Implications of a Dependency on a GPLed Package

Christian Sigg <christian <at> sigg-iten.ch> writes:

Dear Marc

2. Can non-GPL compatible packages for R even be created (even if "pure R"), 

based upon the interpretation

of the GPL that Christian has postulated?

I am not trying to offer my own interpretation of the GPL, which is entirely 

irrelevant. (If anyone is

interested, I fall on the "pragmatic" side of the spectrum). I tried to give 

an accurate rendition of the

interpretation of the FSF as stated in the GPL FAQ, which is relevant for the 

R project. Even though the

legal basis of their reasoning seems unclear, to me, the argumentation as 

given in the GPL FAQ is quite

clear and can be summarised as:

1. In an interpreted environment, using library/module/package functionality 

in a program implies

dynamic linking of the library and the program.

2. Because of the dynamic linking, the library and the program effectively 

form a single program.

3. Therefore, if the library is released under the GPL (without any linking 

exception), the program has to

be released under a GPL-compatible license.

Of course, others can disagree with my understanding the argument given in the 

GPL FAQ, and I might not have

stated the FSF position correctly. 

If others read the same sections of the GPL FAQ and come to a different 

conclusion w.r.t. the FSF

interpretation of the GPL, I would be interested to hear their reasoning, 

unless this is considered

off-topic for R-devel.

Best regards,
Christian

Ok here's a possible different conclusion and reasoning. Here's the 1st 
paragraph of the FAQ, in full :

    http://www.gnu.org/licenses/gpl-faq.html#IfInterpreterIsGPL

   " When the interpreter just interprets a language, the answer is no. The 
interpreted program, to the interpreter, is just data; a free software license 
like the GPL, based on copyright law, cannot limit what data you use the 
interpreter on. You can run it on any data (interpreted program), any way you 
like, and there are no requirements about licensing that data to anyone. "

Therefore, R only packages (no C code) are just data. The GPL doesn't apply to 
them. R only packages depending on other R only packages are just data depending 
on data.

You appear to have been concentrating on the 2nd paragraph onwards of that FAQ. 
The 2nd paragraph starts :

   " However, when the interpreter is extended to provide ?bindings? to other 
facilities... "

There, "bindings" could be seen as related to "linking" at C level. Especially 
since later in the same paragraph the distinctions "statically" and 
"dynamically" are made. Therefore that FAQ's 1st paragraph could be interpreted 
as the only relevant one from that FAQ, for R only packages.

The view that R code is just data is quite widespread and agreed upon, iiuc.

Does that help?

Matthew