Glossary · Information modeling
Semantic control
In information architecture, semantic control is governance that preserves the intended meaning of terms, states, signals, requirements and instructions across systems and changes.
- Information architecture
- Ontology
- Semantic web
In one sentence
Semantic control keeps the meaning of terms, states, signals, requirements and instructions stable across systems, versions and teams.
Example
A manufacturer keeps one controlled definition of “safe state” for each machine type and links the PLC state list, the risk assessment and the operating manual to it.
How it applies
- Technical documentation: Controlled terminology, glossaries with stable identifiers and reviewed definitions are the core tools. Each term used in safety-related text should have one approved meaning.
- Integrated systems: Signals, states and fields need the same discipline as words: a documented meaning, unit and owner, and a review whenever a change could alter their interpretation. This is what schema governance alone does not cover.
- Evidence: Traceability from requirements to implementation and instructions only works if the linked items mean the same thing at both ends.
- AI and retrieval: Knowledge graphs, IRIs and explicit definitions give AI systems a fixed reference, which reduces misinterpretation when content is retrieved or generated.
Semantic control vs. semantic drift
Semantic drift is the problem: meaning changes while labels stay the same. Semantic control is the countermeasure: it makes changes of meaning visible, reviewed and propagated.