Re: Still need a hint for undecorated windows

[Bill Haneman <Bill Haneman Sun COM>]

Philipp Lohmann - Sun Germany - ham02 - Hamburg wrote:

>Carsten Haitzler (The Rasterman) wrote:
>  
>
> ...
>
>Undoubtedly. But my original point was just to say that
>applications/toolkits have to grab the pointer, because else the WM will
>mess their popups up. I just reacted to Mr. Pennington saying "apps want
>to grab the pointer".
>  
>
I'll just note that this behavior turns out to have bad consequences for
accessibility. For instance, if a user is trying to use an onscreen
keyboard (one of the most common cases where this turns up problems) in
'dwell mode', any pointer grabs will disable the onscreen keyboard. If
the user can only point but cannot click (for instance the user is using
a head tracker), then s/he is locked out.

XInput can be used to work around this by driving/controlling all
onscreen keyboard devices via something other than the core pointer, but
this introduces a configuration nightmare in practice for all sorts of
reasons. Multiple HIDs work OK via XInput as long as you want them "all"
to drive the core pointer, but if you don't, you're venturing into
unsupported-land. So pointer grabs make it very difficult to support
some configurations. Keyboard grabs are worse, you want to avoid them at
_all_ costs...

I am not suggesting an alternative, because I know failing to grab the
pointer breaks stuff too (for everybody). I am just saying that one
would be very desirable.

best regards

Bill

>Kind regards, pl
>
>  
>