Notification spec issue: Ability to assign an icon *and* an image to a notification

[Jeremy Whiting]

On Thursday 25 June 2009 08:48:24 Lubos Lunak wrote:
> On Thursday 25 of June 2009, Aurélien Gâteau wrote:
> > Note that turning org.kde.VisualNotifications into
> > org.freedesktop.VisualNotifications won't help you configure the way
> > notifications are presented.
> > To get to configure how notifications are shown through KNotify, you
> > would need to make KNotify internal DBus interface the new spec.
>
>  Which incidentally would be a very nice outcome of this all, eventually. A
> good reason not to block the name.
>
> > At least the bubbles would integrate better into your desktop if we make
> > KDE adopt o.fd.Notifications.
>
>  I don't have any strong argument against supporting the Galago spec in KDE
> as a sort of a legacy support. I do have issues though with this discussion
> reminding me a lot of the first discussion those years back.
>
> > And you can (and should) complain to
> > FooApp developer for spamming you with useless notifications.
>
>  Right. I keep forgetting that there's always also the complicated way. I'm
> sure the developer will exactly agree with me on how FooApp should work.

I may completely misunderstand here and if I do let me know.  I've been using 
kttsd notifications in some apps for a long time (and have recently been 
maintaining kttsd itself). I can easily set up e.g. konversation, and kopete 
to send new message notifications to the kttsd, to be spoken out of my speakers 
instead of popping up in bubbles. I don't want or need konversation or kopete 
to stop sending these notifications.  I don't want or need all notifications to 
go to kttsd either.  I can, as a user, easily choose which ones I want spoken, 
and which I want bubbles, and which I want to ignore.  I think that's the 
beauty of knotification.  Therefore your suggestion to complain to developers 
of apps to not send so many notifications is foolish.  Different users want and 
need different amounts of notification from their apps (why are there loglevels 
in almost every daemon app that has existed including the linux kernel 
itself?).

thanks,
Jeremy

> > >> Suppose you are using a library which provides a class named Foo. This
> > >> library has been in use for 5 years, but you think the class really
> > >> should have been named FooBar. Would you ask the library maintainers
> > >> to break binary compatibility to get the class renamed?
> > >
> > >  This is such a bad analogy in practice that I even don't know where to
> > > start.
> >
> > Care to explain why class <=> DBus interface is a bad analogy? Both
> > provides an API you can call, both have backward-compatibility concerns.
>
>  I take that back, it's actually quite a good analogy:
> - in both cases, things may break if reserved symbols are used by something
> that should not use them
> - in both cases, there are technical means to do the rename while keeping
> backwards compatibility
> - in both cases, things often break more often than once in 5 years anyway
>
> > >> KDE developers use the KNotification class, other developers use
> > >> libnotify. They don't care about the DBus name. Why change just for
> > >> the sake of change, especially since not all parties consider this
> > >> change to go in the right direction?
> > >
> > >  If they don't care about the name, the question then should be 'Why do
> > > we need this discussion at all?'.
> >
> > Or maybe it should be "Why do some of you want to change it and create
> > more work for themselves and for others?"
>
>  To save us from more work, of course. If this was done properly in the
> first place, we wouldn't need this discussion now. If you want to know some
> of the reasons in more detail, please just see this thread.