[racket-dev] LGPL

Three hours ago, Neil Van Dyke wrote:
> Eli Barzilay wrote at 03/03/2011 10:10 PM:
> >   Distributed under the same terms as Racket
> 
> Would it be good practice overall to pick a specific, more limited,
> license for contributors to use?

I have no idea.  The "distributed ... same as" is something that seems
to be doing the most effective job ATM, which is why I used that.  To
clarify the sense of effectiveness -- I basically wanted to get things
sorted out for the debian distribution and minimize any future need to
deal with this whole licensing thing again.


> I believe that a copyright holder permitting the convenient
> "distributed under the same terms as Racket" essentially grants full
> (nonexclusive) rights to the unnamed distributors of Racket,
> potentially bypassing the intent of the open source licenses.

IIUC, it delegates to however racket is distributed as, which plt
controls, not some unnamed distributors.


> Some individual contributors will believe strongly in open source
> licenses, and there's various good reasons why licenses are used
> rather than public domain in most cases.  I'd also bet money that
> the open-ended granting of rights would draw more consternation from
> corporate lawyers whose OK is needed to release code, compared to a
> well-trod standard license with limitations, like LGPL 2.1 or 3.

I didn't even try to figure out if the current racket license is 2.1
or 3 -- that would force me to learn the differences, including
knowing all kinds of obscure facts about possible mismatches between
the version when something happens or whatever.  I'm happy to not know
about it.


> I didn't mind giving full nonexeclusive rights to that CSV parser as
> "same terms as Racket", just for the sake of expedience on that
> small contribution.  But, going forward in the Racket new world
> order, I'd like to be comfortable that license stuff is being done
> in the best way overall.

Given the above, I'm clearly not a good choice for getting advice...
(But if you see something that is wrong in the tree, then I'll do
whatever's needed.)

-- 
          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
                    http://barzilay.org/                   Maze is Life!