Re: comments on current spec

[Tim Janik <timj gtk org>]

On Sun, 30 Apr 2000, Tim Janik wrote:

> On Fri, 28 Apr 2000, John Harper wrote:
> 
> > |> _NET_CLIENT_LIST[_STACKING]
> > |> 
> > |> 	should this only include currently mapped windows?
> > |
> > |definitiely not. as in the old spec, the window manager has to provide
> > |pagers/tasklists with all windows it actually manages, independant of
> > |their skip-from-task-list/hidden/unmapped states, otherwise they simply
> > |can't keep track of windows temporarily switching states (i.e. the
> > |tasklist would loose track of iconified windows).
> > 
> > I think I should have been clearer: I meant windows that the wm is
> > aware of, but have been withdrawn by the client. They're not visible
> > and they're not iconified, so there doesn't seem to be any point in
> > putting them in the CLIENT_LIST properties?
> 
> ok, got it. yeah, withdrawn windows should be in the client list,
> iconified ones should of course.

ok, that didn't make a whole lot of sense ;) next attempt:

yeah, withdrawn windows should _not_ be in the client list, iconified
ones should of course.

> > 
> > 	John
> > 

---
ciaoTJ