The information we acquire is delivered through various channels; they range from formal source, such as news papers, to informal sources. In the latter case, we learn new information, from our peers or other (social) sources. When it comes, however, to the Internet sources for many people finding the appropriate source or answer becomes a nightmare; and after they find it, they are very likely to forget the answer or the source or both. Rose and Levinson define three types of users' goals: resource seeking, navigational and informational. The last one is the one allows users to aggregate, generate and share the knowledge.

PureC will combine various techniques, currently research within the semantic web domain, such as semantic annotations, rich user profiles, social semantic collaborative filtering, knowledge discovery and sharing.

The research in the PureC project will concentrate on the recommendation techniques, based on community, semantic annotations; we will also research on referencing fine-grained elements, such as web clips or fragments of multimedia streams, on the Web.

Compared to existing solutions, PureC will allow more than just annotating and sharing references to whole web pages; it will enable users to discover, retain, and share the knowledge, consisting of fragments of various information sources, including multimedia streams. Furthermore, PureC will expand the information each user gathered themselves, through recommendations based on discoveries of their peers in the social network.