Has GNOME rejected Canonical help?

Dave Neary

March 7, 2011

Through the fall-out from the Unity decision [ http://blogs.gnome.org/bolsh/2010/10/25/ubuntu-to-move-to-unity-as-default-desktop-for-11-04/ ], and now the fall-out from the packaging of Banshee on Natty [ http://www.networkworld.com/community/canonical-alters-banshee-agreement ], I have repeatedly read Canonical & Ubuntu people say “We offered our help to GNOME, and they didn’t want it”.

Exhibit #1 [ http://philbull.livejournal.com/58459.html ] :

For starters, some people in the GNOME community moan about how Ubuntu doesn’t pull its weight upstream.They then make it difficult for Ubuntu-y folks to contribute things upstream.

Exhibit #2 [ http://derstandard.at/1280984081593/Ubuntu-We-have-no-plans-to-fork-GNOME ] :

For the app indicators we also had a lot of community involvement, it was based on a Freedesktop.org spec, worked on with consultancy from KDE, we invited GNOME developers to participate in the Freedesktop discussion and proposed them to the GNOME community for inclusion, but it’s not up to us, if they take it or not

Exhibit #3 [ http://itmanagement.earthweb.com/osrc/article.php/12068_3926371_1/Jono-Bacon-Defends-Ubuntu-An-Insiders-Perspective.htm ] :

Where tensions between Canonical and GNOME have occurred, according to Bacon, is in Canonical’s desktop innovations for improved usability, such as the Ayatana indicators [ http://www.markshuttleworth.com/archives/347 ] for sound and social media, and the new Unity desktop, all of which were submitted to GNOME and rejected, leaving Canonical to develop them outside the GNOME project. [...] Asked whether Canonical could have developed its usability modifications within GNOME, he replies, “To be honest with you, I don’t think it could have been done. The fact that nothing’s been accepted is a pretty reasonable indicator that the two projects have widely different directions.”

Exhibit #4 [ http://www.markshuttleworth.com/archives/611 ]:

We committed to build Unity [...] because we had ample reason to believe that the trajectory of the alternatives was going to fail. And it did fail – Gnome 3 looks much more like the vision we painted with Unity than the original vision [...] I am sorry that a few Gnome leaders have blocked Gnome’s adoption of Unity API’s, and the stress that will cause, but I feel proud that we had the guts, and the capacity, to design and deliver something wonderful.

I have seen and heard this mentioned by others too, but cannot find any others right now – additional pointers in comments would be welcome!

So – given that GNOME is a project which scores very highly as being Open By Rule [ http://webmink.com/2011/02/21/is-gnome-open-by-rule/ ] (disclosure: I put together the evaluation of GNOME for Simon), I thought I would go back through the archives and see how true this was.

Looking at what was actually proposed for inclusion in GNOME from Ayatana work, libappindicator was rejected [ http://mail.gnome.org/archives/devel-announce-list/2010-June/msg00001.html ] because (quoting directly from the release team’s decision):

it doesn’t integrate with gnome-shell
probably depends on GtkApplication, and would need integration in GTK+ itself
we wished there was some constructive discussion around it, pushed by the libappindicator developers; but it didn’t happen
there’s nothing in GNOME needing it

I went back to see where the discussion happened for the libappindicator proposal [ http://mail.gnome.org/archives/desktop-devel-list/2010-February/msg00036.html ]. There was a discussion, some over & back, Ted was (as usual) forthcoming & helpful, and things appeared to be moving approximately in the right direction. There were some issues over copyright assignment, and the discussion petered out. No feedback I could see from the GNOME Shell team – positive or negative – to depending on the library.

Now, this is hardly ideal. I would love to see debate on why there wasn’t a more in-depth debate on using libappindicator in GNOME Shell. Was this ever proposed? If so, where? I can’t find any reference. Was there any reaction other than “we don’t think it’s an issue” to the copyright assignment issue? Perhaps there was a lot more discussion in another forum that I haven’t linked to – on the release-team list, on IRC, or elsewhere? Comments, please!

I would love to point to other instances of work which has been proposed upstream from Canonical and which has been rejected, but my (admittedly, brief) search has not turned up much useful stuff. I can’t find any online reference to displeasure with the GNOME Shell vision, or proposals of alternatives, nor can I find situations of “Paper Cut” patches being rejected because they were from Canonical or Ubuntu. In fact, the one reference I found to the UX hackfest in 2008 from Mark seemed quite positive about the whole thing [ http://www.markshuttleworth.com/archives/223 ].

There are apocryphal stories about patches submitted twice by different people before they were accepted, other stories about people being “impossible to work with”, design feedback being ignored, and more – I would love to see some evidence of this, or some documented criticism from 2008 of some of the GNOME Shell design documents. I hear often that some of the design decisions were unacceptable, but ask which ones, where the discussion took place, or how much effort was spent trying to get things changed, and hand-wavy “lots of stuff” type answers [ http://www.markshuttleworth.com/archives/611/comment-page-1#comment-345825 ] is what you get back.

I would really like to shed some sunlight on this – if we do not have publicly archived references to places where these disagreements have happened, then there are a couple of possible conclusions we can draw: either insufficient effort was made to collaborate, or the effort was made, and GNOME Shell is not sufficiently transparent for the developers and designers to be accountable.

So please – pile in on the comments. I want to know of instances when GNOME has (allegedly) refused contributions or help from Canonical, with links to Bugzilla, mailing lists, even IRC logs or wiki pages. Let’s get to the bottom of this & see if we can’t solve the problem.

Updated to clarify that the reasons for rejecting libappindicator were not mine, but were copied from the release team decisions, after reading Aaron Seigo’s response [ http://aseigo.blogspot.com/2011/03/collaborations-demise.html ]

3:14 pm