[Kde-accessibility] Timeframe for KDE 4

[Bill Haneman]

Hi Gary:

I agree with you, as I'm sure you saw from my recent post (I hadn't read
your reply yet).  Switching to DBUS makes sense, but I believe that we
will get better traction short-term by proceeding with the Qt/KDE to ATK
approach which Harald had already partially implemented last year.

ATK is much friendlier to in-process work than any of the IPC mechanisms
are likely to be, IMO, and ATK only pulls in glib which is a much
lighter set of dependencies and therefore may be acceptable long-term. 
Note that this still would not mean a direct glib dependency for KDE,
only for the accessibility specific QAccessible-to-ATK bridging
component.

This, by the way, is the way the most recent OpenOffice work has been
going - bridging from a toolkit-specific accessibility implementation
(UNO Accessibility API) to the in-process ATK interface, which is then
exported out of process by the ATK-to-AT-SPI bridge.  When the
ATK-to-AT-SPI bridge and AT-SPI framework itself gets converted to use
DBUS (assuming there is still sufficient motivation to do so), then all
the desktop apps will "just work" under the new framework, including the
assistive technologies.

Bill

On Mon, 2006-02-20 at 22:38, Gary Cramblitt wrote:
...
> 
> I believe there is ALOT more to this.  The AT-SPI currently requires CORBA 
> libraries.  There is a desire to switch it to DBUS.  To do this properly, we 
> need an IDL to DBUS compiler.  AFAIK, nobody has begun work on that.  See 
> this thread:
> 
> http://lists.kde.org/?l=kde-accessibility&m=113469521917985&w=2