In other words, can we channel folksonomies, tagging, user-created knowledge into one coherent structured Web? A Semantic Web? Tom Gruber seems to think so. In Collective Knowledge Systems, he proposes the Semantic Web vision points to a representation of the entity - for example, a city - rather than its surface manifestation. Therefore, one of the problems that we've always had accessing the Web's content is the difficulty in differentiating the city of Paris from the celebrity Paris Hilton when using a search engine.
"What can happen if we combine the best ideas from the Social Web and the Semantic Web?" - Tom Gruber
In many ways, harnessing Web 2.0 technologies and refining them for the Semantic Web has been speculated a great deal. How do we move from collected intelligence to collective intelligence? There are three approaches to realizing the Semantic Web. Here they are:
(1) Expose structured data that already underlies unstructured web pages - Site builders would generate unstructured web pages from a database and expose the data using standard formats (think FOAF)
(2) Extract structured data from unstructured user contributions - Manually dentify people, companies, and other entities with proper names, products, instances of relations
(3) Capture structured data on the way into the system - A "snap to grid" system in which users enter structure to the data and helps users enter data within the structure. (Think of automatic spell check).
Where do librarians come in? We have always used our training to structure content, package it, and disseminate to our users. In our article, Dean and I argue that the catalogue is very much an analogy for how the Semantic Web can organize information in a way that the current Web is unable to do. Recent developments in RDA from the library side offer a promising glimpse into the possibilities for Web 3.0. True, we are only surmising. But let's not prevent us from creating.