Previous sections barely skimmed the richness of OWL, though they have shown that regardless of the complexity of the constructs, they remain valid RDF/XML. In fact, if you open up the Wine ontology that the OWL group uses for its examples, located at http://www.w3.org/TR/owl-guide/wine.owl, you'll find that it validates. You'll want to turn on the graph option first, and you should be prepared to wait because wine.owl is quite large.
So, when should you use just RDF Schema and when should you use OWL?
If you're defining a fairly simple vocabulary primarily for your own use (and I use RDF/XML for a dozen different little applications at my site), and if you're concerned primarily with the striped nature of RDF/XML, you'll most likely want to just define your vocabulary in RDF and RDFS.
However, if you're documenting a model of a specific domain and you hope to encourage others to use it and, best of all, be able to use the data to make sophisticated queries, you're going to want to use OWL to take advantage of its many inferential enhancements.
Before we leave this chapter, we'll take a quick glance at a couple of editors specialized for ontologies. Take what you've learned in this chapter out for a spin.