W3C home > Mailing lists > Public > semantic-web@w3.org > July 2008

Re: RDF/OWL/Sparql and native XML databases

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Wed, 9 Jul 2008 18:53:05 +0100
Message-Id: <63B36361-887C-48A2-9317-5B6AD81DF58D@cs.man.ac.uk>
Cc: <semantic-web@w3.org>
To: "Johnson, Matthew C. (LNG-HBE)" <Matthew.C.Johnson@lexisnexis.com>

On 9 Jul 2008, at 18:26, Johnson, Matthew C. (LNG-HBE) wrote:

> Hi all,
>
> Admittedly, this is not a semantic-web question, per se, but I’m  
> thinking that this list may be more relevant than xml-dev…but I  
> could be wrong.
>
> Does anyone have any experience or comments on using a native XML  
> database (e.g. MarkLogic, Exist, Oracle XML DB) as a triple store?

Not personally. But you might look at the slides starting:
	http://www.w3.org/2004/Talks/tp-robie/slide1-0.html

It may be a bit dated. There may be a paper floating around.

> If so, I’m curious as to how the translation from a XML model to  
> the RDF model can be achieved.

Probably the easiest is to "implement" a triplestore in XQuery. You  
could do this via a custom XMl triple model. Think about mapping  
triples to relation tables.

>   From my experience, most of these products focus on XQuery (which  
> is great but doesn’t seem powerful enough to understand the RDF  
> model except to possibly extract all rdf:Description elements and  
> then pass them off to something like Jena separately)
[snip]

The Robie approach is to normalize RDF/XML inside XQuery.

Hope this helps.

Cheers,
Bijan.
Received on Wednesday, 9 July 2008 17:50:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:45:29 GMT