Mailing-List:	contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive:	<http://sourceware.cygnus.com/ml/cygwin/>
List-Post:	<mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help:	<mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender:	cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To:	mailing list cygwin AT sourceware DOT cygnus DOT com
Message-Id:	<200005031505.JAA08814@chorus>
Date:	Wed, 3 May 2000 09:05:24 -0600 (MDT)
From:	"13mb80000-HallM(10053584)37x10" <marcus AT bighorn DOT dr DOT lucent DOT com>
Reply-To:	"13mb80000-HallM(10053584)37x10" <marcus AT bighorn DOT dr DOT lucent DOT com>
Subject:	Re: Things you can do with Cygwin
To:	dj AT delorie DOT com
Cc:	cygwin AT sourceware DOT cygnus DOT com
X-Mailer:	dtmail 1.2.1 CDE Version 1.2.1 SunOS 5.6 sun4m sparc
X-Sun-Text-Type:	ascii

> > Does that mean that the two pieces are now separate works?
> >
> > So, I think that there must be some other criteria for separating works
> > other than the existance of alternative implementations and standard
> > protocols.  I can't say quite what the criteria should be, though...
> 
> It wouldn't matter.  You can't retroactively un-violate the GPL.  The
> first time you distributed the two programs without full source, you
> violate the GPL.  *If* later they become two works, then *further*
> distribution would be OK.
> 
> As for the criteria, it's simple.  A court would decide.  Otherwise,
> it's really rather pointless to try to find such borderline cases,
> unless you *like* going to court just to split hairs.  If you don't
> know where that fine line is, just stay clear of it.

I am not considering the GPL implications yet, still thinking about the
meaning of "a work".  I do realize that this has implications as to how
the GPL gets applied, though.  Still, it somehow seems pecular if I write
some code and the number of "works" that I have created can legally
change later on, well after I have written the last line and shipped
it off.

In the original example, creating an extension to a GPL-ed X-server (a
Y server) and a client that requires this extension, you argued that
they collectively are a single work.  But, if somebody writes another
Y server, then suddenly the original pieces are two works.  Even if
the second Y server is implemented 10 years after the first?  Does it
matter how well distributed the 2nd implementation is?  Could you just
start out with 10Mb of all zeros and work through each bit pattern possible,
assuming that somewhere along the way an alternate implementation is
created.  If so, then do you have to actually do that work, or could you
just speculate that such could be created?

It seems that the number of works created should be intrinsic to the act of
creating them, and not affected by subsequent and possibly unrelated
activity.

marcus

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com

- Raw text -