Questions about copyleft-next

secureblueadmin secureblueadmin at proton.me
Tue Jul 1 18:00:12 UTC 2025


> 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.

I see. It requires licensing patches under copyleft-next, but doesn't preclude licensing patches under other licenses as well.

> 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?

I think something got mixed up here. I was making two separate and unrelated points. 

Point 1: Why have an explicit inbound list as opposed to allowing any license approved by the OSI and FSF?

Point 2 is a hypothetical:

Say for whatever reason I want to get around the network licensing clause of copyleft-next. copyleft-next allows me to disregard any condition of copyleft-next that conflicts with the terms of a Compatible License:

>  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.

The GPL contains "no further restrictions" clauses that conflict with the "network distribution" clause of copyleft-next. Therefore to circumvent the network distribution clause, I can write some GPL code, mix it with copyleft-next code, and release that as a new project-ng without the network distribution clause, as permitted by the copyleft-next draft license and as required by the GPL. Now I've created a version of the original code that can be used downstream without adherence to the network clause.

What am I missing here?

On Tuesday, July 1st, 2025 at 10:33 AM, Richard Fontana <fontana at sharpeleven.org> wrote:

> 
> 
> 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