From kde-core-devel  Tue Apr 28 19:21:29 2009
From: "Brian J. Tarricone" <bjt23 () cornell ! edu>
Date: Tue, 28 Apr 2009 19:21:29 +0000
To: kde-core-devel
Subject: Re: KNotificationAreaItem
Message-Id: <49F75739.5090202 () cornell ! edu>
X-MARC-Message: https://marc.info/?l=kde-core-devel&m=124099986002873

Lubos Lunak wrote:
> On Monday 20 of April 2009, Aaron J. Seigo wrote:
>> hi :)
>>
>> there are several limitations inherent to the current xembed based system
>> tray protocol, including:
> 
>  There are more problems with the systray than just technical problems, as the 
> systray is rather broken semantically (or, to be more precise, nobody really 
> knows that the proper way to use it is).

I kinda think it doesn't matter.  The original spec authors might have 
thought they "know best" as to how the systray should be used, and sure, 
I suppose there are some clear (ab)uses that would be incredibly 
annoying to the user, but my philosophy is that if anyone finds 
different ways to use it that end up being useful to the user, then 
that's a fair use of the systray.

 > I'm specifically thinking
> the "the taskbar entry doesn't quite fit me for whatever reason, so I'll roll 
> my own in the systray" mentality,

There's a better solution to this specific problem: set up a mechanism 
to allow apps to add items to the taskbar entry's menu.  MacOS X 
supports this in the dock, and I'm pretty sure the Windows taskbar 
allows this as well, so toolkits that need to be cross-platform can 
expect this to work everywhere.

Of course, then you get into the "what happens if the user doesn't have 
a taskbar or doesn't have a taskbar that supports this?" questions... 
but this kinda boils down to the same "what happens if the user doesn't 
have a systray?" questions people probably threw out back when the 
systray spec was new.

	-brian