Hi Victor --
I quite like these representations - easy on the eye, dead easy to process and certainly no doubt or ambiguity as to what the triples are or what a triple 'is'.
Yes, it's easy for people to read RDF triples -- provided the people concerned are pogrammers.
Then comes the question -- what do conclusions drawn from those triples mean to non-programmers, e.g. business analysts, in the real world? Think of this as Semantics-2, if you will.
In , we have had reasonable success in taking a small authors-and-papers example in RDFtriples, answering real world questions, and in explaining the answers.
In , we have found it much more of a challenge to do the same for a triples version of the well-known Wine ontology.
It seems that what is needed may be a uniform way of mapping triples to a collection N-ary (N>3) "base real-world concepts" that are independent of any particular application, but useful for all applications. (Relational database folks would call these "views")
Does anyone know of any work of this kind please? All pointers gratefully followed.
PO Box 1412
CT 06011-1412 USA
Phone: USA 860 583 9677
Cell: USA 860 830 2085
Fax: USA 860 314 1029
At 06:39 PM 1/15/2005 +0000, you wrote:
Excuse the blatant plug but all three of these formats are available to eyeball on SchemaWeb  if anyone hasn't come across these triple representations before.
Incidentally TriplesML came about only because it is the internal storage of a lightweight RDF parser developed at SchemaWeb . It just seemed right to expose the 'raw' triples on the parser interface. Actually I quite like these representations - easy on the eye, dead easy to process and certainly no doubt or ambiguity as to what the triples are or what a triple 'is'.