[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: KDE Trunk based on Qt 4.6
From:       Tom Albers <toma () kde ! org>
Date:       2009-09-02 10:32:41
Message-ID: 3109570.fWPkWqCON3 () kde ! org
[Download RAW message or body]


At Wednesday 02 September 2009 11:39, you wrote:
> El Domingo, 30 de Agosto de 2009, Tom Albers escribió:
> > Op Sunday 30 August 2009 23:52 schreef u:
> > > On 2009-08-30, Tom Albers <toma@kde.org> wrote:
> > > > I'm sure we can make a cmake check which will give a big fat \
> > > > warning to
> > > distro's to not do that.
> > > 
> > > Which distros will ignore.  Qt is not only used by KDE, and Qt is
> > > expected to keep compability and no behaviour changes.
> > 
> > Thanks. Just gave a whole new look on Debian.
> > I naively thought they would take KDE serious and would not ignore such \
> > a warning.
> 
> You can't fully blame the packagers for this. Qt-people state that if you \
>  develop and test an app with Qt 4.x, it will run the same with Qt 4.x+1. \
> Then  KDE-people put a warning to not use Qt 4.x+1 because they haven't \
> tested this  version.
> 
> Now packagers have to choose whether they should agree with Qt devs or \
> KDE  devs, so they are going to upset someone anyway. Given that 4.x+1 \
> should  work, and if it does not it's a bug, siding with the Qt devs it's \
> the better  option in the long term: if it doesn't work they are doing \
> something wrong,  and they should know it to fix it and avoid it in the \
> future.

Well, the past has proven that QT+1 /can/ break stuff within KDE. Probably \
caused by workarounds we did in KDE to workaround bugs in QT. Those bugs \
were solved and now the workarounds fail. It does not matter why it fails, \
it matters that it /can/ fail.

And Debian announcing it would ignore our warnings hurts. Making a better \
4.x.0 release than we did in the past is therefor not possible, .0 will \
continue to be less than optimal and efforts to change that are made \
impossible by this behaviour.

Best,

Toma



[prev in list] [next in list] [prev in thread] [next in thread]