- From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
- Date: Thu, 17 Jun 2010 12:32:43 +0200
- To: Jonathan Chetwynd <j.chetwynd@btinternet.com>, www-svg@w3.org
Jonathan Chetwynd: > Olaf, > > unfortunately the activate provided by squiggle/batik is merely a > pseudonym for click, > according to my test it responds to click but not enter > iirc cameron has for some time been concerned with the lack of > keyboard navigation. Because it is mentioned, that activate is considered to be a device independent event, one cannot expect, that the definition is very detailed about the question, how to implement such an event into a specific device. In doubt the documentation of the device has to describe how a user creates an activate event. But every device should have an intuitive or simple method to create such an event. For example without a pointing device a user-agent might provide together with the text alternative (elements title, desc etc) a list of possible activations and the user can select the item to activate by tabs or keys within such a list - why not? > > however your response regarding the failure of other UAs to provide > functionality does not prevent you proposing a single suitable > solution with testcase, > for a variety of different input devices. > And it might be hoped that such a discussion might better enable > implementation. > I think, the test can only mention something like: <circle id="b" r="1"> <title>Button</title> <desc> activate to begin the motion of the funny dog. In doubt read the documentation of your viewer, how to activate elements. If a pointing device is available, this means typically to click the element. Other methods maybe are available as well. </desc> <animateMotion xlink:href="#funnyDog" begin="b.activate" ... /> </circle> > It may even be necessary to raise an errata, as the gross failure to > implement may at least be suggestive. > Well, for example my bug report for Opera about this is about 5 years old (bug-174503) ;o) Olaf
Received on Thursday, 17 June 2010 10:37:39 UTC