Skip to content

docs(ai-memory): MemPalace → MIF Level-1 mapping + examples#206

Draft
perlowja wants to merge 1 commit into
modeled-information-format:mainfrom
perlowja:feat/mempalace-level1
Draft

docs(ai-memory): MemPalace → MIF Level-1 mapping + examples#206
perlowja wants to merge 1 commit into
modeled-information-format:mainfrom
perlowja:feat/mempalace-level1

Conversation

@perlowja

@perlowja perlowja commented Jul 3, 2026

Copy link
Copy Markdown

Reopened against main per @zircote's request (2026-07-02) now that develop/v1.0.0 is promoted and stable. Supersedes #65, auto-closed when the develop/v1.0.0 base was deleted. Rebased cleanly onto main (single commit, 3 files under profiles/ai-memory/).

Per the discussion in #60 — the Level-1 first step of the MemPalace → MIF AI-Memory mapping (OKF-valid + cognitive-triad typing).

What's here

  • profiles/ai-memory/MEMPALACE-MAPPING.md — the draft field mapping (open for your feedback).
  • profiles/ai-memory/examples/mempalace-observation.md — a MemPalace entitysemantic observation, with a KG triple rendered as a typed MIF relationship.
  • profiles/ai-memory/examples/mempalace-session.md — a MemPalace drawerepisodic session.

Mapping at a glance

MemPalace MIF type Rationale
entity (person/project/topic) semantic (observation) decontextualized, stably-true world knowledge
drawer (filed_at/source/added_by) episodic (session) time-anchored record of a lived event
procedural (skill) declared gap — no native MemPalace construct
  • Triples: entity-object → typed relationships[] (predicate→type, object→target path, confidence→strength); literal-object → the first-class scalar properties field (now in v1.0.0).
  • IDs: UUIDv5 derived from the MemPalace id under a fixed MemPalace namespace UUID (deterministic across re-exports).
  • Deferred: bitemporal valid_from/valid_toLevel-3 version-DAG sidecar; trust tiers → L2; compression manifest → L3.

Validation

python scripts/okf_validate.py profiles/ai-memory/examples   # conformance + relationship sync + round-trip
python scripts/mif_convert.py roundtrip profiles/ai-memory/examples   # lossless markdown↔json-ld

Draft — the mapping is the discussion basis from #60. Happy to adjust triad assignment, namespacing, or literal-object handling before layering up to L2/L3. Related #60.

Add a draft MemPalace -> MIF AI-Memory profile mapping at Level 1 (OKF-valid +
cognitive-triad typing), with two worked example concepts:

- mempalace-observation.md: a MemPalace entity -> semantic observation, with a
  knowledge-graph triple rendered as a typed MIF relationship.
- mempalace-session.md: a MemPalace drawer -> episodic session (filed_at ->
  created; wing/room -> namespace path).

MEMPALACE-MAPPING.md documents the structure/field mapping, the triple ->
relationship rules (entity-object vs literal-object), the UUIDv5 id derivation,
and what is deferred to Levels 2/3 (bitemporal validity -> version-DAG sidecar,
trust tiers, compression manifest).

Both examples pass scripts/okf_validate.py and the scripts/mif_convert.py
roundtrip lossless check.

Related modeled-information-format#60.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant