This post originated from an RSS feed registered with Java Buzz by Joe Shelby. Original Post: Why doesn't RDF (or XML) have a standard way of saying something else should be XML? Feed Title: Joe's Java Jottings Feed URL: http://www.blog-city.com/bc/ Feed Description: Notes, observations, and occasional other stuff on Java, with concentrations on Swing, XML, and the Semantic (Object) Web. Latest Java Buzz Posts Latest Java Buzz Posts by Joe Shelby Latest Posts From Joe's Java Jottings

Over here, I rant a little about how RDF and the Semantic Object Web (FOAF in particular) are right now still technical failures because of the lack of a standard way to reference another RDF document. 

A single RDF file is not a web.  Its a graph at best, a collection of triples at worse.  A web involves a collection of said files, cross-referenced.  RDF does allow (provided people pick the same URI for a particular subject) multiple files to define properties of a subject.  However, they don't provide means for linking to the other RDF files that discuss that subject or anything that subject's related to.

How multiple RDF files become a single knowledge base is utterly undefined by the standards, and as such, the RDF can't be "surfed" to building up the knowledge the way a true web should (a-la WWW).

FOAF in particular suffers greatly for this.  I shouldn't have to come up with my own URIs for my friends and re-enter all of their data.  I should just have a file that references my friends URIs, which should be URLs to their own information (and their own friend's URIs), and a webcrawler surf the FOAF files to assemble a true graph.  THAT's a web.  FOAF doesn't even work as bookmark file.  Its a waste of time and effort in its current form.