An active document framework is a self-representable, self-explainable, and self-executable document mechanism. A document’s content is reflected in four aspects: granularity hierarchy, template hierarchy, background knowledge, and semantic links between fragments. An active document has a set of build-in engines for browsing, retrieving, and reasoning, which can work in a way best suited to the document’s content. Besides browsing and retrieval services, the active document supports intelligent information services such as complex question answering, online teaching, and assistant problem solving. The client side service provider is only responsible for the retrieval of the required active document. The detailed information services are provided by the document mechanism. This improves the current Web information retrieval approach by raising the efficiency of information retrieval, enhancing the preciseness and mobility of information services, and enabling intelligent information services. A tool for making semantic links in a document and an intelligent browser have been developed to support the proposed approach, which provides a new type of web information service.