octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Octave Forge: Package groups and properties defined, RFC.


From: Julien Bect
Subject: Re: Octave Forge: Package groups and properties defined, RFC.
Date: Thu, 9 Mar 2017 10:43:56 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Icedove/45.6.0

Le 09/03/2017 à 08:34, Olaf Till a écrit :
On Wed, Mar 08, 2017 at 04:31:25PM +0100, Julien Bect wrote:
Le 08/03/2017 à 16:19, Juan Pablo Carbajal a écrit :
On Wed, Mar 8, 2017 at 10:56 AM, Carlo De Falco <address@hidden> wrote:
On 8 Mar 2017, at 10:21, Olaf Till <address@hidden> wrote:

The point was actually that the _package_ must be compatible with
Octaves GPL3+. You're right, if the package contains only m-code, an
arbitrary free software license is enough (GPL3+ recommended). We
probably should reword this. But "open source" won't do, since it
doesn't necessarily mean "Free/Libre Software".
Of course an m-code package must necessarily be "open source" as m-files
are human readable, but compatibility whith the license of Octave is
not required in this case, so, as far as I know, but I am not a lawyer,
distribution of non-free or even proprietary packages is entirely possible.

I'm not saying we should promote non free software, of course, just that it is 
possible.

c.

My comment only applies to external packages. I would not make
compromises on the community packages: these must be Libre software.
Againk, by putting strong constrains on the external packages we will
loose what I consider perfectly valid Libre software under GPlv3
incompatible licenses (like UEPL).
We also loos the chance to add people who do not understand Libre
software, but like the idea, although they do not call it like that.
What about the list of OSI-approved licences [1] ?

Would that be acceptable as a weaker requirement for external packages ?
(I am not a lawyer...)

Any code for oct-files in distributed packages must be compatible with
GPL3+. This is enforced by Octaves license.

For m-code in the external packages, we could just require Free
Software (in the sense of Libre Software) licenses, not necessarily
(but advised to be) compatible with GPL3+. This requirement could
already be 'weak' enough for external packages.

I'm a bit irritated by the existence of different lists of Free
Software licenses. Maybe we shouldn't prescribe any of these lists for
m-code. If a license turns up which is listed only at opensource org,
we could use our common sense.

Ok, sure... I was only trying to provide a practical way for people to know what kind of licence is acceptable or not...

To remain closer to the FSF view of things, we can use their list [1] instead.


If I understand previous messages correctly, this should be acceptable :

a) "GPL-compatible licences" [2] for 1) community packages, or 2) external packages that link directly to Octave (oct-files or mex-files).

b) "GPL-compatible licences" [2] or "GPL-incompatible" [3] for pure m-file external packages.

c) "Nonfree" -> not accepted.


Finally, concerning the terminology issue raised by JP (free/libre/open/etc.) : I think that if we decide to provide pointers to FSF pages as suggested above, then we should stick to the FSF definitions/vocabulary and use the word "free" only.

@++
Julien


[1] https://www.gnu.org/licenses/license-list.html
[2] https://www.gnu.org/licenses/license-list.html#GPLCompatibleLicenses
[3] https://www.gnu.org/licenses/license-list.html#GPLIncompatibleLicenses
[4] https://www.gnu.org/licenses/license-list.html#NonFreeSoftwareLicenses




reply via email to

[Prev in Thread] Current Thread [Next in Thread]