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

Re: Cheops-ng: DFSG free or non-free?

Scripsit "Wesley W. Terpstra" <terpstra@ito.tu-darmstadt.de>
> On Fri, Jan 23, 2004 at 11:27:20AM -0500, Joe Moore wrote:

> > If the package contains GPL'd code that is written by someone
> > other than the main copyright holder (Adtran), then the package
> > is undistributable.

> He includes adns (GPL) in the source tarball; so, that's that.

> This means that they themselves are not allowed to distribute the program
> too, correct? Since they are violating the terms of adns's copyright?

Yes. So it would probably make sense to conspire with the author of
adns to prod him gently with a large pole.

> > If the package contains only code copyrighted by Adtran (or at least, code
> > that is licensed under this strange GPL+advertising license), then the
> > actual license is not the GPL, but a new GPL-like (but GPL-incompatible)
> > license that includes this advertising clause as part of its terms and
> > conditions.

> > In [this case], the license info for the Debian package should reflect
> > that it is _not_ licensed under the GPL.

> Woah. That's crazy!

No - in fact, it's the most straightforward interpretation of the
author's statement.

> > Third, (and this is uncertain legal territory) the work is licensed under
> > the GPL, since the GPL explicitly delimits the "terms and
> > conditions" in its text, and the additional requirements in the
> > cited COPYING file are not legally binding, but are merely a request.

> I take it that this is insufficient grounds for inclusion in main?
> (Due to the uncertainty?)

Indeed. We've had our flamewars about whether what the author is
trying to do is inherently meaningless of just GPL-compatible; but I
don't think anyone has seriously proposed just to ignore the extra
condition. It would be very hard to uphold such an interpretation in
court, and it would be a disservice to our users to claim (implicitly,
by treating the software as free) that the should expect to be able

Henning Makholm                         "This imposes the restriction on any
                                  procedure statement that the kind and type
                             of each actual parameter be compatible with the
                       kind and type of the corresponding formal parameter."

Reply to: