Phil Windley had a recent interview with Elias Torres and Ben Adida on RDFa. This is an emerging standard for adding semantic information to HTML documents. The “a” in RDFa stands for attributes. Rather than creating new documents, RDFa allows you to add RDF-like semantic information to your existing HTML pages via element attributes. A great deal of thought has gone into how to accomplish this without unwanted side effects such as changing how pages are rendered.
As I listened to the interview, I tried to think of how I would apply this to my websites. There are standardized vocabularies for expressing friendship relationships, address and calendar information, audio file meta data, etc. However, little of this is relevant to my websites. I’m ready to jump on the RDFa bandwagon once there are standard vocabularies more applicable to the kind of content I publish.
To learn more about RDFa, listen to Phil Windley’s podcast or read the RDFa primer.
RDF is nice. But it has been around for nearly ten years now (if not over ten years, I can’t recall). There are many, many ways to serialize RDF and RDFa is just the latest salvo. It is not even all that novel as Dublin Core is already commonly embedded in HTML pages (Dublin Core is the most widely used form of RDF).
I know of only two practical applications of RDF. Creative Commons and Dublin Core. And Creative Commons RDF is like an extension on Dublin Core, so it hardly counts.
The basic problem is that RDF is metadata. Metadata is not all that useful.
More metadata does not mean better technology. That’s a myth.
The problem with metadata is that it is wrong, misleading, too general, too specific… you name it… there is never a good match between the metadata and the user query.
A little bit of metadata is great. Too much is a waste of time.
Well, it’s refreshing to hear some criticism of RDF etc. Most people either ignore it or talk about it as if it were mankind’s greatest hope.