Re: ISSUE-64: Should the Window spec define Window\'s place in DOM event flow?

On Apr 3, 2006, at 8:52 PM, Ian Hickson wrote:

>
> On Tue, 4 Apr 2006, Web APIs Issue Tracker wrote:
>>
>> In existing de facto Window UAs, the Window object is an  
>> EventTarget and
>> part of the DOM event flow; for events in the contained Document, the
>> event performs the capture phase on the Window before the Document  
>> and
>> bubble after the Document. Should Window 1.0 cover this? Should we
>> mention it in DOM Events as a requirement on UAs that also implement
>> Window? Or should we defer it to a future spec?
>
> If it happens for every event, but only for Window, not for other  
> global
> scopes, then it should be in the Window spec.
>
> If it happens for only certain events, it should be in the specs that
> define those events.
>
> If it happens for all global scopes and isn't specific to Window it  
> should
> be in the DOM Events spec.

I believe it is the first of these, but I wouldn't know where to find  
non-Window global scopes that nontheless somehow relate to a DOM. Did  
you have any in mind?

Would SVGGlobal be an example of such? I am not sure if any SVGGlobal  
implementations have the behavior I describe, or for that matter  
whether such implementations should be considered as not at least  
trying to implement a Window object.

Regards,
Maciej

Received on Tuesday, 4 April 2006 06:01:19 UTC