> Waiting for the Social Semantic Web

> While there's a lot of hype about Web 3.0, many companies have yet to take any steps towards anticipating the transformations it promises to deliver. This is particularly true with respect to what is called the “Semantic Web”, a movement poised to revolutionize the Web with its vision of structured and linked data. Perhaps this reluctance results from the apparent complexity of the theory behind the Semantic Web.

It's actually quite simple. “Semantics” has to do with meaning, and the Semantic Web deals with changing the way that meaning stored and created on the Web is found, extracted and applied. The Semantic Web is about linking data across the Web, not about links between Web pages. Instead, it links the information stored on Web pages by redefining that information in terms of the relationship between things (like Paris is a part of France and Lennon is a member of the Beatles) and in terms of their properties (like size, weight, age, or price). That way, information published anywhere on the Web will be uniform, interoperable and machine-readable.

As of today, the Web is made to be a source of information for human readers to consult. Our best search engines, for example, do not retrieve information by identifying the information that is on the page, but rather by making inferences about a page's relevance based on the number of links going to and from that page. A search engine using linked data, on the other hand, would retrieve information based on the meaning of the information presented on a page. It would have no problem distinguishing, for example, between Berlin, the German capital city, and Isaiah Berlin, the writer, since, from the semantic point of view, they represent clearly distinguishable objects. By working directly on “meaningful” objects, the Semantic Web will make the information stored on the Web accessible not only to human agents but to intelligent machine agents as well. In order for the vision of the Semantic Web to be realized, data gatherers of all kinds–companies, universities and governments–must publish information in a novel way, one that is not limited to text. This way of publishing would link data sources directly by using shared protocols and “ontologies” (systematic ways to categorize information objects, like people, places, concepts) that describe the meaning of the data. Linking data upgrades the Web from a series of linked documents to a network of linked information objects. If companies have been slow to respond, especially bigger corporations, it is likely because the Killer App, which would work directly on this data in order to transform a process or exploit the connections it reveals, has yet to emerge. This is unlikely to happen until a critical mass of information has been published and linked. Perhaps the question, then, that companies should be asking themselves (at least initially) is not what the Semantic Web can do for them, but what they can do for the Semantic Web. The answer is to publish as much data as possible, as soon as possible, and as precisely categorized as possible.

Growth of the Semantic Web has so far been driven by the scientific community, particularly the life sciences, where information must be shared across domains to solve problems. Among the data communities that have linked up more recently, the Social Semantic Web, or the combination of the Social Web and Linked Data, has emerged as a key sector. Because the Social Web (e.g. Facebook, Twitter) provides access to information in the public domain, linking the data available was possible as soon as tools were developed to export and “semantify” it. Many regard this sector as key to the next generation of applications, even if they have yet to emerge. Given the pace of development, Jim Hendler, a noted Semantic Web researcher, might be right when he claims that Web 3.0 is the Semantic Web and vice versa.
This view is shared by Tim Berners-Lee, often referred to as the inventor of the original Web. He writes, “If HTML and the Web made all the online documents look like one huge book, [Semantic Web technologies] will make all the data in the world look like one huge database”. If Web 1.0 put an unprecedented amount of information at our fingertips, and Web 2.0 enabled users to generate content and information to “self-organize” through crowdsourcing, Web 3.0, as the Semantic Web, would be, as its application to the life sciences already suggests, an engine for collective problem solving and “tasking” where multiple agents can interact and work on the same information.