W3C home > Mailing lists > Public > public-html@w3.org > October 2009

Re: Extend @refid and including of microdata

From: Philip Jägenstedt <philipj@opera.com>
Date: Sat, 17 Oct 2009 20:51:54 +0200
To: "Giovanni Gentili" <giovanni.gentili@gmail.com>, public-html@w3.org
Message-ID: <op.u1yj0sp4sr6mfa@worf>
On Sat, 17 Oct 2009 17:48:50 +0200, Giovanni Gentili  
<giovanni.gentili@gmail.com> wrote:

> Actually in the spec [1] the refid attribute value
> must be the ID of an element in the
> same Document as the itemref element.
> I think it would be useful to include also properties
> defined by elements that are NOT in the same Document:
>   <div itemscope id="amanda">
>     <itemref refid="http://example.org/amandainfo#abc">
>   </div>
> This allows to include a recurring set of properties
> into multiple elements placed in multiple pages,
> without duplicating the info and with only one "source".
> And, as another possibility, microdata could
> be also extended to permit including like this:
>   <itemref type="application/rdf+xml"  
> src="http://example.org/amanda.rdf">
> [1]  
> http://www.whatwg.org/specs/web-apps/current-work/multipage/semantics.html#itemref

Can you outline how the algorithm for associating names with items [1]  
should be adapted to support these suggestions? Should cross-origin  
references be allowed? They could (relatively) easily be handled by  
external scrapers, but cannot be exposed via the DOM API because it would  
allow cross-origin attacks. In the last case, how should the RDF triples  
be represented in the microdata model?


Philip Jägenstedt
Opera Software
Received on Saturday, 17 October 2009 18:50:41 UTC

This archive was generated by hypermail 2.4.0 : Saturday, 9 October 2021 18:45:00 UTC