Re: Transforming RDF into XML (alternative using ShEx)

On Sat, 2016-07-09 at 11:52 +0200, Felix Sasaki wrote:
> Thanks, Jose. From the call yesterday we came up with a rough
> classification of how to structure our discussion
> 
> 1) functionality needed: convert XML to RDF, or the other way round
> 2) way to implement it: with standard technologies (e.g. XSLT on the
> XML side) or with enhancements

The RDF that you want to end up with from any given XML document may
vary greatly depending on your purpose: XML generally represents
information, not knowledge, and you can extract an unbounded set of
knowledge from a given piece of information.

Examples include "make a list of every natural-language word in the XML
document, together with how many times it occurs" or "extract the names
of people from this document, based on interpreting the content of
"person" elements as people's names."

So I don't see that this can be standarized usefully.

> 
[...]
> Now, if people want to stay with 2), they can follow some potential
> best practice; e.g.  how to store RDF output, so that it can be
> processed with standard XML tools. See the mail from Martynas:
> 
> [
> We use however a canonical "flat" RDF/XML layout, in which
> <rdf:Description>s are not nested (default output by Jena writer). By
> limiting RDF/XML to such layout, and using the key() function to
> lookup descriptions etc., the transformation becomes quite
> manageable.
> ]

Yes. RDF/XML seems to be the result of some really creative people
trying to make the worst XML vocabulary they can imagine. A good clue
is that if you design an XML format yourself and define more than one
XML namespace you're probably doing something wrong. I always hoped
Jeremy Carroll's RDF-in-XML work would get wider adoption than it did.

A simple XML vocabulary that doesn't let RDF introduce new namespaces
or element names - e.g. an n3.xml or turtle.xml - would be a big help.

> > mapping XML of dictionary data to RDF

This is always a fabulous example, especially when you add the ontology
matching problem of identifying corresponding lemmata across languages.
peope have been doing it for years, of course, but you have to have a
clearly-defined purpose first, since the mappings aren't 1:1 and
there's considerable subjectivity involved.

Liam

-- 
Liam R. E. Quin <liam@w3.org>
The World Wide Web Consortium (W3C)

Received on Saturday, 9 July 2016 18:16:51 UTC