linux-kernel - RE: GPL vs non-GPL device drivers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 19 Feb 2007 17:29:02 -0800
From:	"David Schwartz" <davids@...master.com>
To:	<trent.waddington@...il.com>,
	"Michael K. Edwards" <medwards.linux@...il.com>
Cc:	"Linux-Kernel@...r. Kernel. Org" <linux-kernel@...r.kernel.org>,
	"Neil Brown" <neilb@...e.de>
Subject: RE: GPL vs non-GPL device drivers

Combined responses

> On 2/20/07, Michael K. Edwards <medwards.linux@...il.com> wrote:
> > There is no legal meaning to "combining" two works of authorship under
> > the Berne Convention or any national implementation thereof.  If you
> > "compile" or "collect" them, you're in one area of law, and if you
> > create a work that "adapts" or "recasts" (or more generally "derives
> > from") them, you're in another area of law.

> As I said, you're having a purely semantic argument.

When someone uses a term that can mean more than one possible thing and then
uses the fact that it's true for one meaning to argue that it must be true
for the other, what else can you do other than point out that the term they
are using has multiple meanings?

> Regardless of what you *call* it, shoving two works together does not
> excuse you from the conditions of the license on one of those works,
> *when you make a copy*.

Nobody is arguing otherwise. Nobody is saying "you don't have to comply with
the GPL". We're saying that the GPL doesn't mean what people think it means.
In some cases it's because of the GPL's wording, but in other cases it's
because the GPL cannot set its own scope.

> And that's the GPL in a nutshell, if you want
> to copy the work, you need a license, if you want a license, you must
> abide the conditions, and one of the conditions is that you may not
> combine it with a work that is under an incompatible license unless it
> is mere aggregation.

Right. And all the cases we are talking about are mere aggregation (that is,
they do not create derivative works).

> Of course, now you're going to argue that there's no such thing as an
> "incompatible license" or "mere aggregation" and that these are just
> words that were made up for the GPL, so they can be ignored.. another
> pointless semantic argument because it doesn't change the very simple
> fact that you don't have any rights to copy the work unless you have a
> license and you don't have a license if you fail to abide the
> conditions of the license.

The issue is about what the license *means* and what its scope is. For
example, if the license said "if you ever use a work that is subject to this
license, you must place every work you create after that under the license",
that would obviously not be enforceable. The question of the *scope* of the
license is critical, and you can't read the license to understand its scope.

> Hang on, you're actually debating that you have to abide by conditions
> of a license before you can copy a copyright work?

Well, there are certainly cases where you can. Necessary step, first sale,
and fair use all provide possible situations where you can copy a
copyrighted work without complying with the license. But more generally, the
argument is over the scope of the license.

The GPL doesn't restrict the distribution of mere aggregations not because
the authors felt this was a wise decision, but because it *can't*. That is
outside the GPL's power. So the question of what's a "mere aggregation" is a
legal question about the scope of the GPL, and you have to look at law and
case law to understand it, not the text of the GPL.

The GPL grants you the permission to modify and distribute the original work
if you distribute the source of the original work. The GPL also puts certain
requirements on the distribution of derivative works. This is, as far as I
can tell, the maximum of the scope it can have.

You are trying to cram this in a simple yes or no box, and it just doesn't
fit. There are questions nobody knows the answers to (such as what rights
you need to distribute a derivative work or whether compiling code makes a
translation).

DS

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/