Re: pan, rotation, move and zoom request events

Absolutely.


Rich Schwerdtfeger



From:	raman@google.com (T.V Raman)
To:	Richard Schwerdtfeger/Austin/IBM@IBMUS,
Cc:	jason@jasonjgw.net, public-indie-ui@w3.org
Date:	03/28/2013 05:56 PM
Subject:	Re: pan, rotation, move and zoom request events



Hit testing will also be important for touch-screen interfaces,
  so the user doesn't have to touch exactly at a small point, but
  can be given a radius of uncertainty by the underlying
  application.

Richard Schwerdtfeger writes:
 > Hi Jason,
 >
 > Zoom should first be driven by hit testing in the browser. If you have a
point within an object the zoom event should be generated at the object hit
and then the object
 > should manage the zoom from that point. The application needs to make
the decision how it zooms once the event is received. This will be
particularly important, for example,
 > if you have something like a flow chart and your browser deems the hit
to be within a line. The application may find the point toward the end of
the line and want to zoom
 > around that point. We will need these concepts when we apply zooming to
SVG and HTML5 Canvas.
 >
 > So, the x,y coordinates are essential. There may be some instances, such
as text, or UI controls the app. may wish to zoom off center of the x,y
coordinate.
 >
 > Rich
 >
 > Rich Schwerdtfeger Th
 >
 > InJason White ---03/11/2013 05:44:43 PM---James Craig <jcraig@apple.com>
wrote:
 >
 > From: Jason White <jason@jasonjgw.net>
 > To: public-indie-ui@w3.org,
 > Date: 03/11/2013 05:44 PM
 > Subject: Re: pan, rotation, move and zoom request events
 >
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

 >
 > James Craig <jcraig@apple.com> wrote:
 >
 > > I think that depends on the application. In most cases, I would expect
the
 > > zoom event to be centered on the x/y coordinates if they were
provided. If
 > > origin coordinates are not provided, zoom would probably be centered
in the
 > > middle of the current view.
 >
 > That's reasonable. Could you perhaps clarify it in the text without
 > unduly limiting what applications do?
 >