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

Re: DRAFT: debian-legal summary of the QPL

On Thu, Jul 22, 2004 at 03:58:13PM -0700, Josh Triplett wrote:
> Sven Luther wrote:
> > Well, so what. This only proves that there are licences which allow
> > proprietary product, and i would never voluntary release code under such a
> > licence, and they are other who don't.
> Neither would I.  However, my issue with the QPL is not that I would
> want to take the software proprietary, but that I might want to

But Brian's interest seem to be.

> distribute Free Software between a few people, giving those people all
> the freedoms expected for Free Software.

And ? What is the problem with that ? You can do it, the only point is that
you have, upon request to give the upstream author (probably anyone in the
chain of upstream authors) a copy of it if they request it. This can only be a
problem with the DFSG 1, if you consider such a thing a fee. But since the
cost to you is nil, i wonder if we can consider it as a fee, and also i
consider the fairness involved in refusing to give this to upstream that he
requests, while you had no problem in taking his work for free.

> If I take a GPLed program, modify it, and distribute the modified
> version among a few people, then as long as those people also have the
> source (or an offer for the source), then no one is being deprived of
> Freedom, and the software is not proprietary.

So what, how does that change with the QPL ? 

> Giving someone a binary without the source prevents them from exercising
> their rights over that software.  Giving someone no program at all does
> not restrict their rights, any more than giving someone no money: I am
> not obligated to distribute to them.

So, how is this relevant to the problem at hand ? 

> That said, I personally think that under almost all circumstances, it is
> a good idea to provide your changes upstream.

So, ...

Anyway, notice that QPL 6 doesn't speak about modification, but work which
link to a QPLed library. Not exactly the same thing.

> >>>Also, i also doubt that this is a way debian is confortable goind, and that
> >>>allowance of proprietary modifications over other considerations is the path
> >>>we are conforable threading.
> >>
> >>You doubt that which is the way Debian is comfortable going?
> > 
> > To make allowance to proprietary modification hoarder, like you seem to be.
> Again, modifications shared amonst a group, with everyone in that group
> having Freedom, are not proprietary.

Well, sure, but what is your moral ground for refusing the same modifications
to upstream ? 

> >>>>offers lots of permission, and asks nothing.  It's more generous than
> >>>>"fair".  The GPL is "fair": it offers many permissions, but some of
> >>>>them can only be exercised if you pass the same permissions on to
> >>>>others.  That is, it's a copyleft.  But it's probably the most
> >>>>restrictive you can be and still be "fair".
> >>>
> >>>Whatever. you want to modify ocaml, and not give back your changes to the
> >>>community. You have no sympathy from me, neither probably from a waste
> >>>majority of the debian project.
> >>>
> >>>Also you lying, claiming consensus, while there is no such thing, doesn't
> >>>endear you to me.
> >>
> >>I don't think personal insults really help anything.  What I see is a
> > 
> > Well, you claimed there was a consensus, while there is clearly no such thing.
> > Thus it is a lie intended to get the maintainer to take the course of action
> > you want through FUD, or at best a misinformed claim you should apologize for.
> The consensus on debian-legal seems to be strongly against the QPL.

Well, i see disenting voices in that conversation, and the consensus you
mention seems to be one of assertion, as it is quite lacking in analysis and
real arguments, don't you think. Some of the participants don't even seem to
have read the QPL, which makes the whole thing somewhat suspisious.

> >>On the other side of that issue, I see you and a couple others calling
> >>the people here lazy, deceitful, and lying.  You suggest that users
> >>should violate licenses because they won't get caught.
> > 
> > I didn't do that, but you obviously neither read nor understood what i said
> > nor the QPL itself, so ...
> To the best of my knowledge, everyone participating in this discussion
> has read the QPL.  I most certainly have, and I know several others

I don't think so, given the obvious cluelessness that comes with many of the
arguments made here. And if you did, you clearly didn't understand it, or
chose to gloss over parts that where not suporting your point. This kind of
attitude clearly casts some suspision of the quality of advice that comes with
debian-legal consensus.

> have.  Feel free to rebut the arguments of others, but please do not
> call people ignorant or accuse them of not reading the license.

I have, and you didn't respond to them when i first voiced them, and have to
this date not yet done so.

> >>The question of whether the QPL is free appears to have firm consensus
> >>from everyone involved in the debate, instead of standing on the
> >>sidelines and screaming.
> > 
> > A, a consensus is one where there is no discordant voice, right ? 
> Consensus is stronger than a simple majority, but it does not
> necessarily unanimous consensus.

Consensus: a general agreement about a matter of opinion.


> >>The question of what to *do* about that -- ask upstream authors to
> >>change their licenses, or modify the DFSG to make this issue explicit.
> > 
> > Bah, sure, i would do that immediately, if i would be given valable arguments.
> > Like that i would not only be ashamed to inform my upstream about this issue,
> > and thus show that debian follows a bunch of uninformed people in their
> > conclusions, i would be lying to them about this consensus issue, and also
> > lowering my future relationship with them over real issues.
> It is difficult to consider the possible avenues by which a licensor may
> abuse a license, when you have a particular licensor in mind who you do
> not believe would do that.  I understand that you don't think your
> upstream would sue people repeatedly and force them to come to a given
> jurisdiction, or request source repeatedly and without compensation, or
> attempt to take proprietary the modifications made by others.  However,
> lack of expected enforcement by one licensor does not make a license
> Free, or make those restrictions less important.

Sure, but if you follow this route, no software would be free, there has to be
a balance somewhere, and common sense should apply also.

> The "Tentacles of Evil" test is helpful to consider here.  Consider the

The various tests are only usefull to muddle the water, forgetting the actual
DFSG, to argue about hypothetical issues of questionable relevance.

> > What much more ? And what do you loose if upstream is allowed to use your code
> > in the main product, and thus everyone profits ? Again, only a code hoarder
> > would reject this kind of clause, and as thus get no sympathy from me.
> The QPL allows the original developer an all-permissive license,
> allowing them to put your code in their proprietary product.  Between
> that, and the patch clause designed to prevent forking, it seems to me
> like the real code hoarders are those releasing code under the QPL.

  "a non-exclusive royalty-free right is granted to the
  initial developer of the Software to distribute your
  modification in future versions of the Software provided such
  versions remain available under these terms in addition to any
  other license(s) of the initial developer."

You don't give a free blank check here, you only give the right for the code
to be integrated back in the original software, and in the case of dual
licencing of said software, like in both the ocaml and Qt case, you give the
right to have the _SAME_ software also relicenced under the other licence,
thus allowing upstream to not have to handle a split patch. In no way does
this allow upstream to use said modifications in the proprietary version, and
not in the QPLed one. Notice that this applies also for QPLed changes to Qt,
which could then be relicenced back under the GPL.


Sven Luther

Reply to: