FOR APPROVAL - Python License Changes

[Lawrence Rosen]

John Cowan wrote:
> From what I understand, only the top-level license is operative.  Other
> licenses are there because their own conditions say they must be carried 
> along with derivatives of the old code, but all old code has acquired
> enough patches that it is no longer available solely on the terms of
> old licenses.
> 
> I may be quite wrong about this.

Indeed, when you receive a GPL program you can assume that the top-level GPL
license is operative for that program. 

You may also assume in today's world that the GPL author has included other
components in that GPL program whose licenses are compatible with the GPL
(as she understands "compatibility"). There may be Apache components (or
derivative works of such components) in that program, and the Apache License
is still operative for those components. If you fail to honor the conditions
of the Apache License, you may suddenly become an infringer. 

For example: "If You institute patent litigation against any entity
(including a cross-claim or counterclaim in a lawsuit) alleging that the
Work or a Contribution incorporated within the Work constitutes direct or
contributory patent infringement, then any patent licenses granted to You
under this License for that Work shall terminate as of the date such
litigation is filed." [Apache License § 3]

That would depend on a patent infringement analysis, including a better
understanding of the Apache "Work" and "Contribution" at issue, despite
anything the GPL or any other top-level license says.

Note that many modern FOSS licenses (but not the Apache License) defensively
terminate both the copyright and patent licenses in the event of a patent
infringement lawsuit, so that you may become both a copyright and patent
infringer just by suing "any entity" for certain patent infringement even
though that software is buried in a GPL or other program.

I sort of like the way this makes software patent infringement lawsuits
potentially more costly for plaintiffs. At the very least, it should make
them look before they leap to sue. 

At least, John, that's the way I analyze software containing multiple
licenses. Maybe other lawyers look at it differently????

/Larry




> -----Original Message-----
> From: John Cowan [mailto:cowan at ccil.org] On Behalf Of John Cowan
> Sent: Wednesday, May 25, 2011 2:53 PM
> To: Lawrence Rosen
> Cc: 'VanL'; license-review at opensource.org
> Subject: Re: FOR APPROVAL - Python License Changes
> 
> Lawrence Rosen scripsit:
> 
> > Van, I've never really understood the Python licensing situation. It
> > was there before I arrived on the scene (guess how old that makes
> it!)
> > and I decided to just let it be without questioning its provenance.
> >
> > Most FOSS software nowadays is composed of contributions under
> > different licenses from many authors. Most distributors aggregate
> > those works and distribute that software under a single license
> > (perhaps even GPL or Apache). So what is unique about Python that
> > it still requires multiple licenses for its distribution? Can't
> > the Python Foundation find a single license that covers the
> > entire Python collective or derivative work that they are now
> > distributing? Hopefully an existing license so that we can limit
> > proliferation....
> 
> From what I understand, only the top-level license is operative.  Other
> licenses are there because their own conditions say they must be
> carried
> along with derivatives of the old code, but all old code has acquired
> enough patches that it is no longer available solely on the terms of
> old
> licenses.
> 
> I may be quite wrong about this.
> 
> 
> --
> With techies, I've generally found              John Cowan
> If your arguments lose the first round
> http://www.ccil.org/~cowan
>     Make it rhyme, make it scan                 cowan at ccil.org
>     Then you generally can
> Make the same stupid point seem profound!           --Jonathan Robie