Semantic layer
Governed metric and SQL abstraction layer
The Core Argument
The article distinguishes metric semantics from formal knowledge representation. Semantic layers keep calculations consistent; ontologies, knowledge graphs, and context graphs help AI systems interpret the world they operate in.
Ontology
Formal concept, relationship, and constraint model
Knowledge graph
Graph representation of entities, facts, and relationships
Context graph
Operational graph for decision context and procedural traceability
Knowledge architecture
Practice of designing durable semantic systems for people and AI
Architecture Comparison
| Architecture | Optimizes For | Primary Question | AI Relevance |
|---|---|---|---|
| Semantic layer | Measurement | What is X? | Trusted calculations |
| Ontology | Meaning | What kind of thing is this? | Concepts, constraints, inference |
| Knowledge graph | Context | What is connected? | Machine-readable domain structure |
| Context graph | Operational explanation | Why was this allowed? | Decision traces, authority, precedent |
Claims
Metric consistency is not meaning
Semantic layers solve governed calculation, not complete organizational understanding.
AI needs context
AI-native architecture needs explicit context, constraints, and inference-ready relationships.
FAQ
Questions are modeled in the companion RDF as named schema:Question and schema:Answer entities.
Is a semantic layer enough for AI?
It is enough for governed metric lookup and calculation consistency, but not for reasoning that requires domain concepts, constraints, causal context, or procedural meaning.
What does an ontology add?
An ontology makes concepts, relationships, properties, and constraints explicit so machines can interpret and reason over domain meaning.
What does a context graph add beyond a knowledge graph?
A context graph captures the operational why: decisions, authority, precedent, procedures, execution history, and supporting evidence.
What should organizations build now?
Organizations should keep governed metrics where those work, then add formal knowledge representation where AI systems must reason about concepts, procedures, constraints, and decisions.
Glossary
The glossary mirrors the RDF schema:DefinedTermSet; each visible term resolves through URIBurner.
Semantic layer
A governed abstraction layer for metric definitions, dimensional models, SQL generation, and reporting consistency.
Ontology
A formal specification of domain concepts, properties, constraints, and relationships.
Knowledge graph
A graph-based representation of entities, relationships, and facts that provides structured domain context.
Context graph
A knowledge graph focused on operational context, decision traces, authority, precedents, procedures, justifications, and execution histories.
Knowledge architecture
The discipline of designing semantic systems that represent concepts, relationships, constraints, context, and inference rules.
HowTo
The article's practical decision logic is represented as RDF schema:HowToStep entities.
Use a semantic layer for governed metrics
Choose a semantic layer when the task is consistent calculation, dimensions, joins, and BI-facing answers.
Use an ontology for domain meaning
Choose an ontology when the task requires explicit classes, relationships, constraints, and inference.
Use a context graph for operational reasoning
Choose a context graph when the task requires decision traces, authority, procedures, precedents, or auditability.
Knowledge Graph Explorer
Explore Knowledge Graph using SPARQL
Choose a named graph and query recipe, edit the SPARQL if needed, then open the encoded URIBurner query. Query resources are also modeled in the companion RDF as schema:SoftwareSourceCode.
SELECT uses text/x-html+tr. DESCRIBE and CONSTRUCT use text/x-html-nice-turtle.