- From: Olli Pettay <Olli.Pettay@helsinki.fi>
- Date: Thu, 24 Sep 2009 15:50:03 +0300
- To: Anne van Kesteren <annevk@opera.com>
- CC: DOM mailing list <www-dom@w3.org>
On 9/24/09 2:54 PM, Anne van Kesteren wrote: > HTML5 currently defines how the Window object interacts with the DOM > event flow. Is it the idea to move this into DOM Level 3 Events? If that > is the case the special case made for the load event should also be made > to apply to the capture phase. See the HTML5 text here: Indeed, the spec shouldn't say "the load event does not bubble to the defaultView", but "the load event does not propagate to the defaultView". or something like that. > > http://www.whatwg.org/specs/web-apps/current-work/multipage/origin-0.html#events-and-the-window-object > > > I think if we decide to include this into DOM Level 3 Events it should > be made normative and not just a "note" (with respect to the load event) > as it is now. > > > The definition of defaultView could also use some work I think. Events > are not dispatched on interfaces after all, they are dispatched on > objects (which implement interfaces). I agree. defaultView is an object which implements AbstractView. > Also note that the CSSWG is > working on a new version of DOM Level 2 Views called the CSSOM View > Module. (I added a proposed agenda item for TPAC to discuss support for > views in general by the way.) But you're not going to propose to remove .defaultView or anything like that I hope. Using .defaultView in the D3E allows defining the "window propagation" without actually depending on any spec which defines window. -Olli
Received on Thursday, 24 September 2009 12:50:55 UTC