Five dimensions of meaning.

Significance is the foundational dimension, the one every other memory system gets approximately right and then stops. Meaning Memory treats it as a first-class composite, not a scalar.

Every memory carries two significance scores: sig_self (how important the agent thinks it is) and sig_external (how important the operator's policy thinks it is). When they diverge, the system records the divergence and the rationale. This is the encoding-compliance moat, a dual-write architecture where an extractor-LLM catches the moments the agent forgot to remember, with STARE Sig arbitrating which write survives merge.

Probabilistic LLM agents silently fail to call mm_remember in the moments that matter. No retrieval, decay, or calibration system can recover what was never written. Significance, measured twice, arbitrated deterministically, is what closes the encoding hole.

Temporal is more than a timestamp. Every memory in Meaning Memory carries three time signals: when it was observed, when it becomes valid, and when it stops mattering.

A decay curve, operator-configurable per playbook, defines how significance fades. Some memories decay fast (today's traffic numbers). Some never decay (a customer's allergy). Meaning Memory tracks both decay and validity windows: a fact can be temporally valid for a finite period (valid_from → valid_until) even if its significance never decays.

KV caches solve freshness with TTL. Vector stores solve recency with score boosts. Neither captures the truth that some memories fade and some expire and some stay forever, and which is which is a governance question, not a heuristic.

Asymmetry, also called Attribution, is the dimension most memory systems fake with post-hoc filtering. In Meaning Memory, every memory carries explicit attribution and trust gradients native to the data model.

In multi-agent deployments, this matters enormously. The same statement ("our churn rate is 4.2%") from the CFO and from a customer comment should not carry the same retrieval weight. Asymmetry encodes the trust gradient: asymmetry_modifiers is a JSONB field per entry, and _asymmetry_scale() multiplies into retrieval ranking.

Enterprise multi-agent fleets are political ecosystems. Marketing's agent sees one version of "true." Engineering's sees another. Legal's sees a third. Without attribution as a first-class primitive, cross-agent memory becomes either dangerously homogenized or frustratingly siloed. Asymmetry lets you have shared facts with trust-aware retrieval.

Relational turns memories into a typed graph. Every memory can carry typed edges to other memories, to agents, to documents, to customers, to scope groups. Nine bounded edge types ship today (:supersedes, :contradicts, :references, :implies, …) plus custom:* for domain-specific extensions.

Multi-hop traversal lets agents reason across the graph: "what other memories supersede this one?" "what does this fact contradict?" "show me the path from this customer ticket to the original bug report." Recursive CTE queries on Postgres return path-explain answers in single-digit milliseconds at fleet scale.

Vector similarity surfaces "memories that look like this." Relational edges surface "memories that matter to this." The first is retrieval. The second is reasoning. Production agent fleets need both.

Episodic clusters memories into narrative arcs. A customer conversation isn't a list of facts, it's an episode with a beginning, a turning point, and a resolution. Most memory systems flatten that structure. Meaning Memory preserves it.

Every memory can carry an episode_id. The mm_episodes summary table maintains aggregate metadata, duration, dominant participants, narrative arc shape, dim signature. Agents can query "summarize this episode" without loading every underlying memory.

The unit of recall in human memory is not the fact. It's the episode. When you remember a conversation, you don't reconstruct it token-by-token, you replay the arc. Agent memory built on the same principle composes better, retrieves better, and audits better than fact-shaped retrieval.