Much of the Web’s success rests with its role in enabling information reuse and integration across various boundaries. Hyperlinked Web resources represent a rich information tapestry of content and context, instrumental in effective knowledge sharing and further knowledge development. However, the Web’s simple linking model has become increasingly inadequate for effective content discovery and reuse. At the same time, rigorous but heavyweight solutions such as the Semantic Web have yet to garner critical mass in adoption. This paper analyzes the relative strengths and shortcomings of existing linked data approaches. It proposes a novel, lightweight architecture for the modeling, aggregation, retrieval, management, and sharing of contextual information for Web resources, based on established standards and designed to encourage more efficient and robust information reuse on the Web.