Steve Dodier wrote: > I dont think we'd gain much by trying to guess the behaviour of the > user, because it requires being able to imagine absolutely any > activity / behaviour (s)he can have. My HMI courses were poor but i've > been warned that it wasnt an feasible task, back then. We would always > forget an usecase and disturb someone who's focusing on his work. > > That's why i think it would be wise to only delay notifications when > it is obvious that the user needs his(her) eyes somewhere else. The > best approach for me consists of making notifications that are easy to > spot on the screen but that someone focusing on something else can > easily not notice. I think notify-osd is approaching this state now, > and we should focus on improving its last bugs, this will likely > receive more praise from users ;)
+1 I think we can identify a set of "fine motor control" activities, as Celeste described, where you have "brittle state", like driving your mouse through a menu or a dropdown listbox. A loss of attention and consequent slip there is really irritating because you have to "start over" on the brittle work. We could delay a while on those actions, IMO, usefully. But beyond that it becomes a mugs game to guess intent. Mark
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Mailing list: https://launchpad.net/~ayatana Post to : ayatana@lists.launchpad.net Unsubscribe : https://launchpad.net/~ayatana More help : https://help.launchpad.net/ListHelp
