Re: What do you win by moving things to non-free?
On Sun, Apr 17, 2005 at 10:39:56PM -0700, Matt R Hall wrote:
> I really don't think he is too fair to my point of view. At least from
> my view, he always seems to prefer attacking me on red herring mail
> editor technicalities or identity issues, instead of covering the
> points I bring up.
Comments on mail etiquette or technical mail issues are a tangent from
the discussion. If you're interpreting simple, helpful suggestions as
"attacks", it may be difficult to have a civil discussion.
> > > I agree that there is a good reason for non-free. I believe the
> > > justification / reason / purpose of non-free is to earmark data
> > > (software, etc.) which cannot be redistributed by CD vendors. Java
(snip)
> That's not my understanding. It seems to me that non-free software is
> stuff that is encumbered in such a way that it's hard to really use it
> or debug it or redistribute it or develop it. This does not apply to
> the GFDL unless it's read in a pretty extreme (non middle-of-the-road)
> fashion.
The text which I replied to, above, says: "to earmark data which cannot be
redistributed by CD vendors". Doing that is, as I said, not the fundamental
purpose of non-free.
> Alright. The difference (as I see it) is that Qmail's limits on
> modification make it substantially less useful. The GFDL limits make
> GFDL content less modifiable, with the advantage of preventing
> censorship.
So if I add a feature to /bin/echo ("echo --irrelevant") to display my
own political manifesto, and distribute it under terms prohibiting its
removal or modification, you'd call that an "advantage"?
> In some ways, the GFDL is more free than a typical free
> license like BSD because GFDL prevents consumers / users / downstream
> people from twisting the author's intent.
The same argument can be used to claim that Qmail is more free, because
its license prevents people from "twisting the author's intent" by
changing it in ways that he doesn't like. I don't see the difference.
> You got me on this one. Forgive my frankness, but I fucked that
> paragraph up without realizing it until I saw your reply. What I
> _really_ meant was, "Using non-free should be doable _FOR USERS_, but
> not to the extent it becomes a knee-jerk reflex or default course of
> action."
I don't like the idea that such important bits are in non-free that almost
everyone will use non-free--I think we're all in full agreement there.
However, I don't believe lowering standards, allowing licenses to apply more
and more restrictions on users, and greasing some hazardously slippery slopes,
is an acceptable solution.
> There is however a balance point. It could be argued that the GFDL is
> approximately as free as the DFSG rules, but in a different way. GFDL
> is a license that grants some more privilege to an author to preserve
> his/her original intent and/or meaning without interference.
The GFDL is not DFSG-free. It's not even close. This isn't a matter
of interpretation of the GFDL, nor of the DFSG. The DFSG explicitly
requires that licenses permit the creation of derived works. The GFDL
explicitly prohibits creating derived works from the GNU Manifesto.
The GNU Manifesto, as bolted onto GNU documentation, fails the DFSG
(and, as a side-effect, so does the stuff it's permanently bolted to).
If you want your works to "preserve your original intent without
interference", that's fine--but that is a fundamentally non-free
objective.
> This could have an advantage for users (DFSG's #1 priority) because
> they can be sure that they have received a document that the original
> creator would approve of. They know the author's original intent and
> philosophical reason for doing something some way. What about that side
> of the GFDL coin? Nobody seems to have mentioned this yet, to my
> estimation.
Free Software may (and regularly do) require that modified works not be
presented as the original. An author can require that a modified work
be clearly marked as such. This is explicitly permitted by DFSG#4, and
is provided for just this reason. Restrictions beyond those (such as
"can not be modified at all") listed in DFSG#4 for the purpose of "the
author's original intent" go too far.
> Another thing, it's not fair for you to think / say people should not
> be here just because they have a different definition of freedom than
> you do.
If your definition of freedom is not in general agreement with Debian's,
then I don't know why you're here. I don't expect everyone to be in
full agreement, but I'm extremely disappointed that there's anyone here
that would actually argue that invariant sections are Free; to me, it
seems as strange as if people here were arguing that shareware is Free,
too (hey, it helps fund programmers, to allow them to continue their work;
that's in "user's interests", isn't it?)
> Also, there are certain things I have posted to Debian lists under my
> own name that turned into really unpleasant boat anchors to past points
> of view that are no longer relevant.
If your position changes over time--as most people's do, on various
issues--you can simply say so. "I used to think that; I changed my mind."
--
Glenn Maynard
Reply to: