Re: Gnome : Aims/Focus issues

[Rodrigo Moya <rodrigo linuxave net>]

> Good point, but the ability to provide a relational view of data
> is critical to the Gnome project as a whole.  Perhaps the gnome-db
> could be broken up into three main areas
>    1. Provider related issues (talking to the core)
>    2. The GDA core (including XML/DTD representation of databases,
>       queries, recordsets, etc and the means of passing messages
>       between providers and clients)
>    3. Client programs (such as worksheets (gdafe classic),
>       report tools, dba tools, etc.)
> 
this is exactly the same organization we've got now.

> It might also be useful to say that a particular provider is
> CORE-vX.X compliant, and the same for a particular client.
> 
right now, each provider has a method for returning the version it was
compiled on (0.0.95, 0.0.96, etc), so this could be used to do what you
say. for the moment, we've got no problem about this since all the
providers support the same set of IDL interfaces, but it could be useful
as soon as some changes are made to the IDL files (some providers may
not be updated for this, for example).

> This is a HUGE task, remember.  The GnomeDb project is different
> from Matthew's excellent Orasoft.org, which is concentrating on
> pumping out products, but less concerned about universality,
> except as a nice-to-have.  That is Orasoft's right (and I am
> becoming close to them : just declaring interests) and we
> need people pumping out those tools : just as we appreciate
> the KDE types who lead the way before Gnome got off the ground.
> 
> However, comparing Gnome and KDE, you find that "traditionally"
> Gnome held back, cleaning up the architecture before "rushing
> to market".
> 
that's why i prefer GNOME over KDE. It's true that KDE has a bigger set
of applications, but this is gonna change as soon as all the GNOME
architecture changes are stabilized (OAF, GConf, Bonobo), which IMO will
mean the addition of hundreds (or thousands) of new applications.

> Gnome-db can and should be friendly with orasoft, the DBI/DBD
> teams, and others providing more specific solutions for the
> near term.  We should encourage the use of open-source solutions
> generally, not just those from gnome-db.  However, my opinion
> is that the Gnome-Db team should be focussed on the more universal
> and elegantly designed, as-few-compromises-from-perfection that
> typically describes GNU projects.
> 
yep, this is our focus!