Spotting GPL violations

[Philip Webster]

Jeroen Dekkers wrote:
> At Sat, 11 Dec 2004 16:07:19 +0000,
> Philip Webster wrote:
> 
>>Jeroen Dekkers wrote:
>>
>>
>>>At Sat, 11 Dec 2004 13:21:59 +0000,
>>>Philip Webster wrote:
>>>
>>>
>>>>Hi,
>>>>A piece of software has come to my attention that may be violating the 
>>>>GPL. Details are as follows:
>>>>- The software is an interface (GUI, intermediary routines) between a 
>>>>proprietary operating system and the Free printing system GIMP-Print.
>>>>- The company distributing the software is not offering source code, and 
>>>>has not licensed the derivative work under the GPL.
>>>
>>>The question is whether this is really a derivative work. "An
>>>interface" doesn't really say anything, it can be a library interface
>>>or it could talk over some protocol using sockets.
>>
>>It appears to be a library interface.
> 
> 
> I can't really see enough out of the descriptions. The problem is that
> I don't know anything about GIMP-Print or RiscOS. I think the only way
> to be sure whether it's a GPL violation is to download the code and
> look at it.
I was under the impression that any code which links directly to a GPL'd 
library is essentially using part of a GPL-licensed work in the form of 
the method names it is calling, and as such by directly linking, becomes 
in itself a 'derivative work' which must be GPL licensed in order to be 
legally distributed.

Is that right? That's why the LGPL was created, wasn't it? So people 
could use Free Software libraries, without having to take a full leap 
into the Free Software community?

Phil