D-Conf

[Mikael Hallendal]

Philip Van Hoof wrote:
> On Sat, 2005-03-05 at 10:48 -0500, Zack Rusin wrote:

Hi,

 > Changes in the existing applications WILL be needed. Lots of. The KDE
 > applications will need MASSIVE changes. Try selling a generic
 > configuration system by telling them:

 > - You guys will basically need to rewrite 25% of your applications
 > - We won't need to do one single thing
 > - Cool don't you think?

So, it would be better to say, you need to rewrite most of your 
applications but to make you happy we will as well so we will both have 
to spend all of our time in rewriting things?

>>- We do not give a damn whether a name has d, g, h, z, u or any letter 
>>you can come up. And before someone asks, yes, that includes all 
>>encodings.
> 
> BTW. What I'm really trying to say is that we need to start working
> together and that I'm getting tired of non-technical discussions about
> the differences between KDE and GNOME and reasons not to work together
> based on non-technical decisions.

We already have started working together but you can't expect things to 
change over night. Both KDE and GNOME has a compatibility assurance to 
support and can't replace huge parts of their desktops within a major 
release. In GNOME this means we can't do anything that would make a 
binary incompatibility within the GNOME 2.X series. I don't know what 
rules KDE are using but my guess is that they are similar.

So for changes to go in it would mean that they would go into GNOME 3.X 
and KDE 4.X.

> Also: I KNOW that the real KDE developers aren't those kids that I've
> been describing. I'm fully aware of that. I do KNOW that the real KDE
> developers are professionals in that aspect.

And in such a decision it's ONLY the real KDE developers that count. So 
you don't have to please the hang-arounds by doing things to cover the 
real background of a library.

If you really want to work on this, do some real work. By looking into 
KDE's need in a configuration system, and also, look into whether GConf 
is everything GNOME need/want as well. There definitely is room for 
improvements in GConf, let's start in that end instead. Define the 
requirements for both desktops (and possibly others), see if GConf would 
be suitable for this instead of the other way around (let's see how we 
can push GConf).

Maybe we come to the conclusion that something new all together would be 
best. Something designed and modelled after our and KDE's experiences 
with current solutions.

Best Regards,
   Mikael Hallendal

-- 
Imendio AB, http://www.imendio.com/