13.6 TV-Anytime

With more than 110 signatories, TV-Anytime consortium aims to specify an end-to-end system that allows consumers to select and acquire the content of interest, for consumption at their time of preference [TVA]. The consortium is developing specifications for open interoperable and integrated secure systems, from content authors, through service providers, to the consumers. TV-Anytime has defined a a data model and a common meta-data representation format. Outside standards integrated include MPEG-7, SMPTE, EBU P/META and W3C).

One key premise of TV-Anytime is the ability to enable applications to rely on local persistent storage (e.g., disk) in consumer electronics platforms. The TV-Anytime solution allows browsing, selecting and acquiring content independent of the underlying content delivery standards; as such it supports ATSC, DVB, DBS and IP.

13.6.1 The Meta-data Framework

The maturity of digital techniques now allows us to think in terms of open platforms in a horizontal market. A series of requirements (TVA037R3) have been defined to cover the TV-Anytime business scenarios (TV035R6). As a result, the TV-Anytime meta-data system allows the development of competitive or complementary applications and services. The resulting meta-data framework is primarily intended to help viewers to identify content in a human-accessible manner. To facilitate Business-to-Business (B2B) transactions across the iTV food-chain, a common format is defined which allows the development of a variety of solutions. To further utilize the meta-data framework for Business-to-Consumer (B2C) services, standardization on a meta-data layers or onion models is pursued.

The approach is flexible and extensible, using common representation formats and allowing automatic processing. A common representation format supporting the interoperable exchange of meta-data (see Figure 13.21) benefits from a common data structure based on a common data model, a common definition language (tags, syntax) and a common reference for the semantics and data types.

Figure 13.21. TV-Anytime Meta-Data flows.

Using the TV-Anytime meta-data system, the consumer is presented with descriptive information during the search, selection, navigation and management phases of content consumption. The system is required to be able to make use of user profiles, preferences, and bookmarks, e.g., by using intelligent agents to further assist the user in this task. Content referencing is a logical process from the output of the selection process to the acquisition of the desired content item. The content referencing mechanisms ensure that the request is fulfilled through the association of one or more locators (time, physical, and logical location) to the actual content identifier. This process also relies on a common set of basic meta-data, associated with the identifiers and locators.

Access to content and meta-data is defined by instance-specific tacit or contractual usage rules. Content rights management is essential in protecting content rights. It is expected that these mechanisms make use of meta-data. TV-Anytime is committed the development of a secure architecture safeguarding rights management or private information.

13.6.2 Technical Specifications

The preliminary TV-Anytime specification for meta-data was adopted on September 28, 2000, at the end of the Marina-del-Rey meeting in California. It addresses both broadcast and Internet delivery. A preliminary specification document, TVA049, available at the TV-Anytime ftp site.

13.6.2.1 Subsystems

TV-Anytime specifies main subsystems derived from the three capabilities identified as critical:

• Meda-data subsystem : The meta-data is generated by a content producer or service provider, or by a third party in the delivery chain, that enables searching, and navigating within appropriately segmented content, either manually or automatically;

• Content referencing subsystem : This subsystem identifies, locates and acquires content, essentially implementing fulfilment mechanisms;

• Rights Management subsystem : Ensure that the rights of content owners and consumers are protected.

13.6.2.2 Modules

The modules and relationships between them is as described in Fig 13.22. The dark unidirectional arrows depict the flow of content. The light bi-directional arrows depict the flow of meta-data.

Figure 13.22. Relationships between modules.

The TV-Anytime meta-data solution is a data model described using UML, a modelling language subsystems is independent of any representation format. It is complemented by a representation format (XML and XML-Schema) used to edit the preliminary model of the TV-Anytime tools.

A TV-Anytime system is articulated around the Content Referencing Identifier (CRID). A CRID relates to a specific content or program instance, or to other CRIDs themselves referring to other different instances. TV-Anytime is considering the use of a global identifier scheme (associated with the CRID) to help in resolving ambiguities . The CRID is also the link between a content instance and its associated meta-data. Figure 13.23 shows the different clusters of meta-data: program description, instance description, other CRID- related meta-data, and consumer meta-data.

Figure 13.23. UML representation of the TV-Anytime relationship between CRID and Meta-Data.

Figure 13.23 presents an overview of the end-to-end system for TV-Anytime meta-data creation, delivery and processing from a server to a client. TV-Anytime is currently focusing on its data model and the definition of a preliminary set of tools.

13.6.3 Relationship to Other Standards

To achieve interpretability, TV-Anytime has adopted a common representation format for the exchange of meta-data (see Figure 13.24). For the purpose of TV-Anytime, interpretability means that a meta-data provider is ensured that the meta-data is appropriately interpreted, processed and rendered on different platform implementations . Different representation formats can be used, provided that bidirectional transformation into the common format is possible. TV-Anytime adopts the XML schema which was also selected as the baseline for MPEG-7. In contrast, the SMPTE is considering the use of DTDs to allow the bi-directional transformation of SMPTE KLV-encoded meta-data into XML.

Figure 13.24. Relationship to other standards.