GPL: Static/Dynamic vs Derivative work and the Conservative position (Re: GPL, yet again.)
First off, please fix your MUA so that it does not delete References:
and In-Reply-To:, and follow Debian list policy and refrain from
Cc:'ing people who have not explicitely requested it.
On Mon, 12 Sep 2005, Alexander Terekhov wrote:
> Yorick Cool wrote:
> > The reasoning is pretty simple. As Sean said, it is based on the
> > understanding that the GPL is a contract.
> Yeah. Except that everybody and his dog knows that the FSF's
> position is that GPL != contract. Moglen and RMS now call it "The
> Constitution" (of the GNU Republic, I suppose).
Whether the GNU GPL is a contract or a license (and indeed, what these
terms even mean) is dependent upon the local jurisdiction in which it
is being enforced. It's not even a terribly interesting question in
this particular aspect, because the GNU GPL doesn't attempt to
restrict usage that isn't already restricted by copyright law.
At least in the US, the critical metric is not "Dynamic linking" or
"Static linking". The critical metric is "derivative work". If a
resultant work is a derivative work of a GPLed work, then it must
comply with the GPL. If it is not a derivative work, then it need not
comply with the GPL.
The argument of the FSF has been (and continues to be, TTBOMK) that
dynamic linking with a GPLed work forms a derivative work when the
binary is distributed. This works back up the chain to works which
link to things which link to GPLed works because the intermediate work
becomes subject to the GPL as well. AFAIK, there isn't any relevant
case law covering this type of derivative work, so it's just the
opinion of the license drafter and a key copyright holder of GPLed
works. Because of this latter point, as far as Debian is concerned, we
should continue to assume that dynamic linking forms a deriviative
work. This is the conservative position, which is why we've been
assuming this for the purposes of distributing works in Debian.
You could say she lived on the edge... Well, maybe not exactly on the edge,
just close enough to watch other people fall off.
-- hugh macleod http://www.gapingvoid.com/batch8.htm