GMEOW Observations Module

• IRI: https://blackcatinformatics.ca/gmeow/slices/observations
• Tier: core
• Group: core

What This Slice Covers

This slice owns 47 terms and contributes 104 mapping or projection rows. Use it when its terms match the native fact you want to preserve; use the linkage tables to see how those facts leave GMEOW for consumer vocabularies.

Dependencies

• gmeow:slices/entities
• gmeow:slices/events
• gmeow:slices/kernel
• gmeow:slices/names
• gmeow:slices/places
• gmeow:slices/temporal

Consumers

• The unified observation stance: every measured, claimed, or sensed value in any slice; the claim layer (GraphRAG provenance design).

Local Map

Examples

Temperature Reading

• Source: slices/core/observations/examples/temperature-reading.ttl
• GMEOW terms: gmeow:Measurement, gmeow:Place, gmeow:ScalarQuantity, gmeow:SoftwareAgent, gmeow:hasUnit, gmeow:methodInstrumentalReading, gmeow:name, gmeow:observationMethod, gmeow:observationResult, gmeow:observedFeature
• External prefixes: xsd

# SPDX-FileCopyrightText: 2026 Blackcat Informatics® Inc. <paudley@blackcatinformatics.ca>
# SPDX-License-Identifier: CC-BY-4.0
#
# Worked example: a measurement is a reified Observation ( P11). A
# gmeow:Measurement records WHAT was observed (gmeow:observedFeature), HOW
# (gmeow:observationMethod), and from WHERE (gmeow:vantage — the sensor). Its
# result is NOT a bare number on the observation: the scalar lives in an
# entity-valued gmeow:ScalarQuantity wrapper that bundles the value with its UNIT
# and uncertainty, so a temperature is never asserted free-floating (P11 — every
# value carries its frame). The unit is aligned to QUDT by reference (P5).
@prefix gmeow: <https://blackcatinformatics.ca/gmeow/> .
@prefix ex:    <https://blackcatinformatics.ca/gmeow/examples/observations/> .
@prefix xsd:   <http://www.w3.org/2001/XMLSchema#> .

ex:sensor a gmeow:SoftwareAgent ; gmeow:name "Rack temperature sensor 7"@en .
ex:room   a gmeow:Place ; gmeow:name "Server room A"@en .

# --- The measurement: observed feature, method, vantage, and an entity result.
ex:reading a gmeow:Measurement ;
    gmeow:observedFeature   ex:room ;
    gmeow:observationMethod gmeow:methodInstrumentalReading ;
    gmeow:vantage           ex:sensor ;
    gmeow:observationResult ex:temp .

# --- The result wrapper: value + unit + uncertainty, not a bare literal.
ex:temp a gmeow:ScalarQuantity ;
    gmeow:quantityValue       "22.5"^^xsd:decimal ;
    gmeow:quantityUncertainty "0.1"^^xsd:decimal ;
    gmeow:hasUnit             <http://qudt.org/vocab/unit/DEG_C> .

Terms

Classes

Properties

Individuals

Linkages

• Rows: 104
• Projection profiles: crmarchaeo, iptc, ivoa, loinc, qb, schema-org, sosa
• External vocabularies: bbc, crm, crmarc, crminf, fibo-fnd-acc-cur, iao, iptc, loinc, np, oa, obi, oboe, obscore, om, prov, qb, qudt, rdf, schema, snomed, sosa, sta, wd

Guide

Observations — the universal claim relator

Slice: https://blackcatinformatics.ca/gmeow/slices/observations · tier: core One reified structure for every measured, sensed, claimed, or asserted value in the graph.

Most vocabularies scatter their claim machinery: SOSA for sensor readings, CIDOC E13 for attribute assignments, bespoke relators for names, rights, and identities. GMEOW collapses them into one structure — [Observation](../reference/classes/gmeow-Observation.md) ≡ [Measurement](../reference/classes/gmeow-Measurement.md) ≡ [Standpoint](../reference/classes/gmeow-Standpoint.md) ≡ [IdentityFacet](../reference/classes/gmeow-IdentityFacet.md) ≡ [NameUsage](../reference/classes/gmeow-NameUsage.md) ≡ [RightsStatement](../reference/classes/gmeow-RightsStatement.md) ≡ [KinRelationship](../reference/classes/gmeow-KinRelationship.md) (Principle 9: every vantage is a co-equal facet; no frame is privileged). An observation is a reified gufo:Relator mediating a vantage (who holds it), an observedFeature (what it is about), and an observationResult (what was found). SOSA/SSN, PROV-O, and CIDOC E13 are aligned by reference (Principle 5).

This slice is the Claim end of the claim spine (Source → Chunk → EvidenceSpan → Claim, Principle 14): an LLM output, a sensor reading, and a census entry are all observations — stored with a vantage, never adjudicated by rank. The slice is also home to the universal Quantity/MeasuredValue construct (ScalarQuantity alias precedent): every scalar value travels as a value × unit/frame × determinacy × provenance bundle, never a bare literal.

The core relator

gmeow:Observation

The universal claim construct: a gufo:Kind under gufo:Relator. EL-axiomatised to mediate at least one vantage and one feature (open-world); the closed-world "exactly one" cardinality is SHACL's concern (Principle 7). Competing observations of the same feature coexist rather than collapse — disagreement is data.

gmeow:vantage

The agent or standpoint the observation is made from — the reified counterpart of the gmeow:accordingTo annotation axis. The flat↔reified pairing is documented, not axiomatised: accordingTo is an annotation property (keeping the OWL 2 DL downcast clean, Principles 2–3), so vantage ⊑ accordingTo is realised in the projection layer. Range is gmeow:Entity: a vantage may be a bare agent or a first-class Standpoint. Non-functional — joint observations are valid.

gmeow:observedFeature

The feature of interest — deliberately owl:Thing-ranged, because anything can be observed: a place, a person, a proposition, a quality value. Per-kind narrowing is SHACL's job, not the open ontology's.

gmeow:observationResult · gmeow:isResultOf

The entity-valued result (a coordinate set, an instant, a quantity) and its inverse — the explicit provenance leg of the ScalarQuantity alias precedent bundle. Uniformly entity-valued: scalar literals live on a ScalarQuantity, never directly on the observation. Non-functional: one observation may yield results in several frames; one consensus result may flow from many observations. A property chain (isResultOf ∘ hasReferenceFrame ⊑ hasReferenceFrame) lets results inherit the observation's reference frame (Principle 11).

gmeow:observationMethod · gmeow:observationType

How it was done versus what kind of act it was — both open value vocabularies (individuals, never subclasses). Method is functional and constitutive: a different method is a different observation. Type is non-functional: a calibrated reading can be both measurement and derived inference. Seeds cover measurement, sensory, standpoint, derived, simulation, identity, naming, rights, kinship, and streaming.

gmeow:Measurement · gmeow:SensoryObservation · gmeow:StandpointClaim

The shared specialisation spine, declared here so every consumer slice (temporal, places, sensory, standpoint) deepens the same parent. StandpointClaim is the assertion form: vantage = the standpoint, observedFeature = the proposition, observationResult = the modality (unequivocal, probable, conceivable, refuted) — the structure the deception and falsehood doctrine builds on.

gmeow:facetSubject · gmeow:facetVantage

The observation-spine bridge bridge idiom: domain relators (IdentityFacet, NameUsage, RightsStatement, VersionMembership) plug their role properties into the core roles via rdfs:subPropertyOf (facetSubject ⊑ observedFeature, facetVantage ⊑ vantage, likewise usageNamer, usageAppellation, statementAbout, membershipAuthority). A generic consumer asks "all observations about Alice" and gets her names, rights, identities, and coordinates without knowing any domain vocabulary. When gmeow:selfAsserted is true, the facetVantage is the person themselves — self-assertion is top authority (Principle 9).

Streams and scalar results

gmeow:Stream

A time-ordered observation sequence (stream-to-trajectory design) — streamOf exactly one tracked entity, composed via streamSample, hosted on a streamPlatform, produced by a streamSensor over a streamInterval. Ordering is implicit in sample timestamps, never an asserted list (Principle 12); deriving a continuous trajectory from the stream is the solver's job.

gmeow:Quantity · gmeow:ScalarQuantity · gmeow:MeasuredValue

Three equivalent names for one construct — the domain-neutral scalar bundle: gmeow:quantityValue (+ gmeow:quantityUncertainty) for the number, hasUnit / hasReferenceFrame for the frame (Principle 11), hasDeterminacy for ontic vagueness, isResultOf for provenance. Temperatures, counts, probabilities, masses — all of them, one shape.

gmeow:MonetaryAmount

Promoted to core in the dependency refactor: money is frame-relative quantity machinery, not finance-domain. gmeow:monetaryValue carries the decimal; gmeow:currency (⊑ hasReferenceFrame, functional) names the currency frame — a value without its currency is ill-formed (Principle 11). Canonical superset of schema:MonetaryAmount and the FIBO monetary amount.

Solver boundary & alignment

Aggregation, calibration, consensus derivation, trajectory reconstruction, and inter-frame conversion are computations, never assertions (Principle 12): the slice records co-existing observations; the solver layer ranks, fuses, or projects them per consumer policy. Suppressing a contested observation is projection-time filtering, never erasure (Principle 10).

Dependencies

Depends on kernel, entities, events, names, places, and temporal. Consumed by every slice that measures, claims, or senses anything — which is to say, by every slice; the AI claim layer (GraphRAG provenance design) is its flagship consumer.