- From: Doug Schepers <schepers@w3.org>
- Date: Mon, 31 Jan 2011 17:44:21 -0500
- To: Chase Douglas <chase.douglas@canonical.com>
- CC: public-webevents@w3.org
Hi, Chase, Stephen- Thanks for the feedback. You both seem to be saying the same thing; please correct me if I'm missing some subtlety. I'm happy to refine the definition of touch area. The question becomes, should we define these as several individual attributes, or as a new data type or object with the angle values. Anyone have any thoughts? Regards- -Doug Schepers W3C Team Contact, SVG, WebApps, and Web Events WGs Chase Douglas wrote (on 1/31/11 2:37 PM): > The one > area of concern I had was about the touch shape. The spec only provides > for radiusX and radiusY, which means that touch shapes must have rigid > angles of 0 or 90 degrees. > > What we have coming out of the Linux kernel is a touch angle and major > and minor axes lengths: > > http://www.mjmwired.net/kernel/Documentation/input/multi-touch-protocol.txt#179 > > This would provide for any angle of touch input. I would like to suggest > following this approach and adding an angle parameter to touchpoints. Stephen M. Webb wrote (on 1/31/11 2:53 PM): > The limitation of the contact area description of the TouchPoint > Interface to axis-aligned ellipses is unnecessarily restrictive. > > I would suggest this datum is better described with a triplet of > semimajor and semiminor axis lengths and rotation angle, at the very > least.
Received on Monday, 31 January 2011 22:44:24 UTC