At the Drupal 7 Release Party just over a week ago, Mark Birbeck gave a presentation on RDFa and the Semantic Web, as Drupal 7 has just launched with native RDFa support. To be honest, I had little understanding of what RDFa actually was before I saw Mark speak, and after the talk I was nerdishly excited by the prospect of what RDFa can do.
RDFa expands out to mean ‘Resource Description Framework in attributes’. With me so far? No, I got lost about there as well. However, Mark went on to explain RDFa in context (which, funny enough, is what RDFa is about)…
The internet is meant to be consumed by people. That is, for the most part, people can cope with differences in layout, spelling, design and words and still absorb the meaning. However, machines have a harder time extracting meaning. This is where frameworks like RDFa step in to create what is known as the Semantic Web, which is boiled down to information marked up in a way that machines can easily understand it by giving that information context.
The main advantage of the Semantic Web is improved search, which pulls in information from across the internet and surfaces it in a way that is useful to users. Examples of this are already being seen in major search engines like Bing, Google and Yahoo! using RDFa to surface information in searches that provide more information than just a result.
For one such example, we did a search for Empire Records on the three mentioned search engines, with each returning rich results thanks to RDFa tagging…
However, one of the issues with RDFa is how do you make sure you are using the same attributes as everyone else? This is handled by RDFa vocabularies. Most search engines use a list of well known vocabularies, including Dublin Core (standard and academic), FOAF (people), hReview (reviews) and GoodRelations (products), with Yahoo! providing a decent list of accepted vocabularies and uses on their site (that is relevant to most major search engines). However, Google goes out of their way to use another system, based on their data-vocabulary.org site and is outlined on their support site. (UPDATE: As on November last year, Google have started supporting the GoodRelations vocabulary for products. Thanks Martin!)
Over the next week I’m going to boning up on RDFa tagging and having a play within our favourite CMSs. I’ve already found a list of plugins, extensions and native support (as mentioned with Drupal 7) and am going to report back on how to implement RDFa tagging in your site in the most painless way possible. In the meantime, head over to Mark Birbeck’s rather indepth articles ‘Introduction to RDFa Part I‘ and ‘Introduction to RDFa Part II‘ on A List Apart to read indepth about RDFa.
If you have any tips or hints to help out with RDFa in WordPress, Joomla, Umbraco or Drupal then comment or drop me a tweet at @Ubelly.




[...] This post was mentioned on Twitter by Luke Murphy, Underbelly. Underbelly said: Just posted an introduction to RDFa and the semantic web. (Couldn't fit everything I wanted in so expect a part 2) http://bit.ly/gQAdW8 [...]
[...] week I started with an introduction to RDFa and what it means in relation to the semantic web. For the past week I have been playing with [...]
[...] Good semantic markup will help a lot. Not only using <h1>, <h2>, etc., for headings, but microformats, RDFa markup and/or HTML5 microdata. Google’s Rich Snippets – which provide more detailed, structured information in the basic search results – use such markup to discover and track structured information. See Luke’s article on Ubelly.com – “An Introduction to RDFa and the semantic web” [...]