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

List:       kde-core-devel
Subject:    Re: Repositioning the KDE brand
From:       nf2 <nf2.email () gmail ! com>
Date:       2009-07-02 0:22:07
Message-ID: aaed7a190907011722y643bfe14k99d4261974eb83a6 () mail ! gmail ! com
[Download RAW message or body]

2009/7/1 Aurélien Gâteau <aurelien.gateau@canonical.com>:
> nf2 wrote:
>
>> * System (or Platform, or infrastructure) technologies like ioslaves
>> or kwallet: While KDE has been the "market leader" for a long time,
>> comparable technologies have caught up or have even become "cooler".
>> Therefore KDE could benefit from "incorporating" them. The "Platform"
>> should move to the layer below KDE - things that are eager to be
>> shared by everyone. You won't love me for that, but honestly: That
>> part of the world should be ruled by GLib and GObjects...
>
> I don't think anyone with experience in Qt and QObject would like to
> rewrite his work with GObject. I had to work with this and it was really
> painful. At least in plain C. It's probably nicer when using higher
> level languages.
>

Handcrafted in C it's painful, but if you view it not just as a
C-language add-on, but rather as a kind of object-oriented runtime,
it's quite fascinating... The nice thing is the kind of dualistic
nature: You can use GObject based APIs just like traditional C APIs
straight away from C or C++, but with introspection and typelibs you
can also create bindings for dynamic languages or even a high level
compiled language like Vala (which is basically a C# dialect) on top.

I wouldn't slag off such a technology as "inferior", like Aaron does.
It's a masterpiece of software engineering the F/OSS community should
be proud of.

Norbert


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