Le 5 mars 06 à 12:39, Damien Pollet a écrit :
Hmm, you're probably right... still, that duplicated title bothers
me :/
If the menus work using clicks only, then a drag of the menu title (in
the menubar) could tear it. The menu title becomes the teared menu's
titlebar... that's Nicolas's "menu dock" idea.
Yes, I far more prefer this solution rather than add an extra grab
bar at the top or the bottom.
But IMHO menus should also work by heldclick-choose-release, and we
don't want that to be interpreted as an attempt to tear the menu off.
Not sure that really necessary, it is just an habit we have get
accustomed to I would say.
That conflicts a bit too with menu entries thought as buttons. I mean
that when you click on a button and you move out, your click is
ignored when you release the mouse button.
That's not true for popup menu or buttons in NeXT menu, it is one of
the minor idiosynchrasy of NeXT UI ;-)
If we decide, we need a first click to active the menu and the second
one to tear it off. That means we are relying on the following
paradigm : click to active, then you can drag activated element. In
NeXT or Mac OS X UI, you can drag non-activated elements unlike old
Mac OS iirc. I'm afraid this choice could trigger various consistency
issues.
Clarity, clarity, clarity. Do you see 'Grab' written on any of
your mouse'
buttons? No = people won't get it, unless you tell them. However,
I'm not
I'm not reading 'Contextual menu' on my second mouse button either :-)
I know it's not really realistic. My point is that we could try to
change the way we currently use mice :
- LMB = select/activate/move/open (with drags and multi-clicks)
- MMB = I don't know...
- RMB = contextual menu
If we support tear-off menus and add a new selected object menu in
the menur (that can be put anywhere on screen like an inspector by
tearing it off), I don't think contextual menu really deserves a
mouse button. imho we spend far more time dealing with complex UI
objects move/copy patterns. Therefore it seems natural to dedicate it
the second mouse button (when there is one).
First of all, yes, these actions should be written on the mouse ;)
Then, they should be redistributed in a way that maximizes
orthogonality... my idea:
- designate button (probably under the index finger)
- activate button (middle finger)
- grab (thumb)
The designate button would only select/unselect things (no
double-click semantics) ; the activate button would open files, or
display a contextual menu (you "activate" the desktop, to act on the
selection) ; the grab button is used to "keep things in your hand and
still use the other fingers". For instance, to move files, you'd
designate them, grab, and still be able to navigate folders while
holding them. Or you could grab a tool from a palette to use it
several times in a row without returning to the default tool...
I would surely like something inspired by such mouse interaction model.
Not everyone has a multi-button mouse :D
Hell, even Apple has, now :-P
PowerBook, iBook and MacBook Pro too ? hmm I'm not sure :-)
Quentin.
--
Quentin Mathé
[EMAIL PROTECTED]
_______________________________________________
Etoile-dev mailing list
[email protected]
https://mail.gna.org/listinfo/etoile-dev
