Re: Re: [iPAQ] Browsers and Mail Clients

Did I say there weren't going to be some interesting problems? :-)

Do a search on "a2x" and you'll track it down....

And did I mention that we've had 3 different speech recognizers running
on Itsy?
                        - Jim

> Sender: handhelds-admin@handhelds.org
> From: Carl Worth <cworth@east.isi.edu>
> Date: Wed, 20 Sep 2000 16:39:43 -0400 (EDT)
> To: handhelds@handhelds.org
> Cc: ipaq@handhelds.org
> Subject: [Handhelds] Re: [iPAQ] Browsers and Mail Clients
> -----
> Pardon the cross-post, I'm shifting this discussion from iPAQ to
> handhelds. All replies should be directed to handhelds.
>
> Jim Gettys writes:
> > > Another question that I was pondering is how to deal with xscribble --
> > > it would be nice if one of the buttons could be a scribble toggle, and
> > > then you could scribble anywhere on the screen instead of just in the
> > > xscribble window. This would save a lot of screen real estate, and make
> > > the machine more useful overall.
> >
> > Actually, you can do better yet... You can write a program that goes
> looking
> > for windows that want input events, and "do the right thing": for large
> > windows, allow scribbling anywhere, for small dialog boxes, pop up a window
> > to scribble in. This can be done the same way that Bob Scheifler did
> speech
> > recognition control of X (the atox program).
> > - Jim
>
> Jim,
>
> I'm really intrigued by this idea. I've been thinking for a long while
> about how it would be best to get "gestured" character events to
> programs and the Right Thing would be to simply gesture on top of the
> window of interest and it would get the character event. The difficult
> problem is that some programs will be expecting both mouse events and
> keyboard events and there is an ambiguity since the stylus is being
> used to input both. In my mind I had decided that a reasonable
> compromise would be to force the user to specify whether the stylus
> event would be a character-stroke or a mouse event. This could be
> through an on-screen or hardware button, (the implementation could be
> a full-screen, transparent stroke recognizer that would be toggled in
> out of minimization via the button).
>
> At least, that's as far as I had gotten with my thinking until I saw a
> couple of posts from you regarding Bob Scheifler's, "Hand Off X!"
> work. Does that work really handle a single window that can accept
> both keyboard and mouse events? (Think dialog with "filename" text
> area and "Load" button). If so, how? And do you have a good reference
> that discusses how this was done? Or at least the definitive source
> archive for atox?
>

--
Jim Gettys
Technology and Corporate Development
Compaq Computer Corporation
jg@pa.dec.com