RE: Can anyone clarify my questions?

(Lars, I'll give it a shot. Feel free to clarify my answers!)

>How can a metadata of an attribute be represented in triples.

The subject-predicate-object structure of a triple is great for storing
objectID-propertyName-propertyValue (or, to bring it closer to your
terminology, objectID-attributeName-attributeValue) triples. A lot of RDF
data becomes more intuitive when you think of it in those terms, but can be
confusing to think of "object" as both the third part of the triple (in the
sense of subject/object) and as the first part of the triple (in the sense
of an object having properties). 

>And how uniquely the machine can extract the information from the 
>web through rdf files for its context.
>  Example:
>          <foaf:person>
>    Now foaf is the namespace where the person's properties 
>and "about person" are defined.

No, you're confusing property (or attribute) names with property (or
attribute) values. foaf here is a prefix representing the URL
http://xmlns.com/foaf/0.1/, the name for a namespace containing property
names like Person, nearestAirport, homePage, schoolHomepage, etc. The
person's properties are listed in a file that uses these names to identify
the person's properties; for example, the FOAF data file
http://www.snee.com/bob/foaf.rdf has one property with the name
foaf:schoolHomepage and the value http://www.columbia.edu/. Other properties
in my FOAF file list the URLs of some of my friends who have FOAF files, and
an automated process could retrieve the resources at those URLs to see what
values those friends of mine have for their foaf:schoolHomepage properties. 

Reification: see 4.1 at
http://www.w3.org/TR/1999/REC-rdf-syntax-19990222/#higherorder.


Bob DuCharme   www.snee.com/bob       <bob@  
snee.com> weblog on linking-related topics: 
http://www.oreillynet.com/pub/au/1191

Received on Friday, 27 August 2004 17:27:16 UTC