New Platform - GNOME-productivity.

[msevior physics unimelb edu au]

Hi Everyone,
             In addition to our GNOME-platform, GNOME-desktop and
GNOME-bindings collections released every 6 months, I propose
we include
a GNOME-productivity software collection as well.

I personally guarentee that GNOME-productivity would at least include
AbiWord and that there will be stable and improved versions of AbiWord
every 6 months for the forseeabe future.

At the current rate of progress we are well on track to offer AbiWord-2.4
for GNOME-2.12.

I believe that developers of other productivity projects will also pledge
to provide stable releases in synch with the GNOME 6-month release cycle.

What do we as GNOME developers and proponents get from this platform?

1. First and foremost a new cutting edge for everyone to hack on. As has
been pointed out recently, we all enjoy creating things with our minds
through software. The GNOME-platform, GNOME-desktop and GNOME-bindings
projects are mostly moving into maintainence mode. GNOME-productivity
provides a vast range of opportunities for hackers to try new stuff and go
one better than proprietry software.

2. By having our own productivity platform under our control we can shape
a new desktop experience far beyond what Apple or Mircosoft can do at a
far lower price point. We already index MS Word documents via Beagle and
wv (the MS Word import library used by AbiWord) and we have a proof of
principle we can do the same for PowerPoint through GOffice. Both Apple
and MS require at least an $500 per desktop for productively software,
which in the case of MS on Apple, can never be as fully integrated as what
we can do. By developing GNOME-platform we can significantly improve ease
of use to our users at a fantastic price point. (Well zero capital outlay,
of course there will be on-going maintenence costs which the GNOME
sponsers may want to bid for.)

3. It's clear that OO.o has been a failure at attracting grass roots
hackers. It is also obvious that OO.o very much has it's own agenda which
is not neccessarily that of GNOME's.

4. The GNOME-productivity code is far more accessible than that locked in
OO.o. Both Novell and RedHat have fulltime OO.o hackers but neither chose
to use OO.o software for their MS compatibility needs (Beagle or Evince.)
Another point of comparison is my own project, AbiWord. The AbiWord
code-base used for it's Unix FE is approximately 200-300 KLOC. This is an
order of magnitude smaller than OO.o-Writer yet we do not have an order of
magnitude fewer features. Our project can be built from scratch in 5
minutes on a modern computer. The barrier for contribution is similarly an
order of magnitude smaller.

Why propose this now? Well frankly until now I did not think that AbiWord
had the maturity to anchor the job of document production that is the
essential task of a modern information worker. However this is no longer
the case. AbiWord-2.4 will have the ability to render and print arbitrary
content within it's documents as well as providing most of the features
required by such workers in their daily use of word processors.

In addition the GOffice project is really picking up steam and contains
lots of useful code which will be of interest to many other projects.
Gnumeric keeps on getting better and better and other projects may also
wish to join.

As I mentioned in an earlier email we have a proposal, GOCollab, to
develop a real-time collaboration tool which, together with our existing
collaborative programs, will leap-frog the currently accepted norm within
the work-place and amongst students. I have already received some great
feedback and suggestions for improvement for GOCollab already. I will
shortly update the document describing the design to incorporate these
ideas.

I purposely do not propose to call this platform "GNOME-Office" because I
do not think we should be locked into thinking the old standards defined
by the marketing men at MS should dictate the useful applications that
users want or need.

Yours sincerely,

Martin Sevior