- From: Simon Pieters <simonp@opera.com>
- Date: Sun, 02 Aug 2009 12:49:00 +0200
- To: "Ian Hickson" <ian@hixie.ch>, "Doug Schepers" <schepers@w3.org>
- Cc: "public-html@w3.org" <public-html@w3.org>
On Sun, 02 Aug 2009 12:19:02 +0200, Ian Hickson <ian@hixie.ch> wrote: > On Thu, 23 Jul 2009, Doug Schepers wrote: >> >> In looking into "activation behavior" for the DOM3 Events spec, I was >> reviewing the HTML5 spec. Currently, the <audio> and <video> elements >> do not have defined activation behavior. >> >> I suggest that they should, and propose that the activation behavior for >> <audio> and <video> is to play if the media is currently stopped or >> paused, and to pause if the media is currently playing. > > There is in fact a comment in the source of the document that says: > > <!-- XXX (awaiting more impl experience) > default activation behavior is to do the play() if paused, pause() > otherwise > --> > > The implementation experience that I am awaiting is basically, does the > above make sense? It would mean that the author could bind an access key > to the play/pause behaviour, but is that really useful, when the author > can just script the whole control if desired? It would mean that if the > author called .click() on the element, it would play/pause the element, > as > well as firing an event. It would also mean that while every other > control > could be controlled by the user agent (when controls="" is set) and the > author (via script), clicking anywhere on the video, whether the element > had a controls="" attribute or not, would always play/pause the video. The YouTube HTML5 demo has linked <video>s. I think this is a clear indicator that it should be possible to put <video> in a link. -- Simon Pieters Opera Software
Received on Sunday, 2 August 2009 10:49:48 UTC