Social Meaning on the Web: From Wittgenstein to Search Engines
by Harry Halpin and Henry S. Thompson
One could hypothesize that the biggest question for the Web is whether multiple agents in a decentralized information space can share meaning via the use of uniform resource identifiers (URIs), such as http://www. example.org. On the hypertext Web, this bet was trivial; most of the time a URI would identify a Web page by virtue of allowing access to the Web page itself. However, even in the Web's earliest stages, URIs were for more than just accessing Web pages: they united the previous disparate protocols of the Internet into a single seamless and smooth space of information, where any network-accessible object could be given a URI. Yet, as Tim Berners-Lee noted in his keynote speech to the World Wide Web Conference in 1994, "To a computer, then, the web is a flat, boring world devoid of meaning. This is a pity, as in fact documents on the web describe real objects and imaginary concepts, and give particular relationships between them." The goal of the Semantic Web, then, is to give URIs to "real objects and imaginary concepts" as well as to the "relationships between them." However, there is a fly in the ointment: a Web browser cannot simply access a real object like the Eiffel Tower via HTTP! So, the original question of what a URI identifies, which we could answer earlier by trivially accessing a Web page, transforms into the open question of how agents can determine what non-Web-accessible thing a URI on the Semantic Web identifies in a decentralized manner. This is the defining problem for the evolution of the Web into the Semantic Web.
