Module decisions for GNOME 2.18

[Vincent Untz <vuntz gnome org>]

Hi everybody,

The release team had a meeting this week. Among other things, we
talked about the proposed modules for GNOME 2.18. Here are the decisions
that, in our opinion, reflect the consensus in the community.

We would like to thank everybody who was involved in the discussion, and
also the authors/maintainers of the proposed modules.


Short summary
=============

Retracted:
 GnomeScan
 tracker
 monodevelop

In:
 devtools suite
 seahorse
 nm-applet
 devhelp
 glade3

Out:
 gnome-main-menu
 anjuta

Details for In & Out
====================

 + devtools suite
  - general consensus within the community that we want this.
  - we can build it with small steps, so no need to wait for a complete
    solution to launch it.
  - the release team hopes having the suite now will also help getting
    more traction for work on the other tools that could be included in
    the future.

 + seahorse
  - positive feedback from the community.
  - maintainers really did a fantastic job to integrate GNOME (adding
    requested features, getting usability reviews, and more generally
    interacting with the community).
  - since seahorse and gnome-keyring-manager are sharing use cases, it
    seems a good goal for GNOME 2.20 to integrate all missing features
    in seahorse and deprecate gnome-keyring-manager (since seahorse is
    more actively developed and maintained). This needs discussion with
    developers from both modules, though.
  - maintainers really did a fantastic job to integrate GNOME. We wanted
    to mention it twice :-)

 + nm-applet
  - positive feedback from the community.
  - we don't introduce an external dependency on NetworkManager, but we
    will need to mention that nm-applet needs NetworkManager (and
    probably a specific version of its API) to be useful.
  - from a release engineering point of view, things would be really
    easier to manage if NetworkManager and nm-applet were splitted.
    Maintainers are encouraged to do so before 2.18.
  - the release team also encourages people to work towards integrating
    the NetworkManager API in an easy-to-use way in our stack.

 + devhelp
  - positive feedback from the community.

 + glade3
  - not a lot of feedback, and the feedback was mixed.
  - having an interface builder in our devtools suite would be most
    useful, and it appeared from the discussions that glade3 is the best
    option there.
  - the release team thinks it's really important that an interface
    builder included in GNOME deals with the current glade files and
    with the future GtkBuilder format, and the work from the glade3
    maintainers on GtkBuilder encourage us to believe that accepting it
    is a safe choice.

 + gnome-main-menu
  - lots of feedback, including strongly positive and strongly negative, with
    the majority currently being negative.
  - no consensus on how this module should be included (should it replace
    the current menu? Should it only be an option?), while this question
    plays a central role when considering inclusion of this module.
  - main reason for rejecting gnome-main-menu is the lack of integration
    of its development with the community: it is not integrated with our
    main developer build tool (jhbuild), there are no release on GNOME
    FTP and thus there were no releases done during this cycle, more
    communication should have happened with usability people and
    maintainers of other relevant modules.
  - for the future, if gnome-main-menu is proposed again, it's quite
    important to discuss what we want to do with the menus (what should
    be the default? Do we need more than one menu layout? etc.). This is
    more than about inclusion of a module like gnome-main-menu, it's
    about one part of the desktop core.

 + anjuta
  - the current 2.x version is in development and can't go in.
  - the 1.2.x version is not maintained. We can't accept a non
    maintained version in a suite.


Vincent

-- 
Les gens heureux ne sont pas pressés.