W3C home > Mailing lists > Public > public-rdfa@w3.org > September 2014

RDFa and SPARQL

From: gig graham | ontomatica <gig.graham@ontomatica.com>
Date: Tue, 16 Sep 2014 17:26:17 -0500
To: <public-rdfa@w3.org>
Message-ID: <000701cfd1fd$3c993b80$b5cbb280$@ontomatica.com>
I have a 2 part RDFa question.



Part 1

For this site:

       http://ontomatica.com/public/test/3_infotext.html

I am using several ontologies where terms are coded. Example:

       edam:data_1177

       iao:IAO_0000027

Unlike schema.org and some other controlled vocabularies which use words, e.g.

       foaf:Person or schema:dataset

one wouldn't know that:

       edam:data_1177 is ' MeSH concept ID'

       iao:IAO_0000027 is ' data item'

I can overload rel='' with multiple space separated CURIEs.

But how can I write RDFa so that I can separate the CURIEs and associate content='' with each?

The reason why it's important is that Google Rich Snippets will parse and retain the value of content=''.

For example:

       http://www.google.com/webmasters/tools/richsnippets?q=http%3A%2F%2Fontomatica.com%2Fpublic%2Ftest%2F3_infotext.html



Google sees property='' and associates content='' with the term.

Here is the constraint.

I need to be able to SPARQL the page.

If I put rel='' and content='' into <span></span>, SPARQL will not properly parse the relationships.

Is there a method to write RDFa where content='' is associated with multiple relationship-terms AND which can be SPARQLed?



Part 2 build on Part 1.

How to nest a relationship in RDFa that can be SPARQLed?

For example, I would like to associate the first term ("Authority: US NLM MeSH") with each term under the category " Semantic Types 
Ontology".

But all attempts so far have failed, and I'm left with a "naked," not-related assertion.

When there is no requirement to SPARQL a page, the method to nest/recurse the RDFa clause is straight-forward.

But I have not been able to use either <span> or <div> to separate relationships in a form that also can be SPARQLed.



---

I am prepared for an answer that "what you have" is the best that can be done; that SPARQL is not designed to process an "expressive" 
RDFa document with multiple <div> or <span>.



But if you know a better way to implement the example that also can be SPARQLed, I'll use it.



/g




Received on Tuesday, 16 September 2014 22:26:41 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:04:54 UTC