Re: Linux Window Managers and Accessibility

"Carl R. Witty" wrote:
> I don't think using clients behind two NATs on one X server is
> contrived at all.  Both my home network and my work network are behind
> NATs, and have fast, low-latency Internet connections; I can easily
> imagine being on the road and displaying clients from both my home
> machine and my work machine on my laptop X server.  The contrived part
> was if both machines had the same IP address and the same hostname.

OK, I agree.  I didn't mean the 2 NATs was in itself contrived, just
the net result of all 3 being the same.

> > We care a lot. We do *NOT* want to change the window manager, and we
> > haven't time to wait for a new WM_* property if that's what it takes.
> > If you are correct that a program cannot unambiguously identify itself
> > with a window manager then we are in trouble, I agree... but I am not
> > convinced that's the case.  Toolkits, yes we have inserted the hooks
> > into GTK+ and Java (which had them already).
> I still must not understand what we're talking about, then.  The
> message that started this discussion asked whether Sawfish can provide
> certain information, or whether Sawfish would have to be modified in
> order to be able to provide it.  Sounds like Brian Cameron, at least,
> was thinking about changing the window manager.

Well, we're not sure but it looks to me as if it's implementable via
existing sawfish features provided sawfish supports most of the extensions (I think it does).  

> How are you envisioning the window manager providing this information?
> What is it providing information to?  (I was under the impression that
> the window manager was providing information through some side CORBA
> channel to the AccessibilityBroker, which would certainly require
> changing the window manager.)

I hope it doesn't have to be CORBA - but we do need C bindings,
callable from the broker, that allow us to get the WM_ properties for
the toplevel windows.  I would not be surprised if we have to write
these bindings, as long as we can get to existing lisp API (and in
that respect I hope we don't need to change sawfish).  This is the
part I've never done...


> Carl Witty

Bill Haneman
Gnome Accessibility / Batik SVG Toolkit
Sun Microsystems Ireland

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]