Gnome Office summary from the Gnome Summit.

[Martin Sevior <msevior mccubbin ph unimelb edu au>]

Hi everyone,
	   Here are my notes fromt the Gnome Office session of the Gnome
summit in plain text form. Since many of the people involved in these
projects could not make it to the summit I thought it was important
to write everything up. I think the the number one thing Gnome Office
needs right now is a person committed to working and advocating
integration between the various projects.

Now that the Gnome 2 platform has been delivered maybe some people who
have been concentrating on "platform issues" might like to look at these
"integration issues". We came up with many ideas of how we can do better
than the competition by integrating these large applications right into
the desktop environment.

Cheers

Martin Sevior

--------------------------------------------------------------------------


Notes from the discussion at the GNOME Office Workshop.

In the IRC discussion we had a goal to make a substantial coordinated
release of GNOME-Office by December. We want these application to be
ported to GTK2/GNOME2 by then.Is this still realistic?


By years end Evolution/AbiWord/Gnumeric will be serious competition for
MS Office in features, be at least as nice to use, and of course be fully
integrated into the GNOME environment.


It is worth noting that the creation and integration of a top notch
Office Productivity package into a Linux desktop is a really, really
serious deal for MicroSoft. They simply cannot afford to give away their
Office Productivity suite - it generates at least 40% of their revenue.
At the same time a MS based corporate desktop costs businesses around
$1000 per seat. Our success will substantially lower costs for businesses
integrated into the GNOME environment.


It is worth noting that the creation and integration of a top notch
Office Productivity package into a Linux desktop is a really, really
serious deal for MicroSoft. They simply cannot afford to give away their
Office Productivity suite - it generates at least 40% of their revenue.
At the same time a MS based corporate desktop costs businesses around
$1000 per seat. Our success will substantially lower costs for businesses
and provide a compelling case for a change.


_Status of GNOME-Office Applications_

_Evolution _

New features planned for a November release followed by a GTK2/GNOME2
port next year.


_AbiWord _

New features: Tables, Footnotes, EndNotes, Revision Marks, Auto Font
configuration/detection, Anti-aliasing, Hidden text, Document properties,
Import/Export filter improvements, Bonobo container for embeddable
objects. Gtk 2/Gnome port, Maybe "plumbing plugin" to do useful things
with interesting text strings (UPS slips, email addresses, Instant
messenger addresses). - beta in a couple of months, Stable release by
December.

_Gnumeric _

        Embeddable bonobo object, GTK2/Gnome 2 port. Stable release by
December.

_MrProject _

        ??

_GAIM_

Separate library backend from Front End. Make it usable from any apps
that want Instant Messenger - maybe via a "pumbing" style infrastructure.

_Agnubis_

        ??

_Gnome-db_

??

_Guppi_

??

_GtkMathView_

??

_GIMP_

??

_Dia_

??

_Sodipodi_

        ??



All the Gnome Office applications are large programs. We all have lots of
work to do to make our own applications as good as they can be. However
we also realize that each project gains from using features from other
projects. We also all gain from a common look and feel from an integrated
look among our projects. So we want to find ways to integrate these
projects to make them easily usable and develop synergy between them.

The classic example is embedding components such as AbiWord in Evolution
to read Word Processing documents or Guppi in Gnumeric to display charts.
Certainly we need to continue this.

Apart from anything else these helps in "Marketing" . Users now expect
Office Suites to have these integrated features.


Some suggestions for integration.

* Have a gnome-office panel?shell? to launch these applications.

* Copy and paste (maintaining formatting) easily from application to
  application. (Concensus was to use (X)Html to X clipboard?)

* Some wizards to whip up useful applications from users.
  (eg Form letters, envolope printing)

* Extract data from Gnome-db/ addressbook/gnumeric/calendar to generate
  letters.

* Extract emails from evolution to write documents.

* Embed Guppi plots in gnumeric/AbiWord

* Embed spreadsheets in AbiWord

* Use AbiWord to generate reports for MrProject

* Use AbiWord to generate reports form gnome-db

* Use the interesting "Plumbing" technology to do useful things from
  arbitrary text strings. eg Click on Hyperlinks, UPS address slips,
  email addresses, instant messenger addresses, file names

* Integration with Galeon/Nautilus - display Office (MS documents, OO,
  WP, *) in the web browser/ Nautilus.

* Integration with Nautilus - display contents of MS word/ Word
  Perfect/SO/OO/HTML/(everything else) documents inside Icons as is
  currently done with text docs. We could do this rather easily with
  AbiWord and would be a genuinely new feature on the desktop.

* Other ideas?

What I'd like is for people to work on is integration between these big
applications. The people who work on these big office apps normally have
their hands full making their own applications work. There is a big
learning curve for new hackers on these big applications. On the other
hand new hackers can make very useful contributions on the glue code
between these apps and in particular a plumbing style architecture.


_Some Action Items._

* Gnome-Office really needs someone to lead these integration issues and
  try to herd cats into useful activities. This person would promote the
  usefulness of working on Gnome Office applications (maybe particular
  integration projects).

* X clipboard format (use (X)html to copy and paste between applications?

* Advocacy for Gnome Office integration projects on gnome-love.

* Work towards cohert dialogs - Gnome 2 UI guidelines.

* Look through the UI guidelines to evaluate our interfaces.

* Menu label/orders

* Glue code/wizards

* Get interfaces to data items doing useful things (especially now
  AbiWord has tables)

* Implement libgsf for Gnome Office apps.