WM: grouping applications (TAI)

[Maciej Pilichowski]

On Wednesday 29 April 2009 23:22:06 Matthew Woehlke wrote:

> Maciej Pilichowski wrote:
> > On Wednesday 29 April 2009 21:45:47 Matthew Woehlke wrote:
> >
> > Matthew, please do not add key bindings to the actions, ok?
> > Because I am bit lost, and besides, we can discuss key bindings
> > later.
>
> Yes. Um... I didn't, did I?

Just a bit :-))

> >> Maciej Pilichowski wrote:
> >>> PS2. And I have difficulties with this:
> >>>
> >>> wm-switch-window-{previous,next}-sequence
> >
> > I mean what do we need sequence switch for. It is redundant imho.
>
> I'm confused, where did "-sequence" come from?

From me, I added it to avoid confusion with historic and spatial. Or 
we didn't talk about sequence at all? :-)) Then I am sorry.

> >> "Smart" just means that it is either spatial or historic,
> >> depending on the container type.
>
> I don't think it's redundant. The point is to be able to bind one
> set of keys and have that set do the "most sensible" thing
> depending on the container, rather than /always/ spatial or
> /always/ historic.

The example below convinced me :-) Ok.

> > For TAI user would have both historic and spatial (for all
> > containers actually), so when I choose "smart" switch what the
> > action will be? For TAI let's say. Why spatial would be more
> > important than historic or vice-verse.
>
> TAI - spatial
> GAI - probably spatial?
> FAI - historic
>
> The reason is that (pardon) historically, "historic" is how users
> are used to window switching in FAI working. And I think there is a
> reason why FAI uses historic by default and not spatial. But for
> TAI, precedent is for spatial, and again I think that makes sense.

Yes. Small thing -- I think GAI should be historic too for very 
similar reason as in FAI. Besides GAI and FAI are rather more similar 
to each other, than GAI and TAI.

> and in  
> floating, spatial actually does historic.)

I don't a reason for such limitation -- this can be really spatial and 
be helpful.

Cheers,