MADS/RDF: Enhancing Authority Data Representation
In the world of library science and data management, maintaining precise control over names, subjects, and titles is essential. While the Metadata Authority Description Schema (MADS) began as an XML schema, it evolved into a more flexible format known as MADS/RDF. By utilizing the Resource Description Framework (RDF)—a standard model for data interchange on the web—MADS/RDF allows for more sophisticated statements about authority records and their relationship to the real world.
The Distinction Between Records and Reality
One of the primary strengths of MADS/RDF is its ability to differentiate between a controlled record and the actual entity it represents. In this system, a distinction is made between Authorities (and their various Variants), which are the managed records used by librarians, and Real-World Objects (RWOs), which are the actual people, places, or things existing in the physical world.
[ไม่มีภาพประกอบ]
Complementing SKOS
Since 2009, the Library of Congress has utilized the Simple Knowledge Organization System (SKOS) to represent bibliographic authority data. SKOS is a general-purpose framework for representing taxonomies and thesauri. However, the Library of Congress discovered that SKOS alone could not capture the full, complex structure of certain authorities, such as the Library of Congress Subject Headings (LCSH).
To solve this, MADS/RDF was designed to complement SKOS rather than replace it. It achieves this by ensuring its classes and properties are subclasses of existing SKOS items. For instance:
- madsrdf:Authority is a sub-class of skos:Concept.
- madsrdf:authoritativeLabel is a sub-property of skos:prefLabel.
Structuring Authority Items
MADS/RDF provides a dual approach to labeling authority items to ensure both human readability and machine precision. It utilizes authoritativeLabels, which are structured strings (for example, "United States--New Jersey--Essex--Montclair"), alongside collections of typed nodes. These typed nodes break the information down into specific categories, such as {Country, State, County, City}, allowing for more granular data retrieval.
Key Facts
- MADS/RDF is the RDF expression of the original MADS XML schema.
- It distinguishes between controlled authority records and Real-World Objects (RWOs).
- It was developed because SKOS could not fully represent the structure of the Library of Congress Subject Headings (LCSH).
- MADS/RDF properties are designed as subclasses of SKOS properties.
- Authority items use both structured strings and typed nodes for identification.
| Feature | SKOS (General Purpose) | MADS/RDF (Specialized) |
|---|---|---|
| Primary Role | General knowledge organization | Detailed authority representation |
| Class Example | skos:Concept | madsrdf:Authority (Sub-class of skos:Concept) |
| Label Example | skos:prefLabel | madsrdf:authoritativeLabel (Sub-property of skos:prefLabel) |
Frequently Asked Questions
What is the difference between MADS and MADS/RDF?
MADS was originally formulated as an XML schema, whereas MADS/RDF is the expression of that schema using the Resource Description Framework (RDF).
Why wasn't SKOS sufficient for the Library of Congress?
The Library of Congress found that a general-purpose form like SKOS could not present the full, complex structure of specific authorities, such as the Library of Congress Subject Headings (LCSH).
How does MADS/RDF handle real-world objects?
MADS/RDF explicitly distinguishes between the controlled records (Authorities and Variants) and the actual real-world objects (RWOs) that those records describe.
What are typed nodes in MADS/RDF?
Typed nodes are collections of categorized data, such as {Country, State, County, City}, used alongside structured strings to provide a more precise definition of an authority item.