Identity and Reference in the Semantic Web

In a 1998 note by Tim Berners-Lee, we read: "... The Semantic Web is what we will get if we perform the same globalization process to Knowledge Representation that the Web initially did to Hypertext". In this quote there is an intriguing analogy between the globalization of hypertext through the Web and the globalization of KR through the Semantic Web. The seamless integration of hypertexts (or local ?webs? of documents) into the global and decentralized information space known as the WWW was largely made possible by a key enabling factor: the introduction of global and unique addressing mechanism for locating and retrieving resources. So the natural question arising from Berners-Lee?s analogy is: what is going to be the key enabling factor for the Semantic Web? What is needed to enable the integration of these local ?semantic webs? into something like the envisaged global Semantic Web? In the seminar, I will argue that the missing factor is a service which, in analogy with what the DNS did for theWeb, can make the addressing mechanism of the Semantic Web global. The addressing mechanism of the WWW was made global by building it on top of the preexisting DNS service; the addressing mechanism of the Semantic Web cannot go global, as no such pre-existing infrastructure is available, and therefore no fast and easy (semantic) integration is readily available. I will propose a solution based on a service, called Entity Naming System (ENS), which is designed to resolve a local URI or even an arbitrary description of a resource (not only, and not mainly, information resources) into a global identifier for that resource. I will discuss the main issues and challenges associated with the design and implementation of a scalable and sustainable ENS. Finally, I will present two simple examples of how existing applications can use the ENS for creating semantically integrated knowledge for the Semantic Web.