Re: RDF Triples in XML, named graphs

On Feb 12, 2004, at 14:34, ext Jeremy Carroll wrote:

>
>
>
> Eric Jain wrote:
>
>>> In this way, the same XQuery could be executed against a knowledge
>>> base and/or an actual TriX instance. One could then think of TriX
>>> as a means of integrating RDF and XQuery.
>>>
>> This is an interesting application. We currently store triples in a
>> relational database. This is quite efficient, but requires us to map
>> RDQL statements to SQL. If TriX works well with XQuery to the extent
>> that XQuery could be used as an equally convenient but more flexible
>> replacement for the RDQL subset that we have implemented, this would 
>> be
>> a strong argument. On the other hand I'm slightly concerned about the
>> performance of XML databases (size of RDF/XML is 7GB). Anyone have any
>> experience here?
>
>
> I would be surprised if the XQuery route was not significantly slower 
> than native RDF query using RDQL or similar (even when translated in 
> SQL).
>
> I think TriX permits such integration of XQuery and RDF, but that does 
> not replace the need for languages such as RDQL.
>
> I suspect one could have an RDQL to XQuery translater, which would be 
> particularly useful for smaller documents in primarily XML 
> environments.
>

I agree 100% with Jeremy.

TriX can be thought of as an "XML Enabler" for folks who want to use 
generic
XML tools with RDF knowledge bases. But that doesn't mean that they 
will get
optimum performance from those XML tools, nor benefit from much of
the power of RDF/RDFS/OWL, etc.

I see alot of opportunities in TriX for integration/cohabitation with
the XML world, but I don't expect XQuery to provide an optimal
solution for querying RDF knowledge bases.

Cheers,

Patrick


> Jeremy
>
>

--

Patrick Stickler
Nokia, Finland
patrick.stickler@nokia.com

Received on Monday, 16 February 2004 03:59:12 UTC