Home > Articles > Programming > Java

  • Print
  • + Share This
Like this article? We recommend

What Doesn't RDF Do?

By now, you may be thinking to yourself: With all that's great about RDF, why bother with any other kind of representation? Obviously, it would be impractical to reduce the sum of all human knowledge to a bunch of three-word sentences. RDF could tell you whether a play is a comedy, but it can't tell you if it's funny. RDF could tell you where to buy a dozen roses, but it won't smell as sweet. RDF belongs alongside normal data—it's the secret ingredient that makes data manageable.

Using RDF doesn't necessarily mean putting it at the heart of your high-use systems. For instance, a bibliographic database with specially defined fields for title, author, and publish date will most likely be more efficient and scalable than an RDF-based system. This database could be populated out of an RDF source, however. Such a scenario would follow the same pattern (partial decomposition) as expressed in my book.

Of course, this isn't all there is to RDF—I've barely scratched the surface—but I hope I've given you an idea of what RDF is and how useful RDF can be for modeling and representing metadata. The PRISM working group (Publishing Requirements for Industry Standard Metadata) has built a metadata standard on top of RDF that incorporates the Dublin Core metadata set plus its own elements and a bunch of best practice information. If your business is publishing, or even if it isn't, this is a great place to start for some real-world examples. Of course, the RDF pages at the W3C site are where to go to find all the canonical information on RDF.

  • + Share This
  • 🔖 Save To Your Account