Documents compressed subform references, canonical identity, bounded recursion, complexity accounting, lineage, and the interpretability trade-offs of structural reuse.
Research documentation: this page interprets a cited research source and defines evidence requirements. It does not claim a released Teleodynamic AI implementation.
Purpose
The source defines ReEntry as a bounded reference into a registry of recurring subforms so repeated structure can be shared without duplicating its complete representation. This may reduce storage and expose repeated motifs, but it introduces identity, recursion, and accounting questions.
Canonical identity
Assign immutable content or node identifiers, record the referenced registry revision, and prevent a later mutation from silently changing the meaning of historical decisions. Distinguish a shared immutable subform from a copied snapshot.
Safety and termination
- Reject cycles or define an explicit cycle semantics.
- Set maximum expansion depth, visited-node, and evaluation-step limits.
- Make evaluation order deterministic or record it.
- Define reachability, retention, retirement, and garbage-collection policy.
- Verify that registry corruption cannot create unbounded recursion.
Complexity and interpretability
Publish physical storage cost, unique-node complexity, expanded semantic complexity, and explanation depth. Compression can make a graph smaller on disk while making a human derivation deeper or harder to inspect.
Replay requirement
A historical event must resolve against the exact registry revision used at decision time. Replay against a mutable current registry is not an evidence-preserving reconstruction.
Scope
This starter page defines the questions, boundaries, evidence, and failure modes that should be recorded before a capability is presented as supported.
Engineering considerations
- Identify the source, version, target environment, and owner.
- Separate observed values from estimates and externally reported values.
- Record trade-offs, unsupported cases, and fallback behavior.
- Link performance statements to a compatible benchmark methodology.
Verification questions
- What exact artifact, revision, backend, and environment were reviewed?
- Which assumptions could change the result?
- Which data should be retained so another engineer can reproduce the conclusion?