Learning Paths
Learning paths sequence recipes, canonical examples, term pages, and external adoption targets for common implementation goals.
| Path | Audience | Goal |
|---|---|---|
| Model a Person Without Flattening Identity | Developers importing contact, profile, or biography data. | Start with a person, add names and contact points, then keep authority links and display choices separate from identity. |
| Model a Contested or Attributed Claim | Developers handling evidence, provenance, or disagreement. | Represent claims by vantage and evidence rather than overwriting facts into a single global truth slot. |
| Publish Web Structured Data | Developers projecting native GMEOW into web-facing JSON-LD. | Model documents, events, people, and organizations natively, then inspect which fields project cleanly to broad consumers. |
| Ship Offline GTS Documentation | Developers distributing GMEOW snapshots or local docs. | Treat the distribution file, embedded docs, profiles, segments, codecs, and lineage as first-class graph facts. |
| Audit AI and Graph-RAG Pipelines | Developers recording extracted facts, chunks, and tools. | Connect generated claims to source chunks, evidence spans, tools, model context, and provenance before consumers see them. |
Model a Person Without Flattening Identity
Audience: Developers importing contact, profile, or biography data.
Start with a person, add names and contact points, then keep authority links and display choices separate from identity.
Steps
- Read Model Person Names Without a Preferred-Name Slot.
- Copy and adapt Agent Sortals, Person Names, Contact Points, Authority Links.
- Inspect
gmeow:Person,gmeow:PersonName,gmeow:NameUsage,gmeow:ContactPoint,gmeow:displayable. - Check adoption targets
schema,foaf,vcard,wikidata.
Model a Contested or Attributed Claim
Audience: Developers handling evidence, provenance, or disagreement.
Represent claims by vantage and evidence rather than overwriting facts into a single global truth slot.
Steps
- Read Model Contested or Attributed Facts.
- Copy and adapt Contested Authorship, Notability Assessment, Import Lineage, Software Release.
- Inspect
gmeow:StandpointClaim,gmeow:Attestation,gmeow:accordingTo,gmeow:confidence. - Check adoption targets
prov,crminf,wikidata.
Publish Web Structured Data
Audience: Developers projecting native GMEOW into web-facing JSON-LD.
Model documents, events, people, and organizations natively, then inspect which fields project cleanly to broad consumers.
Steps
- Read Publish Documents for Schema.org Consumers, Model Events and Participants.
- Copy and adapt Web Presence, Wedding, Post And Membership, Located Place.
- Inspect
gmeow:Document,gmeow:Event,gmeow:Participation,gmeow:Organization,gmeow:Place. - Check adoption targets
schema,prov,geo,org.
Ship Offline GTS Documentation
Audience: Developers distributing GMEOW snapshots or local docs.
Treat the distribution file, embedded docs, profiles, segments, codecs, and lineage as first-class graph facts.
Steps
- Read Describe Offline GTS Distribution.
- Copy and adapt Dist Package, Import Lineage, Licensed Dataset.
- Inspect
gmeow:GTSDocument,gmeow:GTSProfile,gmeow:GTSSegment,gmeow:usesTransformCodec. - Check adoption targets
dcat,void,spdx.
Audit AI and Graph-RAG Pipelines
Audience: Developers recording extracted facts, chunks, and tools.
Connect generated claims to source chunks, evidence spans, tools, model context, and provenance before consumers see them.
Steps
- Read Model Graph-RAG Dataset Lineage.
- Copy and adapt Grounded Claim, Lillith Dataset, Lillith Pipeline, Agent Trajectory.
- Inspect
gmeow:Dataset,gmeow:Chunk,gmeow:ExtractedEntity,gmeow:EvidenceSpan,gmeow:usedModel. - Check adoption targets
prov,dcat,schema.