"Michael K. Edwards" <[EMAIL PROTECTED]> writes:

> On Thu, 13 Jan 2005 17:02:52 -0500, Brian Thomas Sniffen
> <[EMAIL PROTECTED]> wrote:
> [snip]
>> Why are copies OK, and derivative works not?  I see GPL 2b talking
>> about any work that in whole or in part contains the Program.
>> Eclipse+Kaffe contains Kaffe, GPL 2 then exempts mere aggregation --
>> which this is not.  It also exempts separate sections *when
>> distributed separately* -- and explicitly covers them when the Program
>> is distributed as a whole.
>
> GPL section 2 is all about "works based on the Program", read in the
> context of copyright law.  Later in that section, it says that "the
> intent is to exercise the right to control the distribution of
> derivative or collective works based on the Program."  Eclipse+Kaffe
> isn't a "work based on the Program" because it isn't copyrightable as
> a derivative or collective work.  In other words, you can't combine
> Eclipse and Kaffe, by running Eclipse on Kaffe, and then sue me for
> copyright infringement for doing the same.

It's very clear that GPL 2 refers to "any work that you distribute or
publish, that in whole or in part contains or is derived from the
Program or any part thereof".

> The "mere aggregation ... on a volume of storage or distribution
> medium" bit may clarify the contract writer's intent to some degree,
> but it doesn't automatically make other ways of combining two works
> copyrightable.  To get at the cases the FSF is shooting for, they
> would have to use terms of art instead of "derivative or collective
> works", and would have to insert far more draconian provisions to
> create an action for breach of contract when GPL and non-GPL works are
> combined.

They did: the word is "copy".

> The question "does linking create a derivative work" won't really be
> settled in the US unless and until a new case arises, is adjudicated
> in a court of fact, is appealed, and is properly analyzed with
> reference to precedent by a circuit court.  I think (IANAL) that the
> outcome is quite predictable in light of precedents I have cited from
> several circuits.  Ultimately, linking against the public interface to
> a published library doesn't detract from the rights in authorship that
> copyright law was created to protect, and US appellate courts are
> surprisingly good at sticking to the purpose of copyright in the face
> of legal and technical contortions that attempt to leverage a
> copyright monopoly beyond its legitimate bounds.

But distributing copies of that library does infringe the intended
rights of a copyright holder.

-Brian

-- 
Brian Sniffen                                       [EMAIL PROTECTED]

Reply via email to