Re: GUADEC Results

[Hubert Figuiere <hfiguiere teaser fr>]

On mercredi, avril 11, 2001, at 05:38 , Almer S. Tigelaar wrote:

> Hmmm, in that case the question "what is gnome office" remains
> unanswered?


OK. Actually this is my first post on the list but I plan it to make it 
valuable by giving my own opinion on the topic.

NOTE: licensing issues will NOT be discussed here.
NOTE2: any technical error is solely comming from me. Corrections are 
really welcome.

What is Gnome-Office ? This is a real good question. As I only 
subscribed today, I have read the whole list archive and I was 
dispointed to not see any answer for this question. So I'll try to 
provide answer, that is my own  personnal opinion of what Gnome-Office 
should be.

Gnome-Office should be a set of software that run under GNOME (but not 
specifically exclusively, more on this later) to provide a complete 
productivity suite, aimed at business user (user in a corporate 
environment). See it as a software distribution like the various Linux 
distributions: a set program developped by different users put together 
in a consistent way.

What program should it have ?
-----------------------------

Basically Word processor, Spreadsheet, presentation software, personnal 
database, messenging, web browsing, PIM, Plotting and Diagramming.

Why corporate environment ?
---------------------------

Because the target is corporation. They are actually the people that 
rules the IT world by chossing Microsoft products, that is why most 
users take Microsoft products for home (and most often don't buy them 
either). People that does use the same they have at work are mostly 
geeks or Apple[*] fanatics.
If you bring gnome office into corporations, you win a battle.

How should it be build ?
------------------------

Component of the gnome-office suite should be quite similar to what 
GNOME is: a set of program and libraries that bring foundations for this 
programs. Some are part of GNOME already, others aren't. These 
components should be consistent.

Who will develop them ?
-----------------------

Good question. Volunteers or companies, depending on who commit to do 
the work. See the GNOME model. But it is very important that we get a 
central organization to provide project management. And also (paying) 
support services should be provided.

What do we already have ?
-------------------------

Note: I am not counting OpenOffice currently as I don't feel it yet like 
a GNOME application. This is a technical matter.

-Word processor: Abiword. Abiword is one of the exceptions here as it is 
a cross-platform product that have a GNOME front-end of good quality, 
but also run on BeOS, QNX, Windows and soon on MacOS (X). Curiously 
Abiword is a component of Abisuit whose little brother, Abicalc, has 
been completely eridacted from the surface of the earth.

-Spreadsheet: Gnumeric. This software is currently developped by Ximian 
that has already committed to bring GNOME Desktop to your computer :-)

-Presentation software: none.

-personnal database: Gnome-db.

-messenging: Evolution, also from Ximian and Balsa

-web browsing: Galeon, but it depends on Mozilla engine (aka Gecko) 
which is not the best GNOME solution. Given currently difficulties of 
implementing a web browser today, I won't blame them :-) They still do 
better than Mozilla itself.

-PIM or groupware client / calendering, eventually with PDA 
connectivity: Evolution (again), Gnome PIM (already part of Gnome), all 
using Pilot-link (independent package, integrated with Gnome in 
Gnome-pilot)

-Plotting: GUPPI seems to be the only candidate

-Diagramming: Dia is here.

And the other ?
---------------

-Gimp, Sketch, Sodipodu belong more to an "artist" suite, even if they 
can (and should) use Gnome-office libraries.

-GnuCash is a little bit different as it is targetted at personnal or 
SOHO user, not really at corporate user. It is also encouraged to use 
and integrate seamlessly with the rest of Gnome-Office. Not that I want 
to black ball it out of Gnome-Office. (actually I'm a GnuCash user too)

So what is missing ?
--------------------

A presensation software. We will see later how we can easily build one.

A import/export architecture that will allow do have common code to 
import foreign document within Gnome-office. Just because there are 
composite documents that may not be imported by a single program. I 
mainly think about StarOffice, AppleWorks and other integrated office 
suites, as well as Word document that may include Excel spreadsheets and 
charts, etc. Note that to succeed, the number of file format supported 
must be as wide as possible.

A programming language that Joe Average or a church secretary can use, 
and that is compatible with already established language like VBA. 
Gnome-Basic is aimed at this. BTW this should NOT be the only solution 
as not everybody (including myself) feel confortable with Basic.

Shall it run exclusively on GNOME ?
-----------------------------------

The answer is difficult. Given the name, it should run on GNOME. But I 
don't think If we really want Gnome-office to be successful, 
Gnome-office should be able to run on *as much platforms as possible*. 
On this point, OpenOffice is already forware because it runs on Linux 
x86 (soon PPC), Solaris, Win32 and MacOS X. And the crossplatform 
availability, mostly on Windows, is a real "selling" point. Lot of 
corporation have dropped non Wintel computer of users desk, and 
corporation that require UNIX Workstation for some purpose still have 
Windows requirement for Microsoft Office <grin>.

I have seen this in a company I previouosly worked: Solaris developpers 
had a SUN Workstation for work AND a Wintel PC for messenging (they did 
go within some proprietary Windows only messenging system that give the 
clueless level of the IT director on this matter) and office needs (they 
choose Word as the standard for document publication, not counting on 
drawbacks). I'm pretty sure that is they had a good alternative 
solution, they would have gone with it.

So I must say that Gnome-Office should not be tied with GNOME, and I 
don't think porting GNOME to non X11 platform will be the solution to 
everything. And Gnome-Office can help bringing Gnome to the desktop, not 
the opposite way.
Cross-platform development should not be done as in OpenOffice, but more 
as in Abiword. The goal is to choose the good balance between 
development cost for a new platform (low on OpenOffice) and the platform 
integration (high on Abiword). I think that with Abiword the goal is 
pretty well reached even if better could be done (that is just a matter 
of polishing the whole framework).


How about groupware ?
---------------------

Groupware is an important wizzbang feature of corporate information 
system. Groupware is mostly calendar and contact sharing, and e-mail. 
While Evolution is supposed to provide the client, I feel that we are 
still missing a feature complete server that will help Gnome-Office as 
it will help having them. Currently there are 2 commonly established 
server suite: MS-Exchange and Lotus Notes. The first is Windows only 
(even the Mac client is a piece of junk that prevent from using it 
completely) and the later have client on Linux and other UNIX but is 
still closed on every side (not mentionning the disrespect of standards 
protocols like for the mail). None of them use open standards. So we 
have no chance to be able to integrate with them. A server solution is 
crucial.


I hope that you were not bored by this long first post. Comments are 
really welcome.


Hub


[*] insert your favorite non Microsoft OS here.