W3C home > Mailing lists > Public > www-rdf-interest@w3.org > July 2003

Re: How to handle validation of XML within an RDF document?

From: Arron Ferguson <geek-kit@shaw.ca>
Date: Fri, 11 Jul 2003 02:43:50 -0700
To: leo@gnowsis.com
Cc: "'Rdf-Interest'" <www-rdf-interest@w3.org>
Message-id: <3F0E86D6.7090708@shaw.ca>


Thank you very much. Your answer made it quite clear. I was wondering if 
XSLT fit into it the equation somehow. In answer to your question, yes I 
have read the Primer and gone through the model, syntax and a few of the 
others although not as thoroughly as I went through the Primer. That 
seemed to be the one unanswered question in my mind. I could see how it 
all fit in and how triples work. I think the only thing I have a problem 
with is with the literals as I see that as something that goes against 
what RDF stands for: semantic nets. And yes, now it will be happy 
RDF'ing. Thanks again.


Leo Sauermann wrote:

>arron wrote:
>>My question with all of that is how do we get past the fact that 
>>the XML Schema validation is going to fail because we now have 
>>unknown attributes into the tag set and unknown elements in the 
>>tag set. Or is this not an issue because this RDF document is 
>>really the result of a transformation done by an RDF API (kind of 
>>like XSL-FO is really the result of taking some XML, running it #
>>through some XSLT document which results in the XSL-FO document) 
>>and all this is done *after* the validation using XML Schema?
>This is not an issue, if you an application application that takes an
>XML file as input and wants to transform it into RDF for further
>a typical example would be:
>- Accept XML input "X"
>- validate "X" against XML Schema "XML-Schema-for-X"
>- use XSLT stylsheet "X-2-RDF" to transform "X" into "X/RDF", so "X" is
>now in the RDF/XML representation
>- use Raptor/Jena/Sesame or any other RDF Tool to parse "X/RDF" for
>further computing, f.e. storage or inference.
>a few hints:
>- "X/RDF", the rdf representation can be very similiar to the XML
>- especially if you use Namespaces (which I miss in your xml-schema
>- you can use RDFSchema to set a similiar check in the RDF world, but
>there are only few validators for this, checkout
>this is old but does the best validation you can get.
>- RDF is not just a XML dialect, it is a Graph model, so everything is
>expressed as "Triples", this leads to a few architectural principles. I
>assume you have read the 
>happy rdf'ing !
>Leo Sauermann
Received on Friday, 11 July 2003 05:43:36 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:46 UTC