«

Luke Skywalker <> wrote:
>
> On Mon, 07 Feb 2005 17:47:30 -0800, Robert Kern <>
> wrote:
> >Now, that's not to say that they are correct in their interpretation
> of
> >the GPL's terms. In fact, if I had to bet on an outcome, I'd probably
> >wager that the court would hold that only static linking would force
> the
> >program as a whole to follow the GPL terms. However, I certainly don't
>
> >have the money to pony up to run a test case. Consequently, I try to
> >follow the wishes of the copyright holder.
>
> It's strange that something so central hasn't been clarified yet, but
> maybe it's part of the changes meant for V.3.
>
> When you think about it, it'd be like banning any closed-source apps
> from being developed for Linux, since any application makes syscalls
> to the kernel and its libraries.
>
> But the fact is that there are now closed-source apps for that
> platform, and are considered legit since those apps don't include code
> from the kernel, but instead, merely make calls to binary objects. I
> fail to see the difference between making calls to the kernel API and
> making calls to Qt libraries.

The COPYING file for the Linux kernel includes this note:

Linux main COPYING:

: NOTE! This copyright does *not* cover user programs that use kernel
: services by normal system calls - this is merely considered normal use
: of the kernel, and does *not* fall under the heading of "derived work".
: Also note that the GPL below is copyrighted by the Free Software
: Foundation, but the instance of code that it refers to (the Linux
: kernel) is copyrighted by me and others who actually wrote it.
:
: Also note that the only valid version of the GPL as far as the kernel
: is concerned is _this_ particular version of the license (ie v2, not
: v2.2 or v3.x or whatever), unless explicitly otherwise stated.
:
: Linus Torvalds

Tim C

>
> Luke.
> --
> http://mail.python.org/mailman/listinfo/python-list

On Tue, 08 Feb 2005 13:24:35 +1100, Tim Churches
<> wrote:
>: NOTE! This copyright does *not* cover user programs that use kernel
>: services by normal system calls - this is merely considered normal use
>: of the kernel, and does *not* fall under the heading of "derived work".

OK, so according to Linus, the GPL allows a proprietary program to
make calls to the kernel, but TrollTech says the GPL doesn't allow a
proprietary program to make calls to the Qt library.

It's this double-standard that I find confusing, since both projects
are said to be based on the same license. I wouldn't have any problem
if Qt had built its own GPL-derived, custom license, but they claim
it's the same ol' GPL. Hence the questioning.

Luke.

>OK, so according to Linus, the GPL allows ....

No. Pay attention. Linus has his own revised version, to clarify
this point, and in fact /overruling/ the GPL if the point is
clarified differently by RMS or others.

That's the right of their community, it's /their/ code.

>make calls to the kernel, but TrollTech says the GPL doesn't allow a
>proprietary program to make calls to the Qt library.

That's their prerogative, although TrollTech's authority as an
/interpretational/ entity over the GPL means precisely zero. I
wouldn't push this, though, unless you've got a big litigation
budget.

>It's this double-standard that I find confusing, since both projects
>are said to be based on the same license.

Linus doesn't use "the" GPL, he uses "his" GPL, version-whatever.

Anyway, your safe bet:

Follow the copyright holder's wishes.

That's fair. After all, it's free, so they're doing you a damn
big favor.

C//

On Tue, 08 Feb 2005 03:49:40 +0100, Luke Skywalker
<> declaimed the following in comp.lang.python:

>
> OK, so according to Linus, the GPL allows a proprietary program to
> make calls to the kernel, but TrollTech says the GPL doesn't allow a
> proprietary program to make calls to the Qt library.
>
An off-the-cuff comment here...

One thing I see is that removing "GPL kernel 'library'" would
disable the very OS you are attempting to run on... With no OS, your
application concerns are moot. Anyone running the application has to
already have obtained the OS.

OTOH, while removing the QT library may prevent /your/
/application/ from running, it wouldn't kill the OS. It would be a
hassle to code, but if your application could dynamically select from
whatever toolkit is available on the machine, you (and I should emphasis
that this is an impersonal/generic "you" I reference) might be able to
argue an exemption from the QT license.

--
> ================================================== ============ <
> | Wulfraed Dennis Lee Bieber KD6MOG <
> | Bestiaria Support Staff <
> ================================================== ============ <
> Home Page: <http://www.dm.net/~wulfraed/> <
> Overflow Page: <http://wlfraed.home.netcom.com/> <

On Mon, 07 Feb 2005 20:56:44 -0800, Courageous <> wrote:
> Follow the copyright holder's wishes.
>
> That's fair. After all, it's free, so they're doing you a damn
> big favor.

I'm terrible sorry to extend this topic even furter, but it's silly,
not to say dull to think like that.

in the first place, it's not that they're doing charity. they have
plans. if they release Qt under GPL for non-comercial use, it's
because they want to increase the user base and so be able to earn
money with the comercial fee later.

And if i'm going to write software for it, when there's plenty of
alternative that actualy works flawless under windows, why should i
stick with an option that i don't even know to wich extends i can use
the damn thing? What it they revoke this license? what it windows
longhorn has a non-backwardcompatible GDI API and a newer version of
Qt must be used, and that newer version does not have a gpl version?

If i'm going to commit to something, i like to know the lengths the
other side gona commit also.

What you said was like "Hey! it's free food! who cares if it's
rotten?" sorry, but it's just too homer simpson for me.

Gabriel B. wrote:
> What it they revoke this license [on Qt]?

They can't. It's the GPL.

> what it windows
> longhorn has a non-backwardcompatible GDI API and a newer version of
> Qt must be used, and that newer version does not have a gpl version?

What if Wx does that? What if Tk does? What if GTK does?

> If i'm going to commit to something, i like to know the lengths the
> other side gona commit also.

They have: they're licensing Qt to you under an _irrevocable_ license,
the GPL. If that's not a commitment, nothing is.

On Mon, 07 Feb 2005 20:56:44 -0800, Courageous <>
wrote:

>
>>OK, so according to Linus, the GPL allows ....
>
>No. Pay attention. Linus has his own revised version, to clarify
>this point, and in fact /overruling/ the GPL if the point is
>clarified differently by RMS or others.
>
>That's the right of their community, it's /their/ code.
>
>>make calls to the kernel, but TrollTech says the GPL doesn't allow a
>>proprietary program to make calls to the Qt library.
>
>That's their prerogative, although TrollTech's authority as an
>/interpretational/ entity over the GPL means precisely zero. I
>wouldn't push this, though, unless you've got a big litigation
>budget.
It should also be pointed out that the FSF's interpretation of the GPL
with respect to Qt means absolutely zero. If TrollTech publishes an
interpretation of the GPL and announces it to any interested in
licensing their software, I suspect that the courts will take that
into consideration. This won't help that at all if it isn't a legally
valid interpretation, but it establishes that you *knew* what their
interpretation was when you agreed to the terms to distribute their
copyrighted software.

>
>>It's this double-standard that I find confusing, since both projects
>>are said to be based on the same license.
>
>Linus doesn't use "the" GPL, he uses "his" GPL, version-whatever.
>
>Anyway, your safe bet:
>
>Follow the copyright holder's wishes.
>
>That's fair. After all, it's free, so they're doing you a damn
>big favor.
>
>C//
Scott Robinson

>It should also be pointed out that the FSF's interpretation of the GPL
>with respect to Qt means absolutely zero.

Indeed. It would be the court that would have to decide what the
language of the GPL means, given the substantial body of case
law as the court sees it.


>... but it establishes that you *knew* what their interpretation ...

But it doesn't. They'd really need to put it into their license
expressly.

Anyway, we digress, and are in agreement.

C//

Luke Skywalker wrote:
>
> OK, so according to Linus, the GPL allows a proprietary program to
> make calls to the kernel,

As I understand things, it's not the GPL which allows
this, it's Linus himself who allows it. If Linus
hadn't explicitly said that, the GPL might be interpreted
as disallowing it.

--
Greg Ewing, Computer Science Dept,
University of Canterbury,
Christchurch, New Zealand
http://www.cosc.canterbury.ac.nz/~greg