[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Do I have to release the patch for a GPL software under GPL?

From: Gordon Burditt
Subject: Re: Do I have to release the patch for a GPL software under GPL?
Date: Mon, 15 May 2006 06:01:23 -0000

>   Now, I, Evil Bill Fence Door, copyright this patch, sell it with
>   onerous copy protection, and for $1,000,000 a copy.  The license
>   that comes with it prohibits re-distribution of the patch.  Note
>   that I'm *not* re-distributing any GPL-licensed software.
>But you _modified_ a GPL licensed work (section 2 of the GNU GPL), and
>now are distributing the modifications to this work.  It is completely
>irrelevant what the form of the patch is, your patch does not work
>without the GPLed work, and cannot be used without it so it is a
>deriviate work.  

Copyright law defines no such thing as a "deriviate work".
Copyright law and the GPL define a "derivative work", and
this definition has nothing to do with your definition of it.

The GPL Version 2 contains:

>The "Program", below,
>refers to any such program or work, and a "work based on the Program"
>means either the Program or any derivative work under copyright law:
>that is to say, a work containing the Program or a portion of it,
>either verbatim or with modifications and/or translated into another
>language.  (Hereinafter, translation is included without limitation in
>the term "modification".)  Each licensee is addressed as "you".

A derivative work is a work *CONTAINING THE PROGRAM OR A PORTION OF IT*.
Now, most *real* patches, especially context diffs, contain a portion
of the program to be patched.  But you can come up with patches
that don't.  "diff -e" does a pretty good job if the lines added are
completely new and not modified lines from the original program.

Incidentally, you can apply a diff -e patch to any file with a sufficient
number of lines.  I didn't say the result would be useful, but the
GPL definition of "derivative work" says nothing about it being useful.

                                                Gordon L. Burditt

reply via email to

[Prev in Thread] Current Thread [Next in Thread]