New Google Tech Fuses SEO and Semantic Web

By Dee-Ann LeBlanc (@deele) May 18, 2009

7 comments

As we've discussed before, the semantic web is supposed to help with issues where computers just have no idea of what they're seeing in any context. The problem was getting people to take advantage of this technology in any real numbers.

Well, if you care about Search Engine Optimization (and who doesn't?) then it's time to wake up and smell the semantics. Otherwise, your search results will plummet as SEO gold goes to those who take advantage of Google's new tool: the Rich Snippet.

And Dries Buytaert of Drupal (news, site) seems to agree completely.

What's A Rich Snippet?

When you look at a page of Google search results, you typically see a snippet of material from that page. The goal of this snippet is to help you see at a glance if that page is what you're really looking for.

Without understanding anything about the content itself, that's all Google can do. Or, at least, it was. Google Rich Snippets allow you to give context to your pages through one of two popular semantic web technologies: microformats and RDFa.

Right now you can apply Rich snippets to people and to product reviews, with more types coming in the future.

An Example of a Rich Snippet

We'll offer a quick example here. Say that a savvy online publication wants to make sure that when a name appears in their articles, people and search engines can easily understand the context of what they're seeing.

In particular, let's take the fairly common name of Jane Smith. In order to make sure that you understand that they're talking about Jane "J.J." Smith, popular author, the site editors can use either microformats or RDFa to annotate each mention.

Say that they originally had:

<a href="http://www.example.com/">Jane Smith</A>

With microformats, you specify that you're defining a person with contact information with the term "vcard." You use the div and span tags to logically group the information, with div used for multiple pieces of information and span used for single pieces of information.

Using Google's "Marking Up Structured Data" documentation, they might change this link to:

<div class="vcard">
   <span class="fn">Jane Smith</span>
   <span class="nickname">J.J.</span>
   <span class="url">http://example.com/</span>
   <span class="role">Author</span>
<a href="http://www.example.com/">Jane Smith</A>
</div>

Now they've added the context of Jane's full name, her nickname, her web site, and that she's an author. To show this same information in RDFa, they might have:

<div xmlns:v="http://rdf.data-vocabulary.org/#" typeof="v:Person">
   <span property="v:name">Jane Smith</span>
   <span property="v:nickname">J.J.</span>
   <span property="v:url">http://www.example.com/</span>
   <span property="v:role">Author</span>
<a href="http://www.example.com/">Jane Smith</a>
</div>

Drupal is Taking a Leadership Role

Does all of this code make you nervous? Adding this information can be a lot of work, and whether you have to type every character by hand or your web content management system offers you tools to make it easier will ultimately make a huge difference. Yet a lot of Content Management Systems haven't been showing too much interest in the semantic web.