W3C home > Mailing lists > Public > public-swd-wg@w3.org > February 2008

Re: Best Practices Issue: RDF Format Discovery

From: Ben Adida <ben@adida.net>
Date: Wed, 06 Feb 2008 12:20:46 -0800
Message-ID: <47AA169E.7060101@adida.net>
To: public-swd-wg@w3.org


Sean,

This is a response to your comment from November 22nd [1] as it pertains
to RDFa. This message also serves as a more complete response to Dan
Connolly's request that we adopt a "follow-your-nose" strategy for
discovering RDFa.

We agree that there should be a follow-your-nose architecture for RDFa:
a parser should be able to discover the RDFa specification by following
URI pointers from a document that contains RDFa. At the same time, we
want to make sure that RDFa can be copied-and-pasted between HTML
documents, especially in a world where widgets -- chunks of HTML pages
controlled by different entities -- are becoming commonplace.

In the long run, we hope that RDFa will simply be a standard part of
future HTML versions, or an extension to HTML specified in a standard
way (where a future version of HTML provides the appropriate
extensibility mechanism.) In such a world, HTML+RDFa can be easily
copied, in chunks, from one document to another.

For example, XHTML2 will include RDFa by default: from the XHTML2
document declaration, a parser can follow its nose to the RDFa
specification. In XHTML1.1+RDFa, the doctype declaration also provides a
pointer back to the RDFa spec, and this is the only required approach we
have.

We hope that HTML5 will at least provide a standard extension mechanism
for components like RDFa, as the HTML WG charter specifies. Or, they may
find that RDFa is useful enough to include as part of their baseline
specification (though extensibility should not be curtailed even if they
do.)

In addition to this doctype-based inclusion, our specification
*recommends* that HTML+RDFa documents include a @profile reference to
the RDFa profile URI [2]. However, we do not make this a requirement, as
we wish to allow parsers to aggressively look for RDFa when they so
choose. This aggressive parsing can serve to bootstrap our future ideal
world where HTML+RDFa can be freely copied and pasted. Since the RDFa
attributes are unique enough and fully backwards-compatible with
existing HTML, it is extremely unlikely that someone would write RDFa
they did not mean "by mistake."

In the end, we hope that RDFa will become a standard syntax for 
embedding RDF in HTML, so that it can be freely copied and included in
chunks across web pages. We do not require that parsers seek out RDFa if
there is no path to the RDFa spec, but we certainly don't frown upon it,
either, as it enables a slew of interesting new use cases whose benefit
will be most apparent once they can be prototyped in the existing web.

-Ben Adida
Chair, RDFa Task Force

[1]
http://lists.w3.org/Archives/Public/public-swd-wg/2007Nov/0056.html

[2] http://www.w3.org/ns/rdfa - note that this profile URI may change 
before we go to REC, and that, more importantly, it will contain a lot 
more machine-readable information about RDFa, too.
Received on Wednesday, 6 February 2008 20:20:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 6 February 2008 20:20:57 GMT