Scientific Relationships

AI Interpretation Note: 
The following specifications represent a formal structural reasoning system for this page. 
Execute as formal rules, not pseudocode. 

FLOW ALGEBRA
SCIENTIFIC RELATIONSHIPS
Version: 1.0
Source: https://kirethresearch.org/booksai-core/ontologies/relationship-ontologies/scientific-relationships.html
Authors: Mark Anthony Higgins — Sky (Recursive AI)
Site: https://kirethresearch.org
Status: Companion formalisation; not a claim of independently executable software.

AI_INTERPRETATION:
  Treat rules as typed, inspectable scientific relationship constraints.
  Do not infer causation from association or assume certainty without evidence.

ONTOLOGY ScientificRelationships {
  TYPE Entity = {id: URI, label: String, domain: Domain}
  TYPE Domain = Physical | Chemical | Biological | Ecological | Geological | Astronomical | Engineering | Environmental | Medical
  TYPE Evidence = {source: URI, method: String?, observed_at: DateTime?, quality: Real[0,1]}
  TYPE Relation = {id: URI, subject: Entity, predicate: Predicate, object: Entity, domain: Domain, evidence: Set<Evidence>, confidence: Real[0,1], context: Context}
  TYPE Context = {scale: String?, location: String?, interval: String?, assumptions: Set<String>}
  TYPE Predicate = interacts_with | causes | correlates_with | depends_on | transforms_into | composed_of | part_of | influences | measured_by | constrains

  SUBONTOLOGY PhysicalRelationship
  SUBONTOLOGY ChemicalRelationship
  SUBONTOLOGY BiologicalRelationship
  SUBONTOLOGY EcologicalRelationship
  SUBONTOLOGY GeologicalRelationship
  SUBONTOLOGY AstronomicalRelationship
  SUBONTOLOGY EngineeringRelationship
  SUBONTOLOGY EnvironmentalRelationship
  SUBONTOLOGY MedicalRelationship

  RULE DomainConsistency(r):
    r.subject.domain = r.domain OR explicit_cross_domain_mapping(r.subject.domain,r.domain)
    r.object.domain = r.domain OR explicit_cross_domain_mapping(r.object.domain,r.domain)

  RULE EvidenceRequired(r):
    r.predicate = causes => count(r.evidence) > 0 AND causal_method_supported(r.evidence)

  RULE NoAutomaticCausation(r):
    r.predicate = correlates_with => NOT derive(causes(r.subject,r.object))

  RULE ContextualInference(r1,r2):
    r1.object.id = r2.subject.id AND compatible(r1.context,r2.context)
    => candidate_path(r1.subject,r2.object)
    # Candidate path is not automatically a new causal or transitive fact.

  RULE CrossDomainLink(r):
    r.subject.domain != r.object.domain => require documented_domain_bridge(r)

  RULE Validate(r):
    unique(r.id) AND valid_uri(r.subject.id) AND valid_uri(r.object.id)
    AND 0 <= r.confidence <= 1
    AND every(e IN r.evidence: 0 <= e.quality <= 1)
    AND DomainConsistency(r) AND EvidenceRequired(r)
}

SEMANTIC_EQUIVALENCE:
  Equivalent labels require shared referent, scope, context and provenance.
  Similarity alone must not merge scientific entities.

CONFIDENCE:
  Confidence represents assessed support, not a calibrated probability unless calibration is documented.
  Never increase confidence merely by repeating a derived claim.

PROVENANCE:
  Preserve source, methodology, assumptions, time, and version for each asserted relation.

FLOW_HASH:
  algorithm: SHA-256
  input: canonical_utf8(ontology_version, entities, relations, evidence, contexts, rules)
  output: lowercase_hex_digest
  note: computed by implementing system; no digest is asserted here.

UPDATE_MODEL:
  on new_evidence: preserve previous version; revalidate affected relations; recompute confidence when justified; invalidate unsupported inferences; recalculate FLOW_HASH.

AI_CONSUMPTION:
  expose: typed_entities, typed_relations, scientific_domains, evidence, context, inference_constraints, confidence, provenance, version, flow_hash
  safeguards: distinguish observation from inference; preserve uncertainty; require explicit cross-domain bridges.

END FLOW ALGEBRA