unity-design team mailing list archive

[Scott Kitterman <ubuntu@xxxxxxxxxxxxx>]

"Ted Gould" <ted@xxxxxxxxxx> wrote:

>On Wed, 2010-06-09 at 00:45 -0400, Scott Kitterman wrote:
>> >On Tue, 2010-06-08 at 18:41 +0200, Martin Gräßlin wrote: 
>> >> As mentioned in my open letter: I want to help you. I am willing to spend my 
>> >> time and expertise on this issue to ensure that we don't end with an utterly 
>> >> broken CSD library in GTK 3.
>> >
>> >Cool, that is great!  But, I'm worried that we're approaching this from
>> >different assumptions.  My reaction there is "let's figure out how to
>> >make CSD work" while yours seems to be "let's figure out how to get
>> >window decorations to work."  So, I guess my question is: If we assume
>> >that the window manager isn't going to draw decorations, how can we make
>> >that works as good as possible with KWin?
>>
>> Why would we ever want to do that?
>
>Because you want applications that are using client side decorations to
>work well under KWin and other window managers.
>
>You can take the stance of "Chrome sucks" or the "Radience GTK+ theme
>with CSD sucks" -- or you can look for ways to make them better.  If you
>choose the "sucks" route I personally believe that your list of things
>you dislike is going to get very long.

No. What I want is for Canonical to back away from this.  CSD is a broken concept that will destroy desktop consistency.  This discussion started with an appeal from Martin for an use case that required CSD. I haven't seen one yet. I've seen several cases of things that "couldn't be done without CSD" that are already supported with kwin and no CSD (the chromium top window layout for one).

Currently I'm unaware of any reason why CSD is needed and stacks of reasons why it's a poor engineering choice. 

If Canonical is completely wedded to CSD and unwilling to consider alternatives, please let me know and I'll stop wasting my time on this thread. 

Scott K