What is Descriptive Metadata?
Descriptive metadata identifies, explains, and supports discovery of an asset. Common fields include its title, caption, keywords, subjects, creator, and other information about intellectual content.
How Descriptive Metadata works
Descriptive fields represent what an asset is about and how people should recognize it, rather than how its bytes are encoded or administered. Records may use free text, names, dates, classifications, and controlled subject terms, with different schemas expressing similar ideas at different granularity. Search engines index these values for discovery, grouping, and display. In media operations, catalogers and automated classifiers enrich the record while governance determines which value is authoritative.
Key facts
- Controlled vocabularies can assign stable identifiers to subjects even when display labels change or have translations, improving joins and recall beyond literal keyword matching.
- Mappings among schemas such as Dublin Core, MODS, and IPTC fields can be lossy because cardinality, role qualifiers, and field semantics do not always align one to one.
- Embedded metadata travels with a file, while repository records and sidecars can describe compound or changing context; synchronization rules are needed when more than one copy is editable.
When Descriptive Metadata matters
Index descriptive metadata when users need to search, filter, or recommend assets by meaning. Inconsistent vocabularies and missing fields reduce recall, so validation and controlled terms may be necessary.