Re: Bug#354216: upstream license patched in debian package

[Kel Modderman <kelrin@tpg.com.au>]

Eduard Bloch wrote:

> #include <hallo.h>
> * Kel Modderman [Fri, Feb 24 2006, 11:05:37PM]:
>  
>
> > Package: sl-modem-source
> > Version: 2.9.9d+e-pre2-2
> > |Severity: grave
> >
> > +sl-modem (2.9.9d+e-pre2-1) unstable; urgency=low
> > +
> > +  * New upstream pre-release (closes: #327588)
> > +  * added sv.po from Daniel Nylander (closes: #330436)
> > +  * slmodemd.1 manpage from Erik Schanze (closes: #335369)
> > +  * defining URB_ASYNC_UNLINK if not set, needed by recent kernels
> > +  * applied patch for class/simple_class transition from Gentoo lists with
> > +    some fixes (closes: #325799)
> > +  * changed the MODULE_LICENSE string to "Dual BSD/GPL" since the license
> > +    is actually a BSD license clone (closes: #327545)
> >    
> ...
>  
>
> > How can you justify this change?
> >    
>
> I though I have written that above.
>  

Can you please clarify at all? What makes a license "clone" an 
interchangeable license, especially since you are the one responsible 
for the actual license change, and not "Smart Link Ltd."? What do they 
think about it? Is it your lawful right to change it? Please write a 
brief reason for the change, but a bit more informative than the reason 
stated in debian/changelog (#327545 revealed nothing).

>  
>
> > I believe it is offensive to the linux kernel developers to circumvent 
> > the license exportation of the smartlink module in this manner.
> >    
>
> I believe something different. I believe it is quite offensive to the
> Linux users when some kernel developers deliberately replace interfaces
> with incompatible ones with strace access limitation. Looks similar to
> tactics known from patenting interfaces and file formats.
>  

Well, this will not be discussed further here, its been well thrashed by 
many others.

>  
>
> > drivers/amrlibs.o and modem/dsplibs.o are binary blobs. drivers/st7554.c 
> > (ST7554 USB Smart Link Soft Modem driver) depends on these binary blobs.
> >    
>
> First: only on one of them, amrlibs.o. The other file just have been
> forgoten in the driver source and will be removed, thanks for noticing.
>
> Second: Yes, this is a non-GPLed driver which is to be loaded with the
> Linux kernel. And?
>
>  
>
> > In recent linux kernel code, some advanced functions require a GPL 
> > compatible license, drivers/base/class.c:-
> >
> > |****EXPORT_SYMBOL_GPL(class_create_file);
> > EXPORT_SYMBOL_GPL(class_remove_file);
> >    
> ...
>  
>
> > The above mentioned modification of the Smartlink license circumvents 
> > this requirement.
> >    
>
> I know, no need to teach me. But what are you trying to say? Or are you
> just one of the anti-NON-GPL-drivers crusaders, trying to begin a new
> battle?
>  

Absolutely not. I am sorry you think I am taking the time to report what 
I honestly think to be a grave flaw because I want to bignote myself or 
prove a point. My primary and only concern is the integrity of the 
software debian users are provided. Is that o.k?

> Eduard.
>
>  
Thanks, Kel.