Re: no need to keep non-copylefted files that way in a copylefted project. (was Re: FRR package in Debian violates the GPL licence)
On 20/03/2019, Bradley M. Kuhn <bkuhn@ebb.org> wrote:
> This is an example of a common trend I see: social pressure to keep
> non-copylefted code under non-copyleft licenses, sometimes even escalating
> to aggression (as the OpenBSD project did with Linux over wireless drivers),
> while permitting and even encouraging licensors to incorporate the code
> under proprietary licenses, which are much more restricted of copyleft.
Very well put. Thanks.
But there is an important difference here that make things even worse.
The code distributed under a non-copyleft license depends heavily on
copylefted one, so much that it's not possible to run (or even
compile) it without the pre-existing copylefted one (that includes C
headers that are not described by any standard).
So in a way OpenBSD's social pressure might be interpreted as an
attempt to keep a reciprocity of contribution (you got our code this
way, give back your change that same way) in a context where OpenBSD's
favourite license do not grant such reciprocity.
This is somewhat ironic, but it's not what is happening here.
On 20/03/2019, Florian Weimer <fw@deneb.enyo.de> wrote:
> The behavior becomes much more reasonable if you assume that a
> proprietary variant of the code exists somewhere, and the authors hope
> to merge back contributions into it, under the original non-copyleft
> license.
That's why I said they could (at most, but I guess Bradley can correct
me) double license the modifications, say as GPL and BSD or MIT.
Downstream, developers of a compatible proprietary variants might then
chose to terminate their own GPL license on both FRR and Quagga and
adapt those specific patches to that tool.
The cons of this is that they are not going to ever be able to
distribute or modify these projects without violating the authors'
copyright.
And tbh, I don't know if such voluntary termination of a copyleft
license can be done privately or should be publicly declared somehow.
Giacomo
Reply to: