On Tue, May 21, 2013 at 05:54:44PM +0200, Antoine Pitrou wrote:
IMHO, copyleft-next should be forward-compatible, not only with
GPLv2
and GPLv3, but also LGPL and perhaps MPL.
Originally I dismissed the idea of forward compatibility with LGPL
(though I naturally thought about it) because of the assumption that
copyleft-next would be a strong copyleft license in some very GPL-like
sense, plus (maybe this was more important) LGPL is itself
forward-compatible with GPL.
This implies to have a "weak" copyleft model, and I think
that's the way
to go, as noted in the previous discussion, given the increasing
importance of libraries and dynamic languages in free software
ecosystems.
It is clear that without intending this to have happened,
copyleft-next is showing signs of evolving into something 'weaker'
than I had originally conceived. This is somewhat disturbing (not in a
bad way necessarily) for various reasons. For one, I believe I
originally assumed that weak copyleft was actually the dead
category. For another, I thought Mozilla did a decent job with MPL
2.0. :)
As for forward compatibility with LGPL, though... this is easy to fix
but I am not sure it is really necessary. It is possible to interpret
LGPL as what I've now inaccurately described as inbound-compatible --
that is it is arguably compatible in the same sense that EPL is
arguably compatible. That may rest on a 'weak' interpretation of
LGPL. And as a policy matter it doesn't seem as compelling -- I take
your point about libraries and dynamic languages, indeed this has been
influencing a lot of my recent thinking, but LGPL is presumably
declining in use in that sphere.
- RF