Re: [XBL] XBL 2.0 Semantic-free XML elements

On Wed, 11 Oct 2006, Cameron McCormack wrote:
> 
> Ian Hickson:
> > Why does it matter? The requirement isn't on the UA to distinguish one 
> > from the other. The only requirement that uses the term "semantic-free" is 
> > one that requires the UA to treat an element as if it was semantic-free, 
> > which is something any UA can trivially do (since it is the default 
> > behaviour when it doesn't know about it).
> 
> The current text says:
> 
>   Similarly, XBL elements (other than the xbl element itself) that do
>   not have a correct xbl element as an ancestor are in error too, and
>   UAs must ignore them, treating them as they would any arbitrary
>   semantic-free XML element.
> 
> but some UAs don't ignore such elements.

The term "ignore" is a hyperlink to the definition of "ignoring", which 
emphasises that this is only ignoring for the purposes of XBL processing.

Is there a way I can define this that is clearer, maybe?


> Also, should such in-error XBL elements still implement the relevant XBL 
> DOM interface?  I think it's problematic for elements to gain and lose 
> interfaces if they are moved about the document.

Yeah; I've updated the spec to be clearer on this.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Wednesday, 11 October 2006 22:44:51 UTC