On Fri, Jan 23, 2004 at 02:02:31AM -0500, Raul Miller wrote:
> On Fri, Jan 23, 2004 at 03:55:52PM +1000, Anthony Towns wrote:
> > It's a requirement that all the programs in main satisfy the requirements
> > of the DFSG. At present it's not a requirement that the text of copyright
> > licenses, or documentation satisfy the requirements of the DFSG.
> Guideline #3 never mentions "programs".
Yet, nevertheless the above is the case.
> > You can answer all aspects of the second question without knowing the
> > answer to the first, except one: "how should the social contract address
> > the issue of non-free".
> Well, in the sense that your "except one" is your first question.
Well, no. The first question is "Should the social contract
require/forbid/allow non-free?" The second question is "How do we describe
our requirements about non-free to best communicate our intentions?"
We can answer the second question right now. If we decide to change our
requirements, we may have to ask it again.
But they're fundamentally different questions.
> > > > > > At the moment the substantive options that have been discussed are:
> > > > > > [ ] Drop non-free
> > > > > > [ ] Limit non-free to partially-DFSG-free software
> > > > > > < > Keep non-free as is (unproposed)
> > > > > > while there are a whole raft of possible editorial changes.
> > > > > Even on that axis, there's more involved than that.
> > > > Really? I haven't seen any of it. Would you care to expound?
>
> > > The descriptive text in the social contract which defines our relationship
> > > with non-free software.
> > That sentence no verb.
> Right. That was the expanded "it".
> > What about it? You're changing the text, but what are you doing that'll
> > actually change how we behave?
> Hopefully, not a whole lot, other than avoiding a few arguments.
Uh. The axis was "substantive changes". You claimed there were more
than what I listed, and give an example that doesn't change anything
that we do?
> > It's easy to point anyone who thinks that to point five of the social
> > contract.
> And you get responses like Andrew's proposal to drop point five of
> the social contract.
Sure. Those are the people who want Debian not to distribute non-free.
> Even in your example, if the motive for "Let's change our goals" is
> "because they're not described in French", they're still not independent.
That would be true if it were the case. It's not, though.
> No, but I can point at people who will point at the wording of the social
> contract when asked why they don't think we should distribute non-free.
Well, no. They point at "Debian will be 100% free", and say that that's
a superior goal to the one we're actually aiming for.
> It's extremely difficult to get people to talk about their motivations.
Well, you're basing your entire resolution on an assumption about other
people's motives, without any support for that assumption at all. If your
resolution stands alone -- it's something that you want to see done in
and of itself -- that's fine; if it's a compromise that's intended to
demonstrate to some of the people who support an approach you don't like
that there's a better way, it's probably a waste of time.
> > > Anyways, there's nothing stopping you from proposing "goals only"
> > > amendments.
> > As I've already said to Branden and Andrew; I think it's better to discuss
> > why we want to do things, what we should do and how we do it, before
> > actually doing anything. I'm well aware that I can propose amendments.
> And how do you propose that happen?
Huh? I'm talking, you're talking, other people are talking. How else
would it happen?
> > Andrew's current proposal is *exactly* what a "goals only" proposal
> > should look like. It states what he wants to happen, and the minimum
> > number of changes to other things that need to be approved for it to
> > happen in a consistent manner.
> And he's provided no rationale for his changes.
> If that's exactly what you're advocating, I think I'm going to have to
> confess that I really don't know what you're advocating.
I have no idea what you think I'm talking about.
What I'm advocating is one single ballot to decide the issue of "What do
we do about non-free?". If we're going to drop it, we should do that: drop
it from the social contract, drop it from the archive, and drop everything
else that needs to go with it. If we're going to keep it, then we should
decide that, and nothing else.
If we want to do other things -- like tidy up the social contract --
we can do that in other ballots as necessary.
> > (Well, perfect but for the usual provisos about not dealing with contrib
> > at all)
> That, and he's proposing specific actions. If specific actions are goals,
> I don't see why different specific actions (which happen to include the
> exact wording of the social contract) are not goals.
No, the question is "Is Debian's goal to distribute all the software we
can, or to be 100% pure free software in everything we do?"
The answer to that question has implications we have to deal with:
changing the social contract, dropping stuff from the archive, dealing
with contrib. A good resolution, IMO, deals with the single question
that's at issue and its implications, and nothing else.
There are other questions, notably "Is there a better way of writing up
the social contract?", but they can be asked elsewhere.
> > (If it _is_ a factor, then, like Andrew has, we should make the minimum
> > number of changes to remedy whatever that problem is)
> Even if that problem is poor grammar?
I find it pretty difficult to believe that you really think the grammar
of the social contract is so bad that it actively hinders people
understanding Debian's purpose wrt non-free. This doesn't match my
experience reading it, or seeing other people read it, even remotely.
Maybe you have to explain it, and maybe you get people arguing about what
"Debian" actually means, and maybe you get people saying "but wouldn't
it be better if Debian really *was* 100% free?", but I can't imagine
anyone saying "You want to know about Debian and non-free? Well, make
sure you don't read the social contract, it's just confusing".
Cheers,
aj
--
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.
Linux.conf.au 2004 -- Because we could.
http://conf.linux.org.au/ -- Jan 12-17, 2004
Attachment:
signature.asc
Description: Digital signature