Interesting People mailing list archives

IP: more on GNU license controversy


From: David Farber <dave () farber net>
Date: Tue, 08 May 2001 09:57:01 -0400




Date: Tue, 08 May 2001 09:33:31 -0400
To: Dave Farber <dave () farber net>
From: Manny Farber <manny () manny com>
Subject: Re: IP: GNU license controversy

Here's it is with attribution interspersed.


Manny Farber wrote:

I think I can use similar logic to argue that the "proprietary"
work is also subject to GPL.

Immediately prior to compilation, the modified source code is obviously
a "derived work" and since the GPL is recursive (under 2b below,
any "derived work" falls under GPL), then I could also freely
copy and distribute the modified source code (with or without binaries).

Also see: "the distribution of the whole must be on the terms of this
License, whose permissions
for other licensees extend to the entire whole, and thus to each and 
every
part regardless of
who wrote it."

Brad Templeton wrote:

Indeed, but note the first sentence of the message I wrote.  Humans obey
licences, not programs.  The prorietary program is modifying the GPL'd
source and generating object code.  It never leaves the source around for
you to distribute, though you might hack it to try to extract that source,
and claim it is freely distributable.

Manny Farber wrote:

By saying that you'd have to "hack it to extract the source," it seems
you're implying that the patching problem includes a full-fledged, 
integrated
compiler, presumably one that does not read source files from disk
(otherwise it would be easy to obtain the source files).

So I think you also need to be able to enforce a "no hacking/reverse 
engineering"
provision on your patch.

Brad Templeton wrote:

However, the legal point is that the person who modified the original 
source
is the _user_, not the author of the patching program.  The patching 
program
is a stand alone system not under the auspices of the GPL.

Manny Farber wrote:

Well I thought the whole point was to charge for software that uses GPL 
source.

But I don't think you can make a recipe (patching program)
that calls for two pounds of chicken (GNU source) and then claim that the
resulting entree (the binary from the modified code) is not derived from 
chicken.
The recipe proves that it is.

Why would anyone pay for the recipe when they can just ftp the entree
and eat it for free [assuming someone has "hacked" it to obtain source]? 
And knowing they've circumvented people trying to circumvent
GPL would just make it taste better. Please pass the salt.

Brad Templeton wrote:

The GPL says that anybody who wants to modify and distribute GPLd code must
also GPL the result and provide source.  The author of the patching program
is not modifying and distributng GPL code.  He is only writing a program
which does this, which other people run.

If the other people, the users who run the modifying program, were to
then distribute the results, they would be bound by the GPL, bound to
distribute the source that they don't have.   So they can't distribute
it.  But they don't want to, they just want to run it.




For archives see: http://www.interesting-people.org/


Current thread: