INST0089 Chap.3 Standards and Interoperability
Standards and Interoperability
Standards and Interoperability separates the different agreements that allow records to move and remain meaningful. A standard may govern data structure, descriptive content, permitted values or exchange. Sharing field labels is not sufficient when communities form dates, names or relations differently. Sharing a file syntax is also insufficient when the receiver interprets Creator, Owner or Source in another way.
The chapter treats interoperability as a bounded outcome for defined tasks. A simple discovery portal can accept a lossy mapping that would be unsuitable for preservation or rights administration. Crosswalks are therefore documented interpretations. They compare source and target definitions, state transformation rules and expose information loss.
Local extensions can preserve community meaning, but their mapping and governance must be maintained as standards, vocabularies and systems change.
What this chapter covers
- 01
Structure, content, value and exchange standards
- 02
Semantic and technical interoperability
- 03
Crosswalk design and transformation rules
- 04
Exact, broad and lossy mappings
- 05
Local extensions and documented loss
- 06
Versioning, governance and review
Worked example · free
Crosswalk a local image register
- 2Define whether the record describes a photograph, depicted work or digital reproduction.
- 2Compare the source and target definitions before mapping Maker or Photographer to Creator.
- 2Review Topic tag against the target Subject vocabulary and test identifier persistence.
- 2Document any merged roles, broader terms or nonpersistent links as mapping loss.
Key terms
- Semantic interoperability
- Exchange in which receiving communities can interpret and use the intended meanings.
- Technical interoperability
- Exchange enabled by compatible protocols, formats or machine-readable structures.
- Mapping loss
- Meaning or structure that cannot be preserved when data moves to another scheme.
- Local extension
- A project-specific property or value added beyond a shared standard.
Standards and Interoperability FAQ
Can matching field labels be mapped automatically?
Not safely. Two systems may use the same label for different relationships, such as a rights holder and a physical custodian both called Owner. Compare definitions, record subjects, value rules and examples before declaring equivalence. A label match is evidence to investigate, not proof of semantic identity.
What makes a crosswalk defensible?
A defensible crosswalk stores the source definition, target definition, mapping type, transformation rule, examples and known loss. It also names the use case. That documentation lets later reviewers distinguish an exact relationship from a broad discovery mapping and revisit the decision when either standard changes.
Is a valid export interoperable?
Validity proves only the syntax or structural rules tested. Interoperability also requires interpretable meanings, compatible values, stable identifiers and successful use by the receiver. Test the actual operation, such as filtering, linking or transformation, and report the information that was flattened, broadened or omitted during exchange.
Assessment move
Build a four-layer standards table. Put structure, content, values and exchange in separate rows, then add one example of a rule and one failure for each. This prevents a file format from being treated as a complete metadata standard. Practise crosswalks with a small set of source and target fields.
Write definitions beside labels, classify each mapping as exact, narrower, broader, combined, conditional or unmapped, and explain the consequence for a named user task. Use difficult cases where two local roles merge into one target property or one local field mixes several concepts. Preserve the original data and record every transformation, because a discovery mapping should not become the archival master by accident.
Test technical and semantic interoperability separately. First validate the file and ensure it can be parsed. Then inspect whether dates, identifiers, roles, repeated values and relationships retain their intended meaning. Run a receiver-side task and record the outcome. If the task succeeds only after manual explanation, documentation or mapping is incomplete. Revise conclusions so they are bounded.
Replace “the systems are interoperable” with a statement naming the supported task, standard version, transformation and known loss. Finally, plan governance. Record who owns the mapping, when it was reviewed, which vocabulary versions apply and what change would trigger another review. Standards and local systems evolve, so an unmanaged crosswalk becomes unreliable even when its original reasoning was sound.
Add a change scenario to each mapping exercise. Retire one target term, split one source field or introduce a new local role, then identify every rule and sample record that needs review. This reveals whether the mapping documentation is operational or merely descriptive. Keep an exception list rather than forcing every value through the nearest available target.
Classify each exception and decide whether to retain a local extension, broaden the target or leave the value unmapped with a clear reason. Revisit that decision whenever a vocabulary or receiving service changes version.