Am Dienstag, 24. Juni 2014, 15:25:29 schrieb Alexander Wagner:
> On 06/24/2014 01:44 PM, Tobias Ellinghaus wrote:
> 
> Hi!
> 
> [...]
> 
> >> If dt is GPL, then the artwork, IMHO being part of dts /code/, has to be
> >> GPL as well. And then GPL applies. In principle, no need to track down
> >> people, and more specifically no need to ask. That's why you specify the
> >> license in the first place: you /don't want/ to be asked every time.
> > 
> > That would only be true if there was a file that explicitly said that
> > everything in the repository was GPL – and even then I am not completely
> > sure. There is a reason why all the source files state that they are
> > under GPL in the header. Besides, it is quite common for artwork, logos
> > or other accompanying data to be under a different license as the program
> > itself. See for example our usermanual.
> 
> You can do this of course but they still need to be compatible.

That is not correct, you can have completely proprietary files in the same 
repository as GPL licensed code. If you think otherwise I would like to see a 
rule that prohibits that. See [0].

> BTW: this makes for a nice argument for metadata within images ;) Adding
> a dc:rights would be a good idea.

SVG already supports that.

> [...]
> 
> > True, for example we have a few parts that were borrowed from others and
> > that are under a BSD license with our changes we put on top being GPL
> > which makes the whole thing GPL. But that still only affects our source
> > code. Graphics like SVG files are not linked into the program and not
> > compiled in.
> 
> I'm not sure that linking and compiling is necessary here to make
> something part of a source. Just think about interpreted languages.
> You'd surely not state that python e.g. is not code and that you
> couldn't license something like Calibre as GPL.

It's the other way round: Even if python is free software you are allowed to 
use it to run non-free scripts. Just by opening files in a GPL program doesn't 
impose ANY restrictions on the opened files.

> I'd argue on that line for the XML which makes the svg. In a way it is a
> bit like "programming" in LabView.

SVG is more like HTML, a description language. You can't program in it, but 
you can embed JavaScript. But that is not the point. It's an external file that 
is not linked into the program. The GPL is quite specific about these things.

> > Note that we already made the implicit assumption that our logo IS in fact
> > under a license compatible with GPL because at one place we DO compile it
> > into the program. But that isn't canonical truth, it's just that we did
> > it that way.
> 
> It seems sensible and if there're some doubts it should be changed to
> something GPLish. Otherwise at some point in time it might get the same
> trouble as Firefox has on Debian.

Actually I think that having a restrictive license on your logo is a good 
thing. You don't want to have derivative work to have your name, and you also 
don't want it to use your logo. But that is just my personal view. With 
regards to darktable I assume that the logo is under some license that is 
compatible with GPL.

> >> As an example from the CC world, which is a bit easier due to the
> >> "steps" they use (but useless for code): CC-BY is compabible to CC-BY-SA
> >> but not vice versa as the latter imposes  a/the sensible limit to the
> >> reuse.
> >> 
> >> Probably, the only point one might discuss, is, wether a logo is code in
> >> the framework of GPL. I'd argue that SVG (icon of dt) is code even more
> >> as it is included in dt's git and used for buidling the package. You use
> >> it within your code in the upper left of the main window not only for
> >> the desktop object. In my book this is code, thus GPL applies, it is
> >> ignorant of the language of code.
> > 
> > This would only be true for files compiled into the program, linked to the
> > program or similar things.
> 
> I'm not sure that compilation or linking is something necessary for
> "code". In fact I'm pretty sure those are no valid criteria. I'm even
> not sure that you could explain to most of the juristic people what
> compiling and linking is. Especially not in their terminology of
> concepts like "copy" and so on.

These are restrictions imposed by the GPL, not me. [1] would be a good 
starting point to read.

> > Files loaded and displayed during runtime are not
> > affected. Otherwise Firefox could only show open source websites.
> 
> I really see a difference between displaying some content and using
> artwork to make up the GUI of a program. I'm not sure how to draw the
> line here precisely but linking and compiling is IMHO to limited.
> Thinking of theme sets e.g.

Good example. Let's look at GIMP which is under GPL. You can load non-free 
themes in GIMP, however the default theme, which is compiled in, has to be 
GPL'd.

> >> It seems, that Wikipedia shares this view: at least they use the DT logo
> >> on the Wikipedia pages about dt. They mention Klaus Staedtler as author
> >> and Wikipedia requires you to use at least something like CC-BY-SA. So
> >> if they don't remove it as being GPL I'd argue that every photographer
> >> might use it as well as the NC stanza is not assumed by them to apply.
> >> (No, Wikipedia is not perfect, but usually they are a bit picky when it
> >> comes to licenses.)
> > 
> > That license statement only says that the file shown there is under
> > CC-BY-SA. It doesn't have to be the very same file we have in git, it
> > might also be a re- draw (which I don't think it is, just saying).
> 
> It is actually drawn from the git. Probably Wikipedia is wrong to do so.
> Maybe they assumed, as I do, that if the program is GPL, its artwork has
> to be as well.

You are contradicting yourself, do they assume GPL or CC-BY-SA?

> >> A bit difficult IMHO will be the images and texts on the dt webpages.
> >> Those are licensed CC-BY-NC-SA. In the context of professional
> >> photography lawyers may argue that using them in this context
> >> constitutes a "commercial" use. Note that this would have to be dts (or
> >> the authors) lawyers. ;)
> > 
> > I don't get this. The authors release their work under some license, they
> > themselves can do whatever they want with it.
> 
> Sure. The authors can do what they want and use whatever license they
> want. Especially in Germany you can't even get rid of your copyright.

I know. :)

> > Even show NC stuff on a commercial site. man dual licensing.
> 
> If the content is not used against the NC part. This usually gets very
> messy. E.g. you can not use some image licensed NC in a commercial
> journal. At least not safely. So all big publishers will refuse to use
> it. This is also a point why, if at all, they like CCBYNC for their
> "OpenAccess" stuff: it's nearly not reusable. That's why many people
> argue against NC as being not OpenAccess. (Access is here defined by the
> Berlin declaration and includes reuse, so they have a very broad view of
> "access".)

The author if a work can't even violate the NC part! No matter how hard he 
tries. It's a restriction HE imposes on OTHERS. He himself is not bound by the 
NC.

> >> In my experience as soon as lawyers see the NC they get very creative
> 
> [...]
> 
> >> considered valid...
> > 
> > (This is getting a little OT, but I personally agree with lawyers saying
> > that a beneficiary concert is still a commercial event.)
> 
> Depends, and this is the main issue with NC: how do you define
> "commercial". As this seems not to be clear (we have a perfect example
> here, as we two disagree completely on its meaning ;) the CC guys
> strongly discourage NC these days, arguing that CC-BY-SA results at the
> same effect, but is safe from a juristic point of view.

"Commercial" is quite a clear legal term I guess. IANAL, but AFAIK it doesn't 
require "wants to make money". But that is way OT and I am not an expert on 
this.

> BTW: In my example, I was not refering to the concert. You could even do
> that part without any money, say musicans working for free, no tickets
> sold, public location and so on. The argument they make is, that the
> selling of beer in presence of NC music violates CCBY_NC_ of the music.
> Similarly, many private radio stations do not play CCBYNC music as
> they're not sure they're allowed to do it. They earn their living form
> advertisments listned to cause of the music they play, right? Thats why
> recently something like GEMA (urgh!) for free music was established...

Prohibiting that kind of use is exactly why people use NC. Even if a radio 
station calls itself "private" doesn't make it less commercial.

> I fear I'd have to give you this argument in German as my English is not
> sufficient for this.

Since this part of the discussion isn't important for the topic at hand we 
should just ignore it for now.

> >> Anyway, considering the NC here and switch to a -SA alone might solve
> >> the issue, even without harming the purpose. Point is the very small
> >> area where the NC actually prevents a commercial use. cf.
> >> http://creativecommons.org/freeworks
> > 
> > It would be next to impossible to drop the NC part of the website as
> > everyone ever having edited it would have to agree.
> 
> Yepp. I see this problem. I just wanted to suggest to think this over
> for the future.

I don't see a problem.

> > And I don't see any problem with having it NC.
> 
> The point is if you use NC in a commerical environment it can get quite
> messy. Usually, you'd advise to drop NC if you feel that such
> environments shouldn't be afraid to use something.
> 
> So for the OPs initial question it would probably be perfect to provide
> some piece of HTML to embed in webpages that uses content that is on a
> sufficiently free license like CCBY or even better CC0, so they are
> safe: "yes, I can use this, the people of the project suggest to use it
> that way" and so on. Especially for promotion the hurdle should be very
> low and it should be safe for your promoters.

Now you are saying that the logo is under a NC license? That's the 3rd license 
you said it had in this mail.

> > If someone with commercial interests wants to use parts of our
> > website he can just contact us/the authors and ask for permission.
> 
> The very idea of a license in the first place is that you do not need to
> ask at all. Otherwise, just set "all rights reserved" and everyone can
> ask. But you surely don't want that. Besides I believe this is a model
> of the (hopefully distant) past.

I hope you understand that there is a difference between private people using 
our work and commercial entities? If someone wants to make money from my work 
[2] I would at least want to be asked.

> > PS: All I wanted to point out in my original mail was that while it is
> > probably no problem at all we might want to clarify the legal state of our
> > assets like the logo.
> 
> This is a good idea of course. I'd also believe nobody who contributed
> these works has any problem with that.
> 
> Still, I'd argue that artwork being part of dts GUI is code and thus has
> to be GPL (or compatible) to avoid trouble later on. I'd also suggest to
> consider this line of reasoning for future additons. Thinking also about
> inclusion in those really free distributions like Debian. Its good for
> the project to have this safe.

You can argue like that, it's still not true. Please read the GPL FAQ.

Tobias


[0] https://www.gnu.org/licenses/gpl-faq.html#MereAggregation
[1] https://www.gnu.org/licenses/gpl-faq.html#IfInterpreterIsGPL
[2] Don't confuse that with the right to sell GPL software. That is prefectly 
ok and not forbidden at all. We are currently talking about our blog posts and 
suchlike.

Attachment: signature.asc
Description: This is a digitally signed message part.

------------------------------------------------------------------------------
Open source business process management suite built on Java and Eclipse
Turn processes into business applications with Bonita BPM Community Edition
Quickly connect people, data, and systems into organized workflows
Winner of BOSSIE, CODIE, OW2 and Gartner awards
http://p.sf.net/sfu/Bonitasoft
_______________________________________________
Darktable-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-users

Reply via email to