Re: What is GNOME office?

[Daniel Veillard <Daniel Veillard w3 org>]

On Wed, Nov 15, 2000 at 03:15:05PM +0100, Xavier Bestel wrote:
> 
> Le 2000.11.15 15:08:52 +0100, Mike Kestner a écrit :
> > Dick Porter wrote:
> >  
> > > Surely "what is GNOME Office" should be "all GNOME applications that do
> > > officey things". If we have more than one candidate for a particular
> > role,
> > > then include them all. Choice is good (particularly when they all
> > interoperate)!
> > 
> > Unfortunately, this thought process doesn't work well in the current
> > GNOME corporate expansion.  Corporations that are considering investment
> > into GNOME will want to know which application is the "official"
> > component of GO, when two or more exist.  Companies need to assure that
> > their payroll dollars are being invested in the most productive manner
> > possible, and this means a single sanctioned suite in most cases. 
> 
> Do we start sacrifying projects because of $$$ ?

 If you believe what he said. I have yet to see this backed up by reality.
No corporation to my knowledge has made even the request to be "the blessed
application". And it looks to me that hackers on project A won't necessarily
work on project B just because it could be considered the "official" one,
which means that parallel development won't necessarily penalize the
project as a whole. And there is so many tradeoff possible when doing any
complex implementation that it is sometime difficult say that A is better
than B. An application suite perfect for a Gnome desktop will probably don't
fit on a embedded system, and vice-versa.

 Up to now deciding the "official" application was not too difficult, the
state of the implementation nearly always raised a clear winner, Nautilus
will superseed gmc for example, I think building the core list for 1.4
wasn't too hard, the only debate so far was whether gpaint should be added
along to gimp. Now if we have the difficult task to pick between 3 full
featured spreadsheet application, well I would rather consider this a
success of the Gnome project than a failure ! As long as we don't select
inappropriate or incomplete applications, this all sounds fine.

 Another important point is to allow applications to cooperate and be
interchangeable, IMHO the following points will be extremely importants:
  - define and use standard and interchangeable formats so that
    application A can be transitionned to B (and back !) without
    loosing time or data in the process
  - componentize (bonoboize) those apps so that they integrate nicely
    in the overall GNOME framework

 Just as an example one of the features I really enjoyed in Nautilus
release was the ability to easilly change or define the default browser
for a given set of pages (mozilla vs. gtkhtml). There is pros and cons
for both and I use them in different context. Once you have standardized
at the format level (HTML in this case) and the component level (Bonobo)
then suddenly the "default" choice ain't a lock in, you gain the
freedom to simply select the best app for you to do the job. Coders will
fight in that framework to get the best app and that sounds just perfect
to me !

Daniel

-- 
Daniel Veillard w3 org | W3C, INRIA Rhone-Alpes  | libxml Gnome XML toolkit
Tel : +33 476 615 257  | 655, avenue de l'Europe | http://xmlsoft.org/
Fax : +33 476 615 207  | 38330 Montbonnot FRANCE | Rpmfind search site
 http://www.w3.org/People/all#veillard%40w3.org  | http://rpmfind.net/