CC vs GPL: how to ensure compatibility and compliance

Some time ago we discussed CC vs GPL in this list. Rob Myers
convinced me
that there was no problem in software using CC-licensed
resources, as they
would not be linked.  If I understand it correctly, his
argument was along
the lines that a GPL videogame with CC graphics and music
was not linking
software, and could be considered "mere
aggregation", the same way GPL,
non-GPL and even non-free programs can be distributed in the
same physical
medium.

(Rob, please excuse me if I am paraphrasing you wrong, I
can't find your
message in the archive.)

This article deals with how to ensure compatibility and
compliance between
GPL and "freedomdefined-compliant" CC licenses.

http://www.linux.
com/feature/119212

I have only read the article cursorily, but I have already
found one glaring
mistake:
> The remaining licenses, CC-BY-SA and CC-BY, stick only
to elements that satisfy the free software definition. But
to be GPL-compatible, a license must also be a copyleft
license -- meaning the license ensures that once a work is
made available to the public, it cannot be taken away again.
CC-BY makes no such requirement, thus it is not copyleft and
is incompatible with the GPL.

(of course that's silly, many non-copyleft licenses such as
BSD are
GPL-compatible precisely by dint of being non-copyleft. It
is the OTHER
non-GPL copyleft licenses which are mostly non
GPL-compatible).

Maybe there are other issues that you the list may want to
raise regarding
Creative Commons licensing.

Regards,

Javier
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

On 9/29/07, Javier Candeira <javiercandeira.com> wrote:
> Some time ago we discussed CC vs GPL in this list. Rob
Myers convinced me
> that there was no problem in software using CC-licensed
resources, as they
> would not be linked.

Of course.. then you run into cases where people have
cc-nc-nd
licensed the localization files with all the text for an
application
in an attempt to unfree code they are obligated to leave
GPLed (I can
provide an example of exactly this if someone wants to see
it).

Even things like this are technically okay, they are
violations of the
spirit of the license and the likely intentions of the
licenses. As
such such actions should be avoided.. and may well be
precluded by
future editions of the GPL should practices like that
become
widespread.

Just something else to keep in mind related to this
subject.
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

On Saturday 29 September 2007 02:10 am, Javier Candeira
wrote:
> Some time ago we discussed CC vs GPL in this list. Rob
Myers convinced me
> that there was no problem in software using CC-licensed
resources, as they
> would not be linked.  If I understand it correctly, his
argument was along
> the lines that a GPL videogame with CC graphics and
music was not linking
> software, and could be considered "mere
aggregation", 

OK, so it might depend on the design of the system right?

If you had a GPL game engine that played content according
to a certain API or 
format. So the game consists of an executable engine say and
a data file say. 
And as long as the data file was in a certain format then
the engine would 
play it... Then the data file could non-GPL compatible.

BUT. If it was all one big executable...?

Consider the former may be along the lines of using the gimp
to process all 
rights reserved graphics or some of the Free video editing
software working 
with all rights reserved video...

> the same way GPL, 
> non-GPL and even non-free programs can be distributed
in the same physical
> medium.

>
> (Rob, please excuse me if I am paraphrasing you wrong,
I can't find your
> message in the archive.)
>
> This article deals with how to ensure compatibility and
compliance between
> GPL and "freedomdefined-compliant" CC
licenses.
>
> http://www.linux.
com/feature/119212
>
> I have only read the article cursorily, but I have
already found one
> glaring
>
> mistake:
> > The remaining licenses, CC-BY-SA and CC-BY, stick
only to elements that
> > satisfy the free software definition. But to be
GPL-compatible, a license
> > must also be a copyleft license -- meaning the
license ensures that once
> > a work is made available to the public, it cannot
be taken away again.
> > CC-BY makes no such requirement, thus it is not
copyleft and is
> > incompatible with the GPL.
>
> (of course that's silly, many non-copyleft licenses
such as BSD are
> GPL-compatible precisely by dint of being non-copyleft.
It is the OTHER
> non-GPL copyleft licenses which are mostly non
GPL-compatible).

Indeed, super wrong... but crazy statements like that seem
to crop up over and 
over in the area of Free works.
>
> Maybe there are other issues that you the list may want
to raise regarding
> Creative Commons licensing.
>
> Regards,
>
> Javier

all the best,

drew

_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

drew Roberts wrote:
> On Saturday 29 September 2007 02:10 am, Javier Candeira
wrote:
>> If I understand it correctly, his argument was
along
>> the lines that a GPL videogame with CC graphics and
music was not linking
>> software, and could be considered "mere
aggregation", 
> 
> OK, so it might depend on the design of the system
right?

That's what I was thinking. The article says if the path to
the resource is
harcoded, then the GPL is violated I don't know enough to
say "bullshit"
right away, but it does seem a bit arbitrary.

> If you had a GPL game engine that played content
according to a certain API or 
> format. So the game consists of an executable engine
say and a data file say. 
> And as long as the data file was in a certain format
then the engine would 
> play it... Then the data file could non-GPL
compatible.
> 
> BUT. If it was all one big executable...?

Yes, that's what I was thinking. Not an issue in a distro
like Debian, where
everything is nicely modularised, but a possible issue in
Windows games that
come packaged as one big fat .exe.

Maybe if you provided a directory tree from which the
executatble can be
built, including the GPLd code and the CC'd resources, the
GPL is followed
(you can build the executable, and the code is not linked to
any non-free
code)... I don't know, and that's why I wanted to discuss it
on the list.

The rest becoming slightly more OT:

>>> The remaining licenses, CC-BY-SA and CC-BY,
stick only to elements that
>>> satisfy the free software definition. But to be
GPL-compatible, a license
>>> must also be a copyleft license -- meaning the
license ensures that once
>>> a work is made available to the public, it
cannot be taken away again.
>>> CC-BY makes no such requirement, thus it is not
copyleft and is
>>> incompatible with the GPL.
>> (of course that's silly, many non-copyleft licenses
such as BSD are
>> GPL-compatible precisely by dint of being
non-copyleft. It is the OTHER
>> non-GPL copyleft licenses which are mostly non
GPL-compatible).
> 
> Indeed, super wrong... but crazy statements like that
seem to crop up over and 
> over in the area of Free works.

Yes. I teach "Introduction to Free Software" and
my students have a knack
for finding seemingly authoritative articles, written by
lawyers, professors
and other impressively credentialed folk, that clearly
explain how one
difference between Free Software and Open Source is that
Free Software has
to be copyleft but Open Source can be non-copyleft. Why my
students fail to
read the FSF's Free Software Definition (which we point to
in every way
possible) I don't know.

Javier
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

Gregory Maxwell wrote:
> Of course.. then you run into cases where people have
cc-nc-nd
> licensed the
[resources]
> 
> Even things like this are technically okay, they are
violations of the
> spirit of the license and the likely intentions of the
licenses. As
> such such actions should be avoided.. and may well be
precluded by
> future editions of the GPL should practices like that
become
> widespread.

I disagree strongly.

Many advocates of free software, including Richard Stallman,
have
expressed the opinion that the "ethical
compulsion" to make software
free is unique to computer programs (or perhaps more
generally to
"things of utilitarian rather than aesthetic
value", as I would
personally put it), and feel quite comfortable with the idea
of
conventional copyright-monopoly-based selling of aesthetic
content.

Some would argue that free content is "better",
but not so strongly
indicated as for programs.

And of course, there are some -- notably some loud voices in
the Debian
project -- who do share your assessment that free content is
an ethical
compulsion (or perhaps they only believe it is a
prerequisite to be in
Debian -- a weaker statement).

Now, I haven't done a scientific survey, but my gut feeling
based on
talking to software developers is that the *majority*
opinion among
*developers* (who are the people who determine what license
software
will be released under!) is the first one: that content
licensing is and
should be unaffected by the license of the software.

If this were not so, there are a number of alarming
consequences (which
have been mentioned before):

* You could not write free documents using Microsoft Word

* You could not write non-free documents using
OpenOffice.org Writer
(i.e. the act of using a free software word processor would
mean giving
up any copyright rights in documents you produce yourself)

* You could not compile a proprietary program using gcc.

* You couldn't compile a free software program with
Microsoft Visual C++

* Logos designed in Inkscape couldn't be trademarked or sold
under a
proprietary copyright license

and so on.

Very few people want these consequences.

Now it is true that game engine and game content boundaries
can become
blurred. In the fuzzy areas, it may well be better to stick
to matching
licenses. But in general, the "license of the data
processed is not
affected by the license of the data processor, nor vice
versa" rule is
awfully useful.

Note that the distinction here is not between types of data,
but rather
  the relationship between two works: For example, in the
computer
program compilation case, the proprietary program being
compiled by the
free compiler may itself be a compiler (exactly the same
type of work as
the work processing it). So this is not the same as the
"content is
dissimilar from code" argument; it's the "program
data licensing is not
affected by the license of the program". It has been
argued in the past
that such constraints would render the license
"non-free", since it
would "discriminate against users or fields of
endeavor".

The thing about delivering "free content" with
"free software" is it can
be more easily distributed -- aggregate-compatible licensing
is all that
is required. Assuming that Debian really does approve the
CC-By and
CC-By-SA licenses (which I really hope they will), then this
is an
acceptable combination to get a game (engine and content)
included into
the Debian distribution. This is a big boost to distributing
it, of course.

However, there are plenty of examples of
"shareware" or "non-free" games
which are based on free engines. Sometimes the developers
create a free
engine with the specific intent of basing non-free games on
it (thus
defraying engine development and testing costs as well as
training a
pool of potential future game developer/designers). I think
most
developers consider this a good situation, though of course,
most want
free games to also exist.

Cheers,
Terry

-- 
Terry Hancock (hancockAnansiSpaceworks.com)
Anansi Spaceworks http://www.AnansiSpac
eworks.com

_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

On Sunday 30 September 2007 10:45 pm, Terry Hancock wrote:
> Gregory Maxwell wrote:
> > Of course.. then you run into cases where people
have cc-nc-nd
> > licensed the
>
> [resources]
>
> > Even things like this are technically okay, they
are violations of the
> > spirit of the license and the likely intentions of
the licenses. As
> > such such actions should be avoided.. and may well
be precluded by
> > future editions of the GPL should practices like
that become
> > widespread.
>
> I disagree strongly.
>
> Many advocates of free software, including Richard
Stallman, have
> expressed the opinion that the "ethical
compulsion" to make software
> free is unique to computer programs (or perhaps more
generally to
> "things of utilitarian rather than aesthetic
value", as I would
> personally put it), and feel quite comfortable with the
idea of
> conventional copyright-monopoly-based selling of
aesthetic content.

I am not so sure about this Terry. In my reading, I have
always come away with 
the impression that it is more like... I know this is so for
code. That is my 
area. I haven't thought deeply enough about other areas to
know if it so 
there so I will refrain from comment onthose other areas.

snip.

>
> Cheers,
> Terry

all the best,

drew
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

<quote who="Terry Hancock" date="Sun, Sep
30, 2007 at 09:45:16PM -0500">
> Many advocates of free software, including Richard
Stallman, have
> expressed the opinion that the "ethical
compulsion" to make software
> free is unique to computer programs (or perhaps more
generally to
> "things of utilitarian rather than aesthetic
value", as I would
> personally put it), and feel quite comfortable with the
idea of
> conventional copyright-monopoly-based selling of
aesthetic content.

And many advocates -- like myself -- disagree. Of course, we
both agree
that the FSF's job is not to get into the middle of this
debate except
where it pertains to documentation for software and computer
programs
that we we agree need to be free.

> Now, I haven't done a scientific survey, but my gut
feeling based on
> talking to software developers is that the *majority*
opinion among
> *developers* (who are the people who determine what
license software
> will be released under!) is the first one: that content
licensing is and
> should be unaffected by the license of the software.

I think that's right. While I think that content should be
free, I
don't believe that free software should be engineered or
licensed so
that it can only produce free content. For many of the
reasons you
describe.

This question as I understand it it is different though, and
quite
specific. It is asking about whether GPL linking provisions
extend to
things like images distributed alongside a program. I can't
imagine that
this could he case but licensinggnu.org is the proper
authority on
these issues and I suggest anyone mail that address sf you
would like a
full answer on this or other GPL questions.

Regards,
Mako


-- 
Benjamin Mako Hill
makoatdot.cc
http://mako.cc/

Creativity can be a social contribution, but only in so
far as society is free to use the results. --RMS

_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

On 9/30/07, Terry Hancock <hancockanansispaceworks.com>
wrote:
> > Even things like this are technically okay, they
are violations of the
> > spirit of the license and the likely intentions of
the licenses. As
> > such such actions should be avoided.. and may well
be precluded by
> > future editions of the GPL should practices like
that become
> > widespread.
>
> I disagree strongly.
>
> Many advocates of free software, including Richard
Stallman, have
> expressed the opinion that the "ethical
compulsion" to make software
[...]

I was specifically thinking of a particular case where
someone took a
GPLed application then made an enhanced version which they
wanted to
distribute under terms more restrictive than the GPL.  This
isn't
permitted by the license because the spirit and presumably
the
original author intends and the language of the license is
designed to
preserve the freedom of the software.

In the case I was thinking of the author of this enhanced
version has
attempted to skirt that requirements of the GPL by placing
critical
program data (such as all the UI text) as well as a lot of
secondary
data (documentation) under cc-by-nc-nd and claiming that
removing it
and replacing it would be a violation of no-derivs.

I was only interested in pointing out that the subject can
be complex
by trying to point out a case where the content can be
easily be
argued to be an essential and inseparable part of the
software, and
where incompatible licenses have been used in a hostile
attempt to
'unfree' the software.

Perhaps you should have spent a couple seconds making sure
you
understood me before going off on a rant.

To me it looks like you just wasted a lot of time attacking
a straw
man position which may not be held by anyone sane at all.

You seem to think that you have some opponents that hold
some rather
extreme beliefs. I've often found that kind of situation to
be the
result of poor communication. I'd suggest you hash out your
differences with whomever you're arguing with, if I knew who
they
were... as far as I can tell they aren't here.
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

On Tuesday 02 October 2007 12:40 am, Gregory Maxwell wrote:
> On 9/30/07, Terry Hancock <hancockanansispaceworks.com> wrote:
> > > Even things like this are technically okay,
they are violations of the
> > > spirit of the license and the likely
intentions of the licenses. As
> > > such such actions should be avoided.. and may
well be precluded by
> > > future editions of the GPL should practices
like that become
> > > widespread.
> >
> > I disagree strongly.
> >
> > Many advocates of free software, including Richard
Stallman, have
> > expressed the opinion that the "ethical
compulsion" to make software
>
> [...]
>
> I was specifically thinking of a particular case where
someone took a
> GPLed application then made an enhanced version which
they wanted to
> distribute under terms more restrictive than the GPL. 
This isn't
> permitted by the license because the spirit and
presumably the
> original author intends and the language of the license
is designed to
> preserve the freedom of the software.
>
> In the case I was thinking of the author of this
enhanced version has
> attempted to skirt that requirements of the GPL by
placing critical
> program data (such as all the UI text) as well as a lot
of secondary
> data (documentation) under cc-by-nc-nd and claiming
that removing it
> and replacing it would be a violation of no-derivs.

It seems a person making this claim to bolster their case
might actually be 
making your GPL violation case for you.

>
> I was only interested in pointing out that the subject
can be complex
> by trying to point out a case where the content can be
easily be
> argued to be an essential and inseparable part of the
software, and
> where incompatible licenses have been used in a hostile
attempt to
> 'unfree' the software.

all the best,

drew
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses

On 10/2/07, Terry Hancock <hancockanansispaceworks.com>
wrote:
> Rant?
>
> Your overreaction to my post is a "rant". I
simply presented a
> contrasting opinion. You said yourself that such
practices were
> legally permissible.

I inadvertently omitted an "if" in my statement. I
can see how my
error might confuse you. I intended to write "Even if
things like this
are technically okay".

I ascribed motivations to the actions in my example.
"in an attempt to
unfree code". Now, you often can't objectively measure
motivations but
when you can it often makes things much easier.

In the case I was referring to the software was inoperable
without the
cc-nc-nd data, and the redistributor specifically stated
that their
intention was to be able to charge people who make
commercial use of
the application. And if you believed their licensing
argument they had
succeeded: the software is inoperable without the cc-nc-nd
part.

> But then you claimed that it violated the
> spirit of the license.

The GPL has a preamble for a reason. The intention of the
license is
abundantly clear.

You may be able to find a technical loophole in the
permissions
granted in the GPL which allows you to claim that you can
combine the
GPLed covered work with a differently licensed work in a
manner which
removes the resulting derivative from the pool of freely
licensed
works, but loopholes like this will be closed.

I was not attempting to comment about cases where your
actions don't
result in the removal of a new version of the software from
the pool
of freely licensed works because I didn't have anything to
say about
that.  I was simply reminding people not to play license
games to
escape the requirements of the GPL, and I'd only commented
at all
because I had just run into to such a case.
_______________________________________________
cc-licenses mailing list
cc-licenseslists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/cc-licenses