Metadata standards and profiles
Metadata standards and profiles explain how to describe a resource such that it can be understood, unambiguously, by others.
- an abstract metadata standard, describes a theoretical information model for a standard, whereas a metadata standard encoding describes how to implement an abstract model, e.g. using XML. There are relatively few standards, usually made by international bodies such as ISO.
- a metadata standard extension, extends the informational model for a standard, typically to cover additional concepts. These extensions are implemented by extensions to corresponding standard encodings.
- a metadata profile, builds upon a metadata standard, typically to standardise how it is used for a particular topic,
discipline or within an initiative. They do this in various ways, such as:
- written guidance/notes
- restricting how, or which, elements can be used
- increasing the obligation for elements (e.g. optional to mandatory)
Note: Importantly, a profile cannot decrease the obligation for an element (e.g. mandatory to optional). I.e. profiles can only make things more restrictive, not permissive.
To ensure compliance, metadata standards and profiles typically include a means to programmatically check records contain:
- any required elements
- any required/controlled element values
- elements in the correct order
- no unexpected or disallowed elements, element values or element structure
Metadata standards typically use XML schemas and schematrons to determine if a record is valid:
- a schema checks the structure of elements in a record
- a schematron checks specific elements in a record using a series of rules and patterns
Notably a schematron can use conditional logic, such as to ensure an element is included if other elements are not.
Metadata vocabularies and code lists
Metadata vocabularies and code lists describe lists of controlled terms such that meaning can be expressed unambiguously and consistently by others.
A vocabulary can take different forms, from simple lists of values to complex graphs. Simple vocabularies are often encoded as CSV files or similar, whilst those that are more complex, typically use linked data concepts, such as the Simple Knowledge Organisation System (SKOS)
A code list is a concept from the ISO 19115 standard. It consists of a code list, optional code space and code list value attributes. The code list, is a URI to a list of terms, of which one is selected as the code list value.
For example, a code list is used to control the types of date in an ISO 19115 record
- the code list for date types is:
- possible code list values for this code list include