Proposed: systemtray-spec

[Aaron J. Seigo]

On Saturday 07 August 2004 05:55, Havoc Pennington wrote:
> On Sun, 2004-08-01 at 01:25, Daniel Stone wrote:
> > On Sun, Aug 01, 2004 at 01:22:09AM -0400, Havoc Pennington wrote:
> > > On Fri, 2004-07-30 at 20:11, Daniel Stone wrote:
> > > > I'm proposing systemtray-spec with this mail.
> > >
> > > I vote no, the spec sucks, it's far too policy-free and thus not
> > > genuinely interoperable.
> >
> > Fair enough.
> >
> > Do we have any alternatives coming up for this sort of notification?
>
> We should, or at least fix what we have.

i'll second this one. having had to work with the implementation wonders of 
the current spec, i'd personally like to see the use of XEmbed DieDieDie. it 
just creates far too many problems with regards to constistency and 
flexibility (if anyone cares, i can enumerate these in gross detail on this 
list as i have done that elsewhere already one or twice =)

i've been mucking about with a DCOP based alternative that provides a much 
cleaner interface between the application and the system tray. if KDE moves 
to DBUS for KDE4 then it will be DBUS based instead (obviously, since the 
next version of KDE after the upcoming 3.3 is 4.0) and could well end up 
being a useful start to a new and improved system tray spec. it's one of the 
things i'm bringing to the table with me @ aKademy (the KDE world conf this 
month) for discussion. if there is interest is such an approach, i'll be sure 
to forward on progress, etc from aKademy to the FD.o group.

i also wouldn't mind hearing from those who won't be there (e.g. any of the 
GNOMEs on the list) what they'd like/need to see in a system tray spec.

-- 
Aaron J. Seigo
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43