The Future of Linked Data
Google, and other search engines, use linked data in web pages to produce rich snippets in search results. However, it is not generally understood how this detailed summary information may be used in the future. Currently, several obstacles must be overcome before the full potential of rich snippets can be realised.
After the implementation of the Web 2.0 initiative, the basic function of the World Wide Web turned to the sharing of data across all boundaries, such as application and enterprise boundaries. Linked data is one of the primary tools used to expose, connect and share information through the Web. To share linked data more effectively, a new data layer is required in networking protocol to make sure the linked data is interoperable between applications and systems. Using rich snippets to increase Web semantics for applications is an important first step in the process of overhauling Internet architecture in the future to accommodate improved data sharing and dissemination.
Rich Snippets Today
Table on the left: One million Web pages sampled from the Internet in June 2010. Percentages in parenthesis: actually used for generating Rich Snippets .
Rich snippets were introduced in 2009 to display small bits of detailed summary information in search results. They were developed by Google but built on open standards, so rich snippets were later adopted by other search engines. Rich snippets can support various markup languages to help identify the information to be presented, such as event or item information.
A study of one million Web pages showed that very little actual markup data embedded in the pages is used to create rich snippets. This is mostly because the data is labeled incorrectly, the data is not formatted correctly and the wrong data is marked up. However, a large market exists that depends on correct rich snippet information.
The Future of Rich Snippets
Some common search queries are notoriously inaccurate, but a well-defined system of linked data and rich snippets can improve searches. This system could also be used to bring Facebook functionality to search results and to enrich searches for multimedia content.
One way to accomplish this goal is to use triple-centric networking, which is derived from content-centric networking. This system correlates linked data packets to interest packets that are shared between nodes. Converting Web scale triple propagation to a triple-centric networking layer could increase the efficiency of using data from rich snippets in search results, which is a vast improvement over the host-to-host networking system used today.
Because rich snippets are highly visible, supported by Google and can affect businesses, it makes sense to use them as an element in linked data. By using extensions for rich snippets, common data can be interlinked with a user’s social graph to provide enhanced search results. The information could be correlated by search engines through the addition of a linked data layer to current network protocol and through the implementation of triple-centric networking.
 K. Goel and P. Gupta. Google Rich Snippets. Semantic Technology Conference, 2010. http://semtech2010.semanticuniverse.com/sessionPop.cfm?confid=42&proposalid=2745