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

Re: Damage-control GR and XFree86 non-freedom

On Fri, Apr 30, 2004 at 05:18:05AM -0400, Nathanael Nerode wrote:
> > It's there because the issue's being worked on, is (or was considered)
> > unlikely to be resolved for sarge, and doesn't seem important to be
> > fixed for sarge.
> > No doubt another example of my hypocrisy and lust for
> > non-free software or whatever.
> Well, why do you think that the Social Contract amendment changes
> appropriate behavior for the *other* sarge-ignore cases and not for this
> one? 

So, I explained this in some other message somewhere around the same
time I posted the above, but I couldn't be bothered looking for that,
so I'll repeat it and expand on it a little.

Basically, licenses can come in three categories:

	* those we're sure are DFSG-free
	* those we're sure aren't DFSG-free
	* those we're not sure about

They can also be divided into licenses that we're sure are meant
to be free, ones we're not sure are meant to be free, and ones that
we're sure aren't meant to be free too.

So if we end up with concerns about a license for some software that's
in main, we end up with:

	* a license that's non-free, but was meant to be free
	* a license that's non-free, that might be meant to be free
	* a license that might be free, and is meant to be free
	* a license that might be free, whose intentions aren't clear

	* a license that might be free, that seems to have been meant
	  to be non-free
	* a license that's non-free, that seems to have been meant that way

DFSG-freeness isn't fundamentally about license texts: it's about whether
the author gives you permission to make derived works, distribute for
profit and so forth. If the author makes a mistake in writing legalese,
and actually intends to give more permission than ends up written down,
that's not a problem for DFSG-freeness, it's just a potential problem
in getting the law to match the authors' intent.

But we can't tell what the author's intention was without asking them --
we'd already misread what they wrote in their license in assuming they'd
meant their to give all the permissions we need, so there's no reason
to expect we're doing a better job now. And in most cases, when we do
actually ask it turns out they did mean their stuff to be free.

> If the only reason for the other sarge-ignore cases was that the Social
> Contract allowed wiggle-room, how was this justified (it's clearly
> programs)?

A license is, literally, permission from the rights holder. If the
author gives the permissions the DFSG requires, it doesn't matter if
that's actually accurately recorded in the debian/copyright file. The
same thing's true if the author doesn't give that permission, of course.

> If these practical reasons ("the issue's being worked on, is (or was
> considered) unlikely to be resolved for sarge, and doesn't seem important
> to be fixed for sarge") are good reasons,

No, those aren't good reasons, any more than they're good reasons to
release software with security holes.

Well, as far as releases go, anyway. As far as unstable goes it's a
different matter: that's where we work to try to get this stuff right.
It's reasonable to say that "the Debian system" is dists/stable/main;
and it might be reasonable to say that dists/testing/main is too.
dists/experimental, dists/unstable, people.debian.org, etc aren't.

> One good answer to any of these questions would do wonders to refute my
> claim of hypocrisy.  :-P

And in any event, what you're accusing me of is "inconsistency" --
hypocrisy is saying one thing and doing another. I've been saying that
the social contract only required us to apply the DFSG to programs for
ages, and I've been doing that.

Also, we got rid of the word "software" from the social contract so that
it would be clear that the social contract applied to everything, not
just programs, even though some folks interpret it that way. There's no
reason to use the GR as an excuse to tell folks who interpret "software"
as only referring to programs they're wrong -- GRs aren't an authority
on the English language, and the discussion's no longer relevant since
its not used in that context in the SC anymore.


Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
Don't assume I speak for anyone but myself. GPG signed mail preferred.

``Like the ski resort of girls looking for husbands and husbands looking
  for girls, the situation is not as symmetrical as it might seem.''

Attachment: signature.asc
Description: Digital signature

Reply to: