W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > November 2011

Mixing-up Semantic Data with HTML Rendering

From: landong zuo <landong.zuo@googlemail.com>
Date: Mon, 28 Nov 2011 13:18:10 +0000
Message-ID: <CACNn1brRo5J8uoL2n0=K=Yo=H500a6ZaPvYdGOK4XN1zmaeu1A@mail.gmail.com>
To: public-rdfa-wg@w3.org
Hi All,



May I share some my sought regarding “RDFa semantics” from the
perspective of Semantic Data Publishing.



When RDF triple is published via XHTML+RDFa syntax, it is supposed
that the “semantic meaning” of the page is independent from the HTML
rendering.

The RDFa extractor would get the equivalent set of triples from XHTML
page as it was published in RDF.



The “semantic meaning” here references to the RDF canonicalization
approach developed by Jeremy J. Carroll [1]. The semantic equivalence
can be matched via Jena graph implementation:

Boolean Graph.isIsomorphicWith(Graph g).



However, RDFa intents to reuse XHTML syntax like “@rel” and “@href”. I
wonder if this would blur some semantic content of XHTML page by
mixing-up non-semantic information like CSS with triples?



For example, a XHTML page should state the link to CSS like

<link rel="stylesheet" type="text/css" href="/Styles/theme.css" />

This would probably suggest creating triples in RDFa extractor like below,

<http://blah.co.uk/... >
<http://www.w3.org/1999/xhtml/vocab#stylesheet>
<http://blah.co.uk/Styles/theme.css>  .

Although this triple does not exist in the original dataset.

The Jena equivalence method would tell a mismatch if the RDFa
extraction is compared to the original RDF.



So, my question is “what should we consider as the RDFa semantics?”

Should publisher be free to publish the semantic data in whatever the
HTML rendering without impact to the result of RDFa extraction?



If answer to this question is Yes, should we consider to distinguish
semantic data from non-semantic data in the same XHTML+RDFa page?

Or should we revise the RDFa processing model to make this
distinguishing to the RDFa extractor?



[1] Jeremy J. Carroll, Signing RDF Graphs, available at
http://www.hpl.hp.com/techreports/2003/HPL-2003-142.pdf





Regards,





Landong Zuo

Semantic Web Developer

The Stationery Office


Web:  www.tso.co.uk

Orders:  www.tsoshop.co.uk



TSO Information and Publishing Solutions

Part of the Williams Lea Group www.williamslea.com



Every day, decisions are made that affect you – do you know enough to
have your say?

Keep up with Official Publishing at: www.haveyoursayonline.net

 P Please consider the environment before printing this e-mail
Received on Monday, 28 November 2011 13:18:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 04:55:18 GMT