Re: GPL link against non-free in original work (Re: Isn't a kde version..)

[Richard Braakman <dark@xs4all.nl>]

Peter S Galbraith wrote:
> 
> Richard Braakman wrote:
> 
> > That won't work.  The additional permission granted doesn't help us,
> > since we also distribute Qt itself. 
> 
> 1- read the rest of the email and you'll notice the further
> change to:

Sorry, elm only showed me the first part because it was confused
by the PGP headers.

I don't see the new wording changing this point, though.
If you wish to specify that the XForms library need not be considered
part of the source code of the program, then it's much better to
just say that and avoid the complex "major components" clause entirely.
I see many people misinterpreting that.

> 2- We're talking about XForms and not about Qt.

Sorry again, there was so much talk about Qt that I got confused :)

> 1- Specifying _which_ XForms was suggested by Brian (as was the
>    entire added paragraph), and I wished you had bought up this
>    _valid_ objection when we discussed this a few months back.

*shrug* I do what I can.  There are lots of things I would like
to have done a few months back.

If this is for contrib then it doesn't matter anyway.

> 2- For now, There are no modified versions of XForms anyway.  I
>    agree that this license would be incompatible with future
>    forks of xforms which were not Copyright (c) by T.C. Zhao and
>    Mark Overmars or GPL'ed, so if Zhao and Overmars ever release
>    XForms under anything else but GPL I'd be in trouble (if they
>    ever release it).  Perhaps replace the sentence with something
>    along the lines of linking with XForm or any of its
>    descendents is okay?

Perhaps XForms (etc etc), or any library compatible with it?
It all depends on what the intent is.  I must admit that I
cannot conceive of a reason to use the GPL at all if you're going
to link to a non-free library, so I have no idea what kind
of exception would be appropriate.

In any case, the license as-is is fine for contrib, I just think
it has a few flaws, the most notable being that it encourages
a common misintepretation of the "major components" clause.

Thanks,

Richard Braakman