Re: GRDDL extraction *to* RDFa

I agree it's a very interesting and motivated use-case.

Yet that use-case crucially depends on an unresolved GRDDL spec issue:
"whether GRDDL transformations may produce RDF in a format other than
RDF/XML." [1]

So let's not throw it in the "Use Cases" document till we reach
consensus on that. Unfortunately, that issue will have to be tabled till
our meeting on the 20th when DanC returns, but I'll put in on the agenda
for that meeting. BenA, are you happy with opening that discussion on
the Sept. 20th meeting?

[1] http://www.w3.org/2004/01/rdxh/spec

Ben Adida wrote:
> Hi all,
>
> I've been tasked with describing my driving use case for joining this
> WG. I will call this use case the "GRDDL extract to XHTML" use case.
>
> When GRDDL is used to extract triples from some XML document, its output
> is usually RDF/XML or N3, or some other such "raw" triple format. In
> some cases, this transformation discards important human-parseable
> information: everything that style and layout bring to the table. This
> use case explains how it might be useful to have GRDDL preserve the
> "human" aspect of a document during the transformation.
>
> Consider Alice, who likes to attend various conferences, talks, or other
> gatherings of social groups in her area. These groups publish their
> calendars in various HTML-based formats: microformats, eRDF, RDFa, or
> some home-grown way to express calendar information. If these calendars
> use GRDDL, Alice can easily add all of these events to her own Google
> Calendar.
>
> However, Alice doesn't want to add *all* events to her calendar. She
> wants to pick and choose which events to attend. She browses to her
> local theater's upcoming events page, scrolls down to the one event she
> likes, and then copies and pastes this event to her calendar.
>
> To enable this copy-and-paste, Alice's browser supports a default
> RDF-in-HTML embedding scheme, RDFa. Then, GRDDL hooks in the page
> indicate how to transform this homegrown XHTML into XHTML+RDFa, while
> preserving the style and layout of the page.
>
> Thus, Alice's RDFa-aware browser can perform the transform even before
> rendering the XHTML. Then, the rendered XHTML+RDFa can provide
> copy-and-paste functionality via, for example, right-clicking on an
> event right in the rendered XHTML+RDFa.
>
> For these features to work, we need to specify a GRDDL transform that
> can output XHTML+RDFa as its serialization of RDF. Ideally, this would
> be indicated right when the GRDDL hook is declared, so that a browser
> knows ahead of time what the GRDDL transform will output.
>
> -Ben
>
> PS: I used Alice instead of Jane because the use case seems different
> than the existing Jane use case... that said, I'm certainly open to
> integrating this into another use case.
>
>   


-- 
		-harry

Harry Halpin,  University of Edinburgh 
http://www.ibiblio.org/hhalpin 6B522426

Received on Friday, 8 September 2006 00:14:30 UTC