symbolic mind

Objects, not tokens.

Today's networks learn distributions over tokens; what they know is trapped in weights. symbolic mind is built on a different substrate: typed objects and the relations between them. 70 GB of source code has been translated into object data layers at multiple levels of abstraction — so what the network learns can be read, validated and diffed, not just sampled.

Finite alphabets at every level

At each abstraction level, the mechanics and algorithms discovered in the corpus are expressed as formalized, closed lists — together with the enumerated variants of admissible links between them. Objects at the bottom; mechanics above them; algorithms above those; systems at the top. Abstraction rises upward, and every level has its own finite alphabet.

Knowledge is the force of gravity.

In this architecture accumulated knowledge acts as an attractive force in the formation of object graphs: the denser the knowledge around an object, the stronger it pulls new structure toward it. Graphs are not assembled by rule alone — they condense around mass, the way matter condenses in a gravitational field. What the network already understands shapes what it can learn next.

Meaning graphs of working discussions.

The first instrument built on this discipline turns the transcript of a working discussion into a graph of meaning: who acts, what connects to what, which rules apply — and, critically, where participants fill the same word with different meaning. "Order confirmed" means a signed specification to the commercial director, a checked box in the CRM to the manager, and a payment fact to the database. All three are right; the argument lasts weeks because the word is one and the meanings are three.

• A closed ontology: 12 node types and 15 edge types — subject-level (Actor, Object, Action, Event, State, Process, Rule, Goal) and meta-level (Problem, Decision, Assumption, Question). The validator rejects anything outside the lists.
• Divergences are flagged, not smoothed over: each interpretation becomes a node with its author's quote, linked by conflicts_with.
• Abstraction is expressed by relations, not stored levels — hierarchy derived from refines chains is reproducible across runs.
• Human edits are protected: a confirmed or human-edited node is never overwritten by the next run — divergence goes into a proposal queue, and the human decides.
• Quality is measured against golden expectations: required nodes, required edges, required divergences, forbidden generalizations, integrity violations.