Re: [Nautilus-list] Nautilus user testing at MIT

[Miles Lane <miles megapathdsl net>]

Nicolas Mailhot wrote:


> There are extensive discussions on this subject in the
> gnome-gui list archive. You can look them up. This is a
> difficult subject since the usual solution (hourglass) is
> basiquely *WRONG* in a real multitasking environment.

I disagree.  What is wrong is blocking any additional
user interaction with the OS/desktop.  The hourglass
spinning is just a way to tell the user that "Hey, I'm
working on it!"  The whole issue of "System modal"
dialogs like Microslosh has is another matter entirely.

Displaying a visual cue for a desktop user that the OS
is taking some action has no connection to whether or
not the OS is running multiple tasks or even serving
multiple users.  At this point, it is probably safe
to assume that one user is looking at each system
display "head."  In the future, XFree86 will support
multiple heads with assigned keyboards and mice for
each head (okay, the keyboard part is hard and might
not happen for a while, but the assigned mouse thing is
easier).  As long as there is a graphical display that
is more or less dedicated to one user's control I/O
input devices, I see no conflict with having graphical
feedback for that users requests from the OS.  In fact,
since the VAST majority of the world's computer users
have been trained to expect this feedback, I'd say it
is obligatory that we also provide it.

	Miles