, I find our experience with
> our social contract over the past few years tends to be a distraction
> from the very principles it's trying to promote -- making the best free
> operating system possible, and getting that into the hands of our users
> and the free software community.

So, let's cut on the thousands of new GR proposals, work together to get a
good set of options for the ballot, and let's vote, and devote our time to the
etch release. This one-new-GR-a-week game is not going to help in this way.


Sven Luther

Reply to: