The future of GNOME.

[Miguel de Icaza <miguel nuclecu unam mx>]

Hello guys,

   This is my proposed roadmap for the future of GNOME development:

   First of all, given that we managed to ship GNOME 1.0, the API of
the libraries can be considered to be stable.  This means that we will
only be releasing gnome-libs for the sole purpose of bug fixing,
updating the translations and updating the inline documentation.

   Since both GTK 1.2 and GNOME 1.0 are out in stable form, the
release of the various GNOME components will be now pretty much
decoupled.  We will no longer have to keep all of the released
packages in sync.

   We might still release a bunch of them under a same name or label,
to distinguish the dates on which they are released.

   In the short term, we will concentrate in making GNOME 1.x
derivatives better featured and stable within the current framework.
This means, we will not be doing any massive redesign of the
internals.

* Developing GNOME

   If you want to start work on new code, I urge you to create a
development branch of your project and put all untested code there.
This will allow us to continue making releases of fixed stuff from
HEAD.  

   This is a bazaar-ish project, so we want to make the new
improvements of GNOME and the bugs fixes available to the public as
soon as they get tried and tested.  Major releases will most likely
include major new components. 
 
* The Bonobo document and component model

   I need to finish Bonobo.  I am very close to have a working version
of it.  Currently it does provide a framework for embedding components
into containers and the container framework.  There are a number of
things to finish: menu negotiation, some simplification of the process
(the user currently needs to use CORBA in a couple of paths, and I
plan on removing this), storage support. 

* The File Manager Future

   An exception to this is the new file manager.  We will start
re-creating the file manager from a new more GUI-aware framework and
we will be pulling existing code from the current "mc" module. 

   My personal plans include:

	- Finishing Bonobo
	- Integrate Bonobo this Gnumeric
	- Finish the printing code for Gnumeric.
	- Work on the new file manager foundation.

   During this time, I will help hackers with their questions and with
supporting Bonobo in both container and embedded forms for their
software. 
 
   Bonobo is particularly important as it will provide off-the-shelf
component programming and will be used in the new file manager to
manage the "views" of specific mount points in the file system.

* The PowerPoint clone project

   We do think that we have the tools to write the foundation of a
usable PowerPoint clone in about 5 days.  We will be dividing the
tasks for writing such a project and the idea is to finish the program
in a week.  So far George, Federico, Chris and I have shown interest
in working in this 5-day project.  

* Calendar server

   We do still need a calendar server.  Finishing the calendar server
is not a big task, it just needs a few weeks worth of hacking.  At
some point the vCal code needs to be converted into a storage backend
only, instead of being the main storage abstraction it currently is. 

Miguel.