Questions about copyleft-next

Richard Fontana fontana at sharpeleven.org
Tue Jul 1 17:33:29 UTC 2025


On Tue, Jul 1, 2025 at 12:56 PM secureblueadmin
<secureblueadmin at proton.me> wrote:
>
>
> > it provides some support for the view that a CLA is not needed.
>
> I should have been more precise, I specifically meant the all-too-common type of CLA used alongside the AGPLv3 nowadays, where contributors have to sign over carte blanche rights to the project owners. Doesn't section 6 preclude that kind of CLA, or at least contradict it?

Section 6 is what you might call an "inbound=outbound" clause, but
just as in the inspirational provision of the Apache License 2.0, it
doesn't do anything to stop a project from using the sort of CLA
you're talking about if it's determined to do so.

Maybe there's a way to accomplish what you're talking about, though.
Like generalizing the copyleft equality clause (as bkuhn calls it) or
something like that.

> > latest "draft" version
>
> Explicitly listing compatible inbound licenses seems like it could cause issues. What about niche licenses like https://opensource.org/license/bsdpluspatent ?
>
> > https://git.copyleft.org/copyleft-next/copyleft-next/src/commit/5fdb9dca6bd9c7ab204976e46ce4663b8d4482c7/Drafts/copyleft-next#L72-L80
>
> Doesn't this make it trivially easy to get around the "affero clause"?

> > and a condition in
> >   the Compatible License conflicts with any of the terms of this
> >   License, You have permission to comply with that Compatible
> > License
> >   condition.
>
> Both GPLv2 and v3 contain clauses stating:
>
> (v2)
> > You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
>
> (v3)
> > You may not impose any further restrictions on the exercise of the rights granted or affirmed under this License.
>
> Ergo per the copyleft-next terms, I have permission to instead comply with this term of the GPL, and ignore the "affero" clause of the copyleft-next license. Am I missing something here or is this a loophole? I believe the EUPL has the same problem if I'm not mistaken.

So what 'Draft' copyleft-next says is:

The following activities do not violate [the no-further-restrictions
requirement]:
"a. Distribution of a Covered Work incorporating material governed by a
Compatible License, but the requirement in section FIXME to provide
Corresponding Source is in no case limited by this permission. If
the Compatible License is a GPL-Family License, and a condition in
the Compatible License conflicts with any of the terms of this
License, You have permission to comply with that Compatible License
condition."

So at least with respect to the Corresponding Source requirement [that
should have been changed to "CCS" which is the defined term used in
this draft version], this is trying to address the concern you're
raising. If you always have the original license requirement to
provide CCS, does it matter whether nominally the license of part of
the source code is BSD+Patent? That's the question, anyway. And
doesn't the same issue otherwise exist in the *GPL* licenses? I can
have a big chunk of some AGPLv3 work be under BSD+Patent, presumably.

Richard


More information about the next mailing list