Hi

Oh wow. Is this issue documented anywhere? Curious :-)

Cheers Dave

Hi

Yikes! What a problem. I suggest raising money from the front page to hire someone to sort that out! :-)

Cheers Dave

On Tue, Jul 24, 2012 at 12:43 PM, Guillermo Espertino (Gez) <gespertino@...400...> wrote:

Afaik there was also an issue with pixman/cairo bitmap downsampling that could be potentially a release-blocking regression. Did the board decide something regarding that problem?

Yes, that is still an issue as far as I am aware. However, that's a much more easily solvable problem to fix. It's also not the board's realm to deal with normal problems that the developer community can more easily straighten out. Honestly, we probably have a few other current blockers bugs as well.

In the end though, the licensing situation is really a much bigger and more complicated problem that will probably require getting legal counsel involved.

Cheers, Josh

On 25 Jul 2012, at 12:52, Krzysztof Kosiński <tweenk.pl@...400...> wrote:

2012/7/24 Josh Andler <scislac@...400...>:

...

On Tue, Jul 24, 2012 at 12:43 PM, Guillermo Espertino (Gez) <gespertino@...400...> wrote:

...

Afaik there was also an issue with pixman/cairo bitmap downsampling that could be potentially a release-blocking regression. Did the board decide something regarding that problem?

Yes, that is still an issue as far as I am aware. However, that's a much more easily solvable problem to fix. It's also not the board's realm to deal with normal problems that the developer community can more easily straighten out. Honestly, we probably have a few other current blockers bugs as well.

In the end though, the licensing situation is really a much bigger and more complicated problem that will probably require getting legal counsel involved.

Most files are annotated as 'released under GNU GPL' and point to the COPYING file for license information, without specifying the license version. Is it safe to assume this means 'GPL v2 or later'? I guess it would be up to the legal counsel to find this out.

The feature of calculations in spinboxes uses GPL v3 code from GIMP. This can only stay in the next release if we really are GPL v2 or later, and needs to be behind a compile-time switch, because some people might want to use Inkscape under GPL v2.

Regards, Krzysztof

---

Unless we can get permission for that code to be released on a v2 license? May be easier than trying to get the rest of the code base v3...

On Wed, Jul 25, 2012 at 1:09 PM, Martin Owens <doctormo@...400...> wrote:

On Wed, 2012-07-25 at 13:52 +0200, Krzysztof Kosiński wrote:

...

because some people might want to use Inkscape under GPL v2.

The GPL only applies to you if you are distributing the licensed work, not if you're using it. That's why it's called a copyright license and not a use-it-right license.

And are people really that paranoid about the GPLv3? On reading, it's a good improvement over v2, the only reason why people would be trying to move away from it would be if they'd been infected with some sort of curious meme.

in any case, retaining the four freedoms and maintaining the copyleft are the most important principles; the exact license isn't very important in the grand scheme.

Okay, I should shut up and be doing vacationy stuff, and will start after this...

I talked with Bradley from the SFC and it sounds like as long as all licenses are compatible, the project/releases would become GPLv3 but the "GPLv2 and later" licensed files don't need to change. If people want to use those v2 in other v2 or compatible projects, it's all good. I personally think putting in compile time flags to make it only use v2 code is not worth any core developers time... if a third party (or any project member) feel strongly enough to put something forth a patch or commit, that's all good for them. I just don't know that as a project we'd really want to maintain a compile time flag like that.

Martin, as mentioned in the board thread (which I don't know if you read or not), the freedoms thing is indeed the biggest concern, not the specific license.

It's sounding like now it's just looking to see if there are any files which are explicitly v2 only or otherwise not compatible with v3. I won't have a chance to look into this until after after the 6th. Anyone else that wants to take it one prior to that is more than welcome to (as mentioned though, I was already planning on doing so... even the old fashioned way of looking in every file manually).

Bradley will be contacting the board/committee to discuss things with us. Note: Aside from the freedoms thing and not caring about the license... we also don't want to have the project come across as being ambivalent towards licensing. We're not that paranoid or worried, we just would prefer to have our i's dotted and t's crossed... we are a member organization of the SFC and it's just smart that we utilize their services and get their guidance to ensure we're steering things in the right direction.

Anyway... I hope you all enjoy the next couple weeks and I will be back in contact in the near future.

Cheers, Josh

On Thu, 2012-07-26 at 21:05 +0000, Dave Crossland wrote:

Is anyone is interested in doing this?

Looking at each file's headers or the commit log?

Using a modified checklicense script, I've pulled the attached statistics.

There is a lot of cleanup required for all these files... someone with commit access needs to go through and clean up some of these obvious problems.

Especially worrying are all the files without copyright headers at all, especially those in ./src (mostly headers though).

2 CMakeScripts,*No copyright* GENERATED FILE 1 CMakeScripts,*No copyright* GPL (v2 or later) 70 cxxtest/cxxtest,*No copyright* UNKNOWN 1 cxxtest/docs,*No copyright* UNKNOWN 4 cxxtest,*No copyright* UNKNOWN 1 cxxtest/sample/gui,*No copyright* UNKNOWN 14 cxxtest/sample/mock,*No copyright* UNKNOWN 1 cxxtest/sample/mock/T,*No copyright* UNKNOWN 18 cxxtest/sample,*No copyright* UNKNOWN 1 cxxtest/sample/winddk,*No copyright* UNKNOWN 1 ./,LGPL (v2.1 or later) 1 ./,*No copyright* GENERATED FILE 2 ./,*No copyright* UNKNOWN 6 packaging/macosx,GPL (v2 from COPYING) 1 packaging/macosx/Resources/bin,*No copyright* UNKNOWN 1 packaging/macosx/Resources/themes/Clearlooks-Quicksilver-OSX/gtk-2.0/Scrollbars_1,*No copyright* UNKNOWN 1 packaging/macosx/Resources/themes/Clearlooks-Quicksilver-OSX/gtk-2.0/Scrollbars_6,*No copyright* UNKNOWN 1 packaging/macosx/ScriptExec,GPL (v2 or later) (with incorrect FSF address) 4 po,*No copyright* UNKNOWN 2 share/attributes,*No copyright* GENERATED FILE 2 share/attributes,*No copyright* UNKNOWN 6 share/extensions,Apache (v2.0) 28 share/extensions/Barcode,GPL (v2 or later) (with incorrect FSF address) 2 share/extensions,BSD (3 clause) 2 share/extensions,GENERATED FILE 2 share/extensions,GPL (v2 from COPYING) 24 share/extensions,GPL (v2 or later) 2 share/extensions,GPL (v2 or later) GENERATED FILE 168 share/extensions,GPL (v2 or later) (with incorrect FSF address) 24 share/extensions,GPL (v3 or later) 4 share/extensions/ink2canvas,GPL (v2 or later) (with incorrect FSF address) 1 share/extensions/ink2canvas,*No copyright* UNKNOWN 10 share/extensions,MIT/X11 (BSD like) 2 share/extensions,*No copyright* GENERATED FILE 82 share/extensions,*No copyright* UNKNOWN 78 share/extensions/test,*No copyright* GENERATED FILE 18 share/extensions/test,*No copyright* UNKNOWN 4 share/extensions/test,UNKNOWN 4 share/extensions,UNKNOWN 14 share/extensions/xaml2svg,MIT/X11 (BSD like) 4 share/filters,*No copyright* UNKNOWN 1 share/palettes,*No copyright* GENERATED FILE 2 share/palettes,*No copyright* UNKNOWN 1 share/patterns,*No copyright* UNKNOWN 93 src/2geom,MPL (v1.1) 18 src/2geom,*No copyright* UNKNOWN 4 src/2geom,UNKNOWN 1 src,GENERATED FILE 296 src,GPL (v2 from COPYING) 5 src,GPL (v2 or later) 3 src,LGPL (v2.1 or later) 8 src,MPL (v1.1) GPL (unversioned/unknown version) 9 src,*No copyright* GPL (v2 from COPYING) 1 src,*No copyright* GPL (v2 or later) (with incorrect FSF address) 130 src,*No copyright* UNKNOWN 27 src,UNKNOWN

Raw Data: http://paste.ubuntu.com/1113233/

Martin,

Hi!

On 27 July 2012 06:37, Martin Owens <doctormo@...400...> wrote:

On Thu, 2012-07-26 at 21:05 +0000, Dave Crossland wrote:

...

Is anyone is interested in doing this?

Looking at each file's headers or the commit log?

Using a modified checklicense script, I've pulled the attached statistics.

Wow, that's AWESOME :-) Great work Martin! :-D

Especially worrying are all the files without copyright headers at all, especially those in ./src (mostly headers though).

I believe the release-blocking problem is what Josh said, getting permission to change any files which are explicitly v2 only or otherwise not compatible with v3 into v2-or-later or otherwise v3 compatibility.

While serious, I think the lack of copyright notices isn't a release blocker, because copyright adheres without notices, and version control (specifically `bzr qlog` and `bzr qannotate`) will reveal who the copyright holders are (and that may be more reliable than notices in the text files.)

I think its most problematic that the COPYING file is just the GPLv2 text - http://bazaar.launchpad.net/~inkscape.dev/inkscape/trunk/view/head:/COPYING - because to me that implies that any file without a copyright notice alongside an alternative license declaration -- all your UNKNOWN lines -- are indeed GPLv2.

I wonder about adding your modifed checklicense script into the repo, moving COPYING to GPLv2.txt and adding a GPLv3.txt file to the repo, and creating a new COPYING to have some explanatory text about the licenses used in the codebase and how to use the checklicense script.

But yeah, seriously good work :-)

On 27 July 2012 19:44, John Cliff <john.cliff@...400...> wrote:

the history for the files is not complete in the bzr repository.

Ahh, a pity. I suppose that makes this quite a lot of administrative work by a human being then :(

On 27 July 2012 02:09, <inkscape-devel.neophyte_rep@...2295...> wrote:

On Thu, Jul 26, 2012 at 12:15 PM, Josh Andler - scislac@...400... wrote:

...

I talked with Bradley from the SFC

Do you mean Bradley Kuhn of the Software Freedom Law Center (SFLC) < https://www.softwarefreedom.org/ >?

Bradley was formerly at SFLC, he is now at Software Freedom Conservancy (SFC)