[openbox] sloppy focus?

Ben Jansens ben at orodu.net
Mon Nov 17 00:30:30 EST 2003


On Sun, Nov 16, 2003 at 05:45:08AM +0000, corey at streamreel.net wrote:
> On Sun, Nov 16, 2003 at 04:01:24PM -0800, Marius Nita wrote:
> > On Sun, Nov 16, 2003 at 03:55:35AM +0000, corey at streamreel.net wrote:
> > Openbox doesn't try to be standards compliant by copying what other window 
> > managers do, or trying to guess what the social definition of "focus follows 
> > mouse" is.
> > 
> 
>   Copying other wm's, or guessing "social definitions" is a far cry from 
>   simply being a little mindfull of certain normalized conventions, which
>   ob3 appears to be, except for the specific singular issue of the ffm 
>   behavior - which makes that particular discrepancy ( I would actualy go
>   so far as to call it a bug ) all the more vexing.  (c8=

Progress is generally not achieved by duplicating other efforts. I was
trying to make FFM better. I think its a horrid hack in the first place
designed by engineers, not "user-interface aware" people of any kind. :)

The idea is that you don't want windows to raise right away because you want
the window you are going to to be raised, and not the windows in between.
This way the stacking order is preserved even though you flicked your mouse
across them.

I applied this same logic to the focus order, which was not even present
when FFM was first applied to WMs. You want to the window you are going to
be using to be moved to the front of the focus order, not all the windows in
between.

When you are using auto-raise, I think in the general case it is safe to
assume you are meaning for the window to raise for you to use it, you are
not intending to use the window without raising it, like in plain old FFM.
So in this spirit, I grouped the two delays together, meaning it became
instead of a raise delay, more of a "use the window under the mouse" delay
to let to flick across windows without messing things up.

I would ask you to try a little experiment for me :) Try shortening the
delay so that it is just longer than the amount of time you use to flick the
mouse across the screen, no extra time to dawdle :) See if shortening this
time does not let you use the new approach. It may mean adjusting to it a
little bit, but moving from a bad interface to a good one usually will.

Maybe I am wrong of course and it will just suck horribly, but I'd
appreciate it if you gave it a try for a bit and see if it can't turn out or
not. Also, maybe you will come to know and love Alt-Tab cycling which
suddenly becomes feasible for you to use :)

> > It implements actual _standards_.

I would also have to agree that FFM is in no way a standard. Rather it is a
focus policy that has been implemented in a number of ways, usually with a
bare minimum and then plenty of options to try cater it to everyone. Kinda
like saying Openbox isn't following the standard of metacity by using all
its code, imo :)

Anyhow enough typing and more playing warcraft :)

Ben
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20031117/c8b6b1a9/attachment.pgp>


More information about the openbox mailing list