Robin Berjon wrote:
On Nov 9, 2009, at 16:41 , Marcos Caceres wrote:
That would be 'application', but not maximized.
Uh, but those can be two different windowing modes, with the chrome
subtly different and different behaviour (e.g. the window can't be
dragged if maximised).
That's UA/OS dependent.
How it is implemented is UA/OS/UI dependent, but it doesn't mean that
there isn't a semantic difference. The differences are:
- show me alongside other apps (windowed mode)
- show me, no other app, but keep the OS UI (maximised)
- show me, and nothing else (fullscreen)
Right.
I'm happy for implementers to map the values we list to whatever makes
sense on their platform, but we need to at least have a vocabulary that
covers the more common modes. All versions of Windows in recent memory
as well as most Linux windowing managers support the three levels above,
only OSX believes that it's a good idea to annoy people who are two
pixels off in clicking on the scrollbar. Without the three levels above,
we can't capture the most usual windowing semantics.
I agree. I wonder if we can leverage some text from CSS. However, it
should not be too hard to specify this.
Or are you thinking about this in terms of the broken OSX UI that can't
tell the difference? If so, I strongly object — it's a usability
nightmare.
Exactly, so stop imposing your dirty Vi command-line view of the world
on the rest of us, Robin! :)
Actually, I'm thinking of usable click-and-drool UIs as my primary use
case.
But seriously, I don't think we need to get to the level where we are
specifying behavior.
No, but we do need a level of semantic description that matches typical
UIs.
Agreed.
