What is an RDA record?

Editor's Notes

• #4: RDA is based on the FRBR and FRAD conceptual models. Those models propose entities that are useful to identify and describe in order to support resource discovery tasks undertaken by the user. * RDA is currently focused on seven entities. They fall into two groups: W/E/M/I for describing and identifying resources; and P/F/C for describing and identifying the agents responsible for various aspects of resources. * The entities are distinct and RDA assigns separate sets of attributes to them. * The entities in the WEMI group are inter-locked when used to describe a single resource. * An Expression must express one and only one Work, and an Item must exemplify one and only one Manifestation. A Manifestation must realize at least one Expression. * The entities in the PFC group have a different relationship. They have a common Agent super-entity. * RDA provides primary or high-level relationships between the entities in the WEMI group, and between WEMI group entities and Agent group entities. These relationships are represented as reciprocal pairs for separate directions to and from each entity.
• #5: RDA refines the high-level FR relationships with a rich set of relationship designators between the separate entities. * The WEMI group has the primary relationship designators for a single resource. * These are related to the Agent group by designators. * RDA also provides designators for relating individual WEMI entities from different resources. * Plus designators for relating individual Agent entities for authority control and navigation. * How many records are represented here? Is there one type of record for each entity? Or is there just one record for a resource? Or is it something in between?
• #6: RDA provides a set of attributes for each of the entities. * The WEMI group has the richest set. * For example, Work has attributes for Identifier, Preferred title, and Tight ascension and declination; the last is applicable to cartographic works. * Expression, Manifestation, and Item have a similar range of attributes. * Every entity has an attribute for identifier, with human-readable values. This is also required for machine identification in non-linked data applications which are not using URIs. Other attributes are also used for human identification, directly and as components of authorized access points. The remaining attributes are used for describing the entity. * RDA provides attributes for human and machine identification of the Agent entities; description is a secondary function. * Is the set of values of the attributes of a single entity a record? Or the set of values for WEMI? Or for all RDA entities relevant to the resource being described?
• #7: We can look at a typical simple case. * This is the WEMI stack for a single resource; there is an Item, and therefore a Manifestation, Expression, and Work. * The Work has two attribute values, for a label such as the Preferred title or Authorized access point and for a descriptive element such as Intended audience. * The Manifestation and Item have label attributes; the Expression and Manifestation have descriptive attributes. * Is this set of values the record for the resource? * The Work is related to a Person such as its creator. In linked data applications, the entity is identified with a URI. * The Expression is related to a different Person such as a translator, the Manifestation to a Corporate Body such as a publisher, and the Item to a Family such as a donor. The Work is also related to another Work such as an abridgment, and the Expression is related to yet another Work such as a review. * Do we extend the set of values to include the identifiers of the related entities? * Each of the related entities has its own label, and related WEMI entities are components of other WEMI stacks. * Do we extend the record to include the labels of related entities? * Each of the related entities has its own descriptive attributes. * Do we extend the record to include other information about related entities? * Each of the related entities may be related to other entities, such as the Person who is producer of the manuscript that abridges the original Work. Where does the record end?
• #8: The table shows the number of relationships or properties that RDA provides in the rdaregistry.info namespace. Relationships are in two categories: those that are intended to link to entities with URIs; and those that are intended to link to text values. The totals in this table are rounded to the nearest 5.
• #9: The record we are trying to scope can be re-arranged and represented as an RDF graph for use in linked data applications. Using the RDF classes and properties for the RDA entities ensures that the graph can be used in a FRBRized application. * The WEMI stack is in the centre, the focus of the data. * It is surrounded by a shell of entities that are directly related to the WEMI stack. * This is the machine-actionable shell of a “record”. * All of the entities in this shell are linked to labels and descriptive values. * This is the human-actionable shell of a “record”.
• #10: RDA provides a parallel set of properties for relationships and attributes that are not constrained to the FRBR entities. * The WEMI stack is collapsed to an entity Resource; this is not explicit because the properties do not refer to any specific entities. * The unconstrained properties can be used to assemble the same “record” for use in an application that is not FRBRized. This record can be generated automatically by applying a semantic map to the machine-actionable shell of the constrained RDA graph.