Re: GPL traitor !

[Tim Smith]

In article <slrnh01dri.bb5.jedi@nomad.mishnet>,
 JEDIDIAH <jedi@nomad.mishnet> wrote:
>     Not really. The only real question is what constitutes a "derivative 
>     work".
> 
>     There are certain people that have an interest in "misunderstanding 
>     this".

Such as the FSF?  A pure copyright license, such as what they claim the 
GPL is, cannot protect against all of the things they want to protect 
against, and so they make claims that aren't backed by copyright law as
to what makes a work subject to GPL.

For instance, suppose you have a GPL library, X, with a publicly 
documented interface.  No other library implements that interface.  They 
say that if you write code that can use X, your code must be under GPL.

However, suppose there is also a library, Y, that provides the same 
interface as X, and Y is under, say, a BSD license.  Now, according to 
the FSF, you can write code that calls X without it being a derivative 
work.

Copyright law simply doesn't work that way.  The existence or 
non-existence of Y is completely irrelevant to the question "is your 
code a derivative work of X?".  Y is relevant when it comes to trying to 
prove the answer, but it does not change what the answer is (for 
example, if you are accused of making a derivative work of X, and Y 
exists, then you can simply claim you copied from Y, not X, and the 
authors of X will probably have a hard time proving it was really X you 
copied from).

-- 
--Tim Smith