On 11/10/05, Phil Shaw <[log in to unmask]> wrote:
> Do you propose a formal mapping from any-property to
> any.property and any:property, or is it purely the responsibility of
> the transformation to make sense of whatever prefix scheme is
> used?
any-property and any.property are equivalent syntactic forms for
encoding a property URI. They both mean lookup the schema URL
corresponding to the part before the punctuation and concatenate it
with the part after the punctuation.
The any:property notation is not supported in embedded RDF.
> There is clearly a dependency between the source format, a
> transformation service, the right to conduct the transformation and
> the longevity of one's data. The widespread adoption of this format
> is probably dependent on having a standard XSLT as part of the
> specification in the public domain (as you have done).
Yes. The intention is to make all the tools, specifications and
examples freely available for any use. I will formalise this on the
website as soon as possible.
> This approach seems a more natural progression from XHTML than
> the W3C discussions I read on this a while back. It's not literally
> embedding RDF in XHTML, more like encoding RDF in XHTML.
>
> Do you see a fit in with W3C plans, a rival or alternative?
>
I have been frustrated with the W3C approach because it hasn't
generally considered compatibility with existing practices to be a
goal. I don't claim to have a complete RDF in XHTML solution, but I
think I enough to be very useful. I'm currently at the Semantic Web
conference in Galway and I've had many discussions with RDF
practitioners, some of whom are W3C staff or members. All of the
discussions have been positive so far and there are lots of people
trying it out with their own RDF systems. Some feedback I've had has
led to experiments with enhancements which I've been writing about on
my blog [1]. I now think it's possible to embed almost all of the RDF
model into XHTML without inventing new markup or breaking
compatibility with existing tools and still retaining an essential
simplicity.
I'm not positioning this in opposition to the W3C efforts. I see it as
something that fills a gap right now. Ultimately it would be nicer to
work with specific new markup with clean semantics and the RDF/A and
XHTML2 efforts will produce that. However, based on the rate of uptake
for XHTML 1, I think that it'll be ten years before either of the
aforementioned formats are seen in significant quantities on the web.
I don't want to wait that long ergo embedded RDF.
Cheers,
Ian
[1] http://internetalchemy.org/2005/11/enhancing-embedded-rdf
--
http://internetalchemy.org | http://purl.org/NET/iand
Working on... Silkworm <http://silkworm.talis.com/>
|