Gideon

[Victor Röder]

Hi,

> On Friday, 10. August 2001 12:20, you wrote:
> > Hmmm, I think the changes between KDE 1.x and KDE 2 aren't comparable to
> > the changes that will be made between KDE 2.2 and KDE 2.3(?) based on
> > Qt3. Sandy is right when he says that we might lose beta users who
> > haven't the "Qt3 KDE" installed. I think matching Gideon to Qt3 is not
> > going to cause many work. So on my opinion we should wait until a certain
> > time (maybe when the KDE team ships the first beta/alpha version based on
> > Qt3) so that there is a user base that uses the "Qt3 KDE". At the other
> > side, Qt3 offers some interesting things. How to other projects (KOffice)
> > answer this problem?
>
> They all switch because the plan for KDE 3.0 is to release in January. The
> switch to Qt 3 will be in 3 weeks (first week of september). See Dirk
> Müller's mail on KDE 3.0 at lists.kde.org or kde-core-devel. Qt 3 isn't
> that revolutionary IMHO. That justifies the switch of us, too. Otherwise
> we'll run into trouble for many things:
> -the source incompatibility isn't estimated as far as it affects us yet
> -we will have to do the porting later which says we're using API methods
> which are away probably later
> -we're using admin to work at the autoconf/automake level which will switch
> the check to Qt 3
>
> Given that and the fact that the source incompatibility mostly affects the
> kde styles and that the extended API of Qt 3 is easier to use and we save
> time by switching right away instead of later, I think that are very valid
> reasons. See my other reply to Sandy.

Yes, but you have forgotten the people who want the test Gideon. Granted 
there is nothing to test in Gideon so that the only users of Gideon will be 
developers the next months. And those developers are usually "armed" with the 
latest Qt and KDE libs.
But as Sandy said: We might get into serious troubles with buggy KDE libs and 
other things. On my opinion we should wait until there exists something 
stable. Speaking of technical things (the new component model) and APIs. I 
think we do nothing wrong if we wait. But many things could go wrong if don't 
wait.

Bye,
	Victor

-
to unsubscribe from this list send an email to kdevelop-devel-request at kdevelop.org with the following body:
unsubscribe »your-email-address«