RDFa 1.1 with a rich snippet example

With RDFa 1.1 making its way out of last call, I looked at the examples from Google’s Webmaster Central to see what RDFa 1.1 brings to those. A typical example is the one on reviews; here s where it starts out in RDF 1.0:

<div xmlns:v="http://rdf.data-vocabulary.org/#" typeof="v:Review">
   <span property="v:itemreviewed">L’Amourita Pizza</span>
   Reviewed by
   <span property="v:reviewer">Ulysses Grant</span> on
   <span property="v:dtreviewed" content="2009-01-06">Jan 6</span>.
   <span property="v:summary">Delicious, tasty pizza on Eastlake!</span>
   <span property="v:description">L'Amourita serves up traditional wood-fired 
   Neapolitan-style pizza, brought to your table promptly and without fuss. 
   An ideal neighborhood pizza joint.</span>
   Rating:
   <span property="v:rating">4.5</span>
</div>

The most important change is that RDFa 1.1 has moved away from using XML namespaces as a syntax for identifying vocabularies. In the simplest case, the @vocab attribute can be used to identify a vocabulary as follows:

<div vocab="http://rdf.data-vocabulary.org/#" typeof="Review">
   <span property="itemreviewed">L’Amourita Pizza</span>
   Reviewed by
   <span property="reviewer">Ulysses Grant</span> on
   <span property="dtreviewed" content="2009-01-06">Jan 6</span>.
   <span property="summary">Delicious, tasty pizza on Eastlake!</span>
   <span property="description">L'Amourita serves up traditional wood-fired 
   Neapolitan-style pizza, brought to your table promptly and without fuss. 
   An ideal neighborhood pizza joint.</span>
   Rating:
   <span property="rating">4.5</span>
</div>

The change looks small, but it means that the author has less complexity to worry about. Of course, the @vocab attribute may be placed somewhere up in the hierarchy, e.g., on the body element; this may become interesting if the same page reviews not only a pizza but, say, cannelloni, too.

It is also interesting to compare the last code extract with the microdata example provided by Google for the same case:

<div itemscope itemtype="http://data-vocabulary.org/Review">
  <span itemprop="itemreviewed">L’Amourita Pizza</span>
  Reviewed by 
  <span itemprop="reviewer">Ulysses Grant</span> on
  <time itemprop="dtreviewed" datetime="2009-01-06">Jan 6</time>.
  <span itemprop="summary">Delicious, tasty pizza in Eastlake!</span>
  <span itemprop="description">L'Amourita serves up traditional wood-fired 
   Neapolitan-style pizza, brought to your table promptly and without fuss. 
   An ideal neighborhood pizza joint.</span>
  Rating: 
   <span itemprop="rating">4.5</span>
</div>

Almost identical in terms of simplicity (or complexity, if you prefer); there is no real difference between the two. Which is, actually, to be expected: indeed, the example concentrates on what one could call a “single-vocabulary” case (whereby I mean that the structured data uses terms from within the same vocabulary). One of the main design goals of RDFa 1.1 (compared to the previous RDFa version) was to simplify such simple cases.
However, the same structured data in HTML may be used for other purposes, in which case one would possibly prefer to add additional terms. As a simple example, one might want to add the address of the restaurant using the vcard vocabulary; this is the case where prefixes do come handy. Indeed, one could add a few lines:

<div vocab="http://rdf.data-vocabulary.org/#" typeof="Review">
   <span property="itemreviewed">L’Amourita Pizza</span>
   Reviewed by
   <span property="reviewer">Ulysses Grant</span> on
   <span property="dtreviewed" content="2009-01-06">Jan 6</span>.
   <span property="summary">Delicious, tasty pizza on Eastlake!</span>
   <span property="description">L'Amourita serves up traditional wood-fired 
   Neapolitan-style pizza, brought to your table promptly and without fuss. 
   An ideal neighborhood pizza joint.</span>
   Rating:
   <span property="rating">4.5</span>;
   Address:
   <span property="vcard:street-address">111 Lake Drive</span>,
   <span property="vcard:locality">WonderCity</span>, 
   <span property="vcard:postal-code">5555</span>, 
   <span property="vcard:country-name">Australia</span>.
 </div>

which is just marginally more complicated than the previous case with the additional vcard properties. One could have sprinkled the code with other vocabulary elements, using Dublin Core or other review vocabularies, etc.

At first glance, this code does not look valid in RDFa 1.1, because there is no prefix declaration for vcard. This is correct, and should be added using the @prefix attribute. Except that… the profile mechanism of RDFa might become handy here: one could imagine Google (or the site hosting the review) defining a snippet profile including all the terms that Google recognizes directly but which could also contain prefixes for a number of well known other vocabularies (whether Google uses it in the search pages or not). If so, the only change to the code could be:

<div profile="http://google.profile.example.org" typeof="Review">
   <span property="itemreviewed">L’Amourita Pizza</span>
   Reviewed by
   <span property="reviewer">Ulysses Grant</span> on
   <span property="dtreviewed" content="2009-01-06">Jan 6</span>.
   <span property="summary">Delicious, tasty pizza on Eastlake!</span>
   <span property="description">L'Amourita serves up traditional wood-fired 
   Neapolitan-style pizza, brought to your table promptly and without fuss. 
   An ideal neighborhood pizza joint.</span>
   Rating:
   <span property="rating">4.5</span>;
   Address:
   <span property="vcard:street-address">111 Lake Drive</span>,
   <span property="vcard:locality">WonderCity</span>, 
   <span property="vcard:postal-code">5555</span>, 
   <span property="vcard:country-name">Australia</span>.
</div>

Actually… the vcard case, specifically, may be even easier. Indeed, RDFa 1.1 has the notion of a default profile, i.e., a small set of vocabulary prefixes that are “predefined”. Although the exact list of those prefixes is not yet fixed, we can expect that vcard will be one of those, being one of the most widely used vocabularies around. I.e., the code may be valid RDFa 1.1 after all…

The key point here: RDFa 1.1 brings in an open architecture that scales to authors adding structured data mixing in with ease whatever vocabulary their application needs. Scaling, the possibility to use several vocabularies with ease is RDFa’s real strength, though this is often forgotten. And that openness matters. Take, for example, the BestBuy case. They started with the simple snippet case (yielding real benefit from it, thanks to Google or Yahoo!) but then they added new features with new vocabularies, and extended the functionality for other uses (e.g., to identify shops where a specific item is to available). As so often, openness counts in the long term, also for structured data…