Re: notification enhancements

[Havoc Pennington <hp redhat com>]

Hi,

Christian Hammond wrote:
> You say there's no way imaginable to use libnotify, but I'm curious why? 

It just won't work. Look at the Mugshot notifications, look at what 
libnotify supports. (If you don't want to install mugshot, just look at 
someone's page on mugshot.org, the local client notifications are about 
the same as the stacker blocks on the web site, except a little more 
complex even)

It isn't just a graphic design issue; Mugshot has all kinds of links and 
thumbnails and a chat log and other stuff in the notifications. It would 
be a substantially worse user experience if reduced to only title + 
summary + some buttons.

Another aspect is that Mugshot has the "stacker" concept where the 
notifications are one block out of a stack of blocks, but you can also 
view the whole stack either in the client or on the web site. With 
libnotify we couldn't do this either. It's important that the blocks 
look similar in popup form, and in the client and web versions of the stack.

Right now there's a tradeoff of course, where the Mugshot fancy 
notifications collide oddly with other notifications, but that's what 
I'm posting about fixing. I don't want to throw the baby out with the 
bathwater, though.

> Why can't this use libnotify? Same reason as Mugshot? Or is this a part 
> of Mugshot?

This one is just a short list of features not in the libnotify protocol 
(bold text in buttons, for example). It is less "extreme" than mugshot.

>     A simple extension of that API would also provide calls or signals to
>     detect idleness and fullscreen app / screensaver. Or perhaps it's
>     simpler than that - maybe when an app is fullscreen or the user is idle,
>     the daemon claims "I am showing something" and thus blocks all other
>     notifications.
>
>
> Not sure I follow this one.. If an app is fullscreen or hte user is 
> idle, notification-daemon should claim it's showing a notification? Why?

I was saying first, you could have methods like IsIdle() or 
IsFullscreenActive()

but then second, maybe instead of having those methods, just have 
notification-daemon lie and say there was an active notification in 
those cases. Then other apps would not show a notification.

>       - take over the notification service from notification-daemon and
>         provide the same D-Bus API
>
>
> This seems like a bad idea. We can do better. notification-daemon is 
> pretty well established and people are used to seeing notifications, so 
> suddenly pulling them out from under the user is probably undesirable.

The point would be to have mugshot display them instead; the user 
experience would be unchanged perhaps, if we got ambitious and 
cut-and-pasted or librarified the theme code.

Havoc