The global TV-Anytime Forum is an association of approximately 146 member organizations which seeks to develop specifications to enable audio-visual and other services based on mass-market high volume digital storage in consumer platforms. The Forum has recently published a version 1.1 Metadata specification to accompany its Content Referencing and System Description specifications, together with XML Schemas and data dictionaries. The TV-Anytime Forum is "developing open specifications for interoperable and integrated systems to allow broadcasters and other service providers, consumer electronics manufacturers, content creators and telecommunications companies to maximise effective use of high-capacity digital storage in consumer devices." Metadata types defined by the specification include: (1) Content Description Metadata, which describes content independently of any particular instantiation of that content; (2) Instance Description Metadata, used for linking content metadata to content; (3) Consumer metadata, "modeled as description schemes for describing usage history information gathered over extended periods of time"; (4) Segmentation metadata, "which supports the ability to define, access and manipulate temporal intervals (i.e., segments) within an AV stream. For the purpose of interoperability, the TV-Anytime Forum has adopted XML as the common representation format for metadata. XML offers many advantages: it allows for extensibility, supports the separation of data from the application, and is widely used. TV-Anytime uses the MPEG-7 Description Definition Language (DDL) to describe metadata structure as well as the XML encoding of metadata. DDL itself is [2001-08] based on XML schema 'W3C Proposed Recommendation'; the MPEG-7 data types and description schemes in the TVA Metadata Specification are currently taken from the MPEG-7 MDS Final Committee Draft (FCD)."

In the context of TV-Anytime, metadata means "descriptive data about content, such as program title and synopsis. We call such metadata 'attractors' because they can attract a consumer to content. Attractors allow consumers to find, navigate and manage content from various sources. In addition to attractors, metadata as defined by TV-Anytime also includes information about user preferences and history. User preference information, such as favorite actors or TV shows, is included within the scope of TV-Anytime metadata to allow software agents to select content on the consumer's behalf. The set of metadata described in this document was selected in order to satisfy the usage scenarios listed in the TV-Anytime business models requirements document R-1."

The metadata specification contains three appendices. Appendix A (TV-Anytime Usage History Thesaurus) "contains a thesaurus for valid terms of the ActionType element in the Usage History Description Scheme. Appendix B supplies the TV-Anytime Genre Dictionary. Appendix C (TV-Anytime Description Schemes) aggregates the TV-Anytime description schemes listed in the document into a single file, tva11.xsd; these description schemes, along with the MPEG-7 DSs [description schemes] referenced herein, constitute the normative set of TV-Anytime metadata, version 1.1. The SP003V11 [S-3] distribution package for the TV-Anytime Metadata specification introduces new technology in particular metadata for segmentation and includes updates in the semantics in order to maintain maximum compatibility with the version of XML in use in MPEG and W3C."