Technology research programme 2026-04-25

Established 2026-04-25 after Catala library+web research surfaced 3 load-bearing discoveries that flipped the Topic 1 scope-approval decision. Rich directive: apply the same research discipline to ALL technologies used in INHERIT v2 + InheritKit.

Origin

Catala Topic 1 scope-approval started with γ winning 85.3% (0.8pp over δ). Rich asked “please look in the library and search online… take the perspective of the worlds leading Catala expert, and heavily incentivise him/her to find an extremely elegant solution.”

Research surfaced:

catala-explain production module (structural validation for δ)
Catala v1.0.0 / v1.1.0 stable (risk mitigation)
catleg French legislative tooling (Akoma Ntoso integration prior-art)

Decision flipped to δ with 4.1pp margin. 14 downstream improvements identified.

Rich inference: similar discoveries likely exist for SHACL, MCP, VC 2.0, JSON Schema, and other session technologies.

Research programme scope

12 technologies across 3 tiers:

Tier 1 (architecture-critical; HIGH expected yield)

#	Technology	Execution plan	File
T1	MCP (Model Context Protocol)	Reserved for in-session deep-dive (HIGH velocity; Nov 2024 launch; AI-vendor integration critical)	T1-MCP-notes.md
T2	W3C VC 2.0 + DID + KERI	Reserved for in-session deep-dive (VC 2.0 recently ratified; Phase-C Step 3 critical)	T2-VC-DID-KERI-notes.md
T3	JSON Schema + Sourcemeta ecosystem	Reserved for in-session deep-dive (INHERIT primary format; commercial partner relationship)	T3-JSON-SCHEMA-notes.md
T4	SHACL (Core + SHACL-AF)	Reserved for in-session deep-dive (Rich’s flagged example; A-52/A-56/A-63 CI critical)	T4-SHACL-notes.md
T5	SKOS + XKOS + OMG Commons Classifiers	Subagent dispatch 2026-04-25	T5-SKOS-XKOS-CMNS-notes.md
T6	PROV-O	Subagent dispatch 2026-04-25	T6-PROV-O-notes.md
T7	Akoma Ntoso + ELI + ECLI	Subagent dispatch 2026-04-25 (Phase-C Step 2 target)	T7-AKOMA-NTOSO-ELI-ECLI-notes.md
T8	ROBOT (OBO Foundry ontology build)	Subagent dispatch 2026-04-25	T8-ROBOT-notes.md

Tier 2 (infrastructure; MEDIUM-HIGH expected yield)

#	Technology	Execution plan	File
T9	OpenAPI 3.1 + Arazzo	Subagent dispatch 2026-04-25	T9-OPENAPI-ARAZZO-notes.md
T10	SSSOM (ontology mapping)	Subagent dispatch 2026-04-25	T10-SSSOM-notes.md
T11	F* + Mopsa (Catala verification stack)	Subagent dispatch 2026-04-25	T11-FSTAR-MOPSA-notes.md
T12	External-ontology alignment targets (CIDOC CRM + FIBO + LADM + BFO)	Subagent dispatch 2026-04-25	T12-EXTERNAL-ONTOLOGIES-notes.md

Tier 3 (supporting; LOW-MEDIUM yield; deferred)

Not included in this programme. Quick release-notes batch-check deferred to future session if needed: OCaml / TypeScript / Python / Cloudflare AI Gateway / Anthropic API / Git/GitHub CI/CD / W3C RDF 1.1/1.2 / OWL 2 (EL/QL/RL/DL) / LegalRuleML + LKIF.

Execution model

Option D2 (hybrid tiered) — chosen 2026-04-25:

8 parallel subagents dispatched for Tier-1 MEDIUM + Tier-2 technologies (T5-T12)
4 Tier-1 HIGH-velocity technologies reserved for in-session deep-dives (T1-T4)
Wait for subagent results; then plan follow-up actions

Research prompt template per subagent:

What are the current stable releases (2024-2026)?
Are there production ecosystem modules analogous to catala-explain?
What integration patterns are emerging with adjacent standards?
Any new features that simplify or transform INHERIT v2’s current approach?
Are there pre-existing INHERIT-adjacent implementations?
What deprecations or superseding standards should we know about?
What’s the architectural impact (HIGH / MEDIUM / LOW)?

Expected outputs per T-file

After subagent completion, each T-file has:

findings section with 3-5 library-ULID-quoted discoveries (if library has relevant items)
web_findings section with URL-cited discoveries from 2024-2026
ecosystem_modules section listing production-modules discovered
integration_patterns section describing emerging patterns
architectural_impact rating (HIGH / MEDIUM / LOW)
recommendations section with specific INHERIT v2 / InheritKit actions

Followup planning

After subagent results return:

Review each T-file for HIGH-impact findings
Identify Catala-equivalent discoveries (production modules / stable releases / integration patterns that shift architecture)
Update relevant Phase-C spec briefs (T5/T6 inform Phase-C Step 1 Catala + Step 3 SSI; T7 informs Phase-C Step 2 Akoma Ntoso; T9 informs SDK spec; T10 informs A-33 SSSOM; T11 informs Phase-1 CI chain; T12 informs A-22 alignment matrix)
Flag items for in-session deep-dive if HIGH-impact surface
Schedule in-session deep-dives for T1-T4 HIGH-velocity technologies when Rich is ready

Why this programme matters

If ANY of the 12 technologies has experienced Catala-level ecosystem evolution:

Architectural decisions currently baselined on older evidence need updating
Phase-C specs (Step 2-4) benefit from research-grounded baseline before authoring
Phase-1 execution can leverage stable-release infrastructure rather than research tooling
Commercial positioning (InheritKit) can reference production-ready ecosystem

If NO technology has material evolution:

Current session architecture is research-validated across the stack
Phase-C specs proceed without need for subsequent-revision
Research investment still yields confidence + audit trail

Estimated yield

Based on Catala precedent (massive yield) + research pattern analysis:

Likely high yield (Catala-equivalent): MCP (HIGH velocity; AI-vendor central), W3C VC 2.0 (recently ratified), Sourcemeta JSON Schema
Likely medium yield: SHACL, Akoma Ntoso, OpenAPI+Arazzo, ROBOT
Likely low-medium yield: SKOS/XKOS, PROV-O, SSSOM, F*/Mopsa, CIDOC/FIBO/LADM/BFO

project_catala_review_plan_2026_04_25 — inspired this programme
project_catala_ecosystem_maturity_2026_04_25 — precedent for what discoveries look like
project_catala_explain_as_commercial_artifact — commercial-impact precedent
feedback_library_web_research_narrow_margins — the pattern being generalised
project_z_research_programme_24_files — parallel 24-Z research programme (library grounding)
feedback_always_check_library_indexed_first — broader discipline

Sources to check (research-request file templates)

Each T-file starts with:

research_questions (6-7 from template)
library_evidence_registered (check ~/off-github/library/indexed/ for relevant items)
web_search_queries (3-5 specific 2024-2026 queries)
github_repos_to_fetch (official + ecosystem repos)
w3c_oasis_omg_pages (standards-body relevant pages)

Timeline

2026-04-25: Programme established; 8 subagents dispatched; memory + T-files created
2026-04-25 (later): Subagent results return; review findings
2026-04-26+: Follow-up actions per T-file findings; in-session deep-dives for T1-T4 if Rich ready
Ongoing: Quarterly refresh (target 2026-07-25) to catch new developments

FINDINGS SUMMARY (2026-04-25 — all 8 subagents complete)

Impact distribution: 6 MEDIUM-HIGH (T6 PROV-O, T7 AKN/ELI/ECLI, T8 ROBOT, T9 OpenAPI/Arazzo, T10 SSSOM, T11 F*/Mopsa) + 1 MEDIUM-LOW (T5 SKOS/XKOS/cmns-cls) + 1 MIXED per-family (T12 External ontologies: LOW CIDOC / MODERATE-HIGH FIBO / MODERATE LADM / HIGH-positive BFO).

Yield:

10 top-tier discoveries (ranked in project_technology_research_synthesis_2026_04_25)
16 new tensions surfaced for Phase-5 queue
5 cross-technology convergent threads identified
3 new citations to add (Nature Sci Data 2025 PROV-O↔BFO; arXiv 2507.16044 AutoMCP; Sartor et al. ACM 2023 ELI↔AKN)
4 ISO-track positioning boosts (BFO DoD adoption + PROV-O↔BFO alignment + FIBO cmns-* consolidation + LADM 5-of-6-parts-published)

Retroactive impact: ZERO revisions to Phase-B Wills v1.3 / Trusts v1.1 / Probate v1.1 locks. Catala scope-δ strengthened. A-22 iv-3 tier assignments all hold.

Biggest strategic shift: ISO-track positioning stronger than previously baselined.

Full synthesis: project_technology_research_synthesis_2026_04_25.md.

Follow-on: T1-T3 in-session deep-dives dispatched 2026-04-25

Per Rich directive: “I think we should do the MCP (Model Context Protocol), W3C VC 2.0 + DID + KERI, JSON Schema (Sourcemeta ecosystem) research now - this could provide even richer pickings.”

T1 MCP — dispatched as subagent (reserved-as-in-session reframed given subagent success pattern)
T2 VC 2.0 + DID + KERI — dispatched as subagent
T3 JSON Schema + Sourcemeta — dispatched as subagent
T4 SHACL — NOT in Rich’s 2026-04-25 directive; remains reserved

TT Claude Memory

Explorer

project_technology_research_programme_2026_04_25

Technology research programme 2026-04-25

Origin

Research programme scope

Tier 1 (architecture-critical; HIGH expected yield)

Tier 2 (infrastructure; MEDIUM-HIGH expected yield)

Tier 3 (supporting; LOW-MEDIUM yield; deferred)

Execution model

Expected outputs per T-file

Followup planning

Why this programme matters

Estimated yield

Sources to check (research-request file templates)

Timeline

FINDINGS SUMMARY (2026-04-25 — all 8 subagents complete)

Follow-on: T1-T3 in-session deep-dives dispatched 2026-04-25

Graph View

Backlinks

TT Claude Memory

Explorer

project_technology_research_programme_2026_04_25

Technology research programme 2026-04-25

Origin

Research programme scope

Tier 1 (architecture-critical; HIGH expected yield)

Tier 2 (infrastructure; MEDIUM-HIGH expected yield)

Tier 3 (supporting; LOW-MEDIUM yield; deferred)

Execution model

Expected outputs per T-file

Followup planning

Why this programme matters

Estimated yield

Related memories

Sources to check (research-request file templates)

Timeline

FINDINGS SUMMARY (2026-04-25 — all 8 subagents complete)

Follow-on: T1-T3 in-session deep-dives dispatched 2026-04-25

Graph View

Backlinks