When running a fullscreen (not maximized but really fullscreen) applications usually people want only to focus on that app and nothing else, so not displaying notifications makes sense (think of videos, presentations, games ...).

(In reply to comment #1)
> When running a fullscreen (not maximized but really fullscreen) applications
> usually people want only to focus on that app and nothing else, so not
> displaying notifications makes sense (think of videos, presentations, games
> ...).

Yes, people use full-screen not just for maximizing the pixels, but also to remove any clutter so they can focus on the task at hand.

The whole idea of a discrete notification system is it allows you to become aware of events you may be interested in as they occur, with minimal disruption.

Wouldn't you want to know for example, that your hard disk started failing, some of your web sites were down, or that long running task you've been waiting on has finished, at the time that event was known, rather than some time later when you happen to check what notifications you've missed. Even worse, if the notification was transient you'll never even know it happened, I don't think that's a good thing.

Notifications have an urgency level so when it comes to videos, games, and especially presentations, there should be the ability to disable display of notifications equalling or below a given level, for some criteria. During that period transient notifications should probably become persistent. Equally transient notifications should probably at least remain on-screen, if not become persistent, when the computer has been idle for some time. But these are separate issues.

Discrete notifications and full-screen applications shouldn't be mutually exclusive, especially as IMHO full-screen apps and distributed services are likely to be used a lot more in the future. Full-screen editors, readers, browsers, ... Are just as desirable as full-screen videos/presentations etc...

As far as I understand it not showing notification while in full screen was not a design choice but an artifact of us hiding all shell chrome to avoid being on top of the fullscreen window. I think showing notifications is orthogonal to being fullscreen. Full screen + busy is the state that doesn't show them.

Although an artifact, the argument not to display notifications is being made more as a side effect of that artifact rather than being based on sound design principles.

Is the argument for allowing full screen to hide notifications stronger than the argument for ensuring they're displayed?

I can see a limited argument for hiding notifications, but I think it is weaker than the argument for showing them and/or allowing the user more control of this behaviour.

I'm not sure there's a reasonable argument for hiding critical notifications without the user setting that behaviour.

I guess the importance of whether notifications are displayed when in full screen depends on whether you believe running apps in full screen is likely to become popular, even normative. I think it probably is, especially given the way people use their mobile devices, in which case the value of the notification system is greatly diminished.

Hm. It appears that we agreed on this behavior (notifications should be visible in fullscreen) a long time ago, but one of the patches never got committed for some reason, and then the bug got closed.

Fixed in git

*** This bug has been marked as a duplicate of bug 608667 ***

Ah, great I didn't find bug 608667, thanks for the update.

(In reply to comment #6)
> Ah, great I didn't find bug 608667, thanks for the update.

I assume you reopened the bug by accident ...

*** This bug has been marked as a duplicate of bug 608667 ***