For efficient access to multimedia content the media data has to be enriched with additional information about the content's semantic description and functionality, e.g. style sheets for rendering. Current approaches for semantic modeling of multimedia content store the information about the content's semantics and functionality in different files and at different locations, which makes the retrieval and reuse of multimedia content very difficult. We have proposed Enhanced Multimedia Meta Objects (EMMOs) as a new means for semantic multimedia modeling that indivisibly combines multimedia content with its description and functionality. In this paper, we give an overview of the structure of EMMA, a query algebra, which is adequate and complete with regard to the EMMO model, enables query optimization, allows the integration of ontology knowledge, and supports distributive collaborative authoring.