Skip to content

Batch 0.4.0 — Concept Selection (No Drafting Yet)

Status: Selected for admission; not yet drafted or reviewed
Date: 2026-07-10
Governing rules: ADR-0018, docs/methodology/crosswalk-runbook.md, docs/crosswalks/candidate-concepts.md

Why a new selection pass

The original matrix-derived queue of 19 candidates is exhausted:

Release Concepts
0.1.0 8 reviewed (original inventory)
0.2.0 +9 (≥3-source rung)
0.3.0 +9 reviewed, 1 off-ramped (Quality Space/Quale)

docs/crosswalks/candidate-concepts.md still lists 540 source classes. Most remaining rows are either (a) already bucketed under one of the 26 reviewed concepts as sub-species, (b) Ungrouped linguistic/domain leaves flagged "likely NOT core", or (c) single-source Ungrouped terms. This pass does not draft crosswalks. It only admits the next batch by re-clustering evidence already visible in the public backlog.

Admission method

Per ADR-0018 and the 0.3.0 precedent:

  1. Prefer multi-source themes whose classes are currently scattered across existing reviewed buckets (or left Ungrouped) rather than inventing labels with no source footprint.
  2. Do not exclude a same-tier candidate on pre-verification "weakness" impressions (crosswalk-runbook.md).
  3. Single- and two-source admissions remain allowed when the term is a source's own core/base concept (ADR-0021 / runbook reviewed-eligibility), but this batch leads with ≥3-source clusters.
  4. Off-ramp remains available later if Step 3 verification shows a source-side equivalence to an already-reviewed concept.

Admitted batch (4 concepts)

# Working label Provisional slug Seed source count Seed evidence (from candidate-concepts.md triage) Why it is not already covered
1 Dependence / Dependent Entity dependence-dependent-entity ≥5 BFO independent / specifically dependent / generically dependent continuant; GFO Dependent / Independent; YAMATO dependent entity, specifically dependent, generically dependent; UFO externalDependence; DOLCE dependent-place (adjacent) Today these rows are split across Continuant-Occurrent, Quality, Information Artifact, and Boundary/Site. No dedicated crosswalk compares dependence as its own organizing distinction.
2 Material / Immaterial material-immaterial ≥4 BFO material entity / immaterial entity; GFO Material_* family; TUpper MaterialObject; UFO material; GUM material-quality ascriptions (adjacent) Materiality is currently folded into Object / Continuant-Occurrent / Boundary buckets. Several sources treat material/immaterial as a top-level partition, not merely an Object subtype.
3 Set / Class set-class ≥3 SUMO SetOrClass / Set / Class; DOLCE set; GFO Set; GUM UMSet / DisjunctiveSet (OrderedSet already moved to List/Sequence) Currently parked under Universal/Type, which focuses on type/kind/category. Extensional set/class collections are a recurring cross-source cluster that Universal/Type deliberately does not settle.
4 History history 2 BFO history; GFO History Both sit under Process today. A dedicated pass can test whether History is a distinct occurrent kind (BFO: history of a continuant) vs. ordinary Process — same two-source admission style used in 0.3.0.

Explicitly not admitted this batch

Cluster Reason
Function Already in scope of disposition-capacity (BFO function is an etiological disposition subtype there).
Feature Three sources, but divergent senses (DOLCE feature → Boundary/Site; YAMATO feature → Quality; TUpper ShapeFeature). Needs a scoping spike before admission; not silently dropped — deferred.
Configuration Largely covered by situation-state-of-affairs (GFO/GUM Configuration*).
Relator Already treated inside relation (GFO/UFO relator evidence).
GUM Ungrouped linguistic leaves (~51) Flagged "likely NOT core"; no cross-source counterpart pattern.
TUpper Ungrouped extrema (Max/Min/ZEX/…), SUMO Graph/GraphElement, GFO Configuroid/Persistant, YAMATO substrate Single-source or not yet showing a multi-source cluster; remain Ungrouped for later triage.
Domain leaves (YAMATO living organism / chemical compound / morphological whole) Explicitly "likely NOT core".

Provisional Feature spike (not in batch)

If a later session wants a fifth 0.4.0 slot, run a scoping-only check on Feature before drafting: confirm whether any source treats Feature as a first-class category independent of Boundary/Site and Quality. Until that check exists, Feature stays out of the admitted set.

Next steps (when drafting begins)

For each admitted concept, follow docs/methodology/crosswalk-runbook.md:

  1. Session branch from current main
  2. Fetch/verify all 8 sources (do not trust this selection table as final source counts)
  3. Off-ramp if a source-side equivalence to an existing reviewed concept appears
  4. Draft in _private/ → YAML SSOT → generate Markdown/TSV/TTL
  5. Maintainer review → core.ttl placement per ADR-0021 → glossary closure

No crosswalk YAML or Markdown for these four concepts is created by this document.

References