* simo wrote, On 22/11/07 15:25: > On Thu, 2007-11-22 at 13:33 +0000, Sam Liddicott wrote: > >> * Alex Hudson wrote, On 22/11/07 13:09: >> >>> On Thu, 2007-11-22 at 12:50 +0000, Ciaran O'Riordan wrote: >>> >>> >>>> Sam Liddicott <[EMAIL PROTECTED]> writes: >>>> >>>> >>>>> The GPL is widely considered a share-alike license where licensors have >>>>> understood that the same terms will propagate throughout the distribution >>>>> chain. >>>>> >>>>> >>>> You're presenting an argument against additional requirements as being an >>>> argument against AGPL compatibility. >>>> >>>> Apache licence compatibility was achieved by allowing people to add the >>>> requirements of Apached licensed code to GPLv3 licensed code. >>>> >>>> >>> Are you sure about that? I don't see anywhere in the GPLv3 which says I >>> can attach extra restrictions in Apache licenses to GPLv3'd code. GPLv3 >>> + Apache doesn't have further restrictions on the GPL that I'm aware of. >>> >>> I think the point is that the GPL always set a maximum level of >>> restriction, and although you could lessen them (e.g., LGPL), you >>> couldn't add to them. That has now changed: the AGPL is the maxima, >>> effectively, and the GPLv3 could be simply written as the AGPL plus a >>> grant of permission. >>> >>> That's not the same as designing the basic license to be compatible with >>> other popular license. >>> >>> >> My current understanding is that the AGPL puts restrictions on GPL3 >> software when and for as long as that GPL3 software is combined with >> AGPL software. >> > > No it adds requirements, no restrictions on what users can do, just > requirements when they are done with it. > > Let's use the right words please. > > >> Please could some kind soul confirm this understanding? >> > > I don't think you can say AGPL add restrictions, no. > > Would you say it added requirements then? (I guess you knew what I meant) >> It leaves me wondering if the next release of Microsoft Windows would >> try a similar but more restrictive clause; >> > > And how this would be relevant? > > Because the FSF would lead the way in a future license subverting the supposedly fixed terms of past licenses. >> I don't like the idea that one license can restrict the terms of >> another license. >> > > In fact this does not happen. The *requirement* is only for AGPL or > combined works with the AGPL. But the work under GPLv3 even when > combined remains under the GPLv3. > The fact that you use the word requirement doesn't change my internal concern, it just makes me repeat it using different words:
I don't like the idea that one license can add place extra requirements on use of software governed by another license. if it works as I understand, then as a USER of GPL3 software (not distributor) I have to provide it's source if I use it with AGPL software to provide a web service. I used to have to provide the source only if I distributed it. If so, then the AGPL is putting extra restrictions on my USE of GPL3 software by requiring that I meet the requirements of another license. Thanks what I meant. Previous licenses have been based on copyright, by granting conditional distribution rights, but section 13 of the seems not to be AGPL such a term, it restricts use regardless of distribution. Is this based on copyright permissions required to "install" the software? Or does 13 affect non-distributing service providers? Sam
_______________________________________________ Discussion mailing list [email protected] https://mail.fsfeurope.org/mailman/listinfo/discussion
