Vocabulary cardinality
The 1,000 tuples are derived as 10 × 5 × 4 × 5 before description and metadata.
The SPINE EBNF-RV grammar for visible temporal, interpretive, didactic and content status
RigVedan represents a compact knowledge unit as four independent classification axes plus a description and optional metadata. The source provides typed enums, a dataclass, single- and multi-line parsing, text and context-stack formatting, dictionary round trips and a compatibility bridge into session-level hermeneutic status.
Drag and inspect the four semantic axes, description and metadata around the typed object.
The intended current model contains three past, three present, three future and one timeless marker.
Create a synthetic browser-local unit and inspect both compact text and object form.
The source parser extracts matching annotations and silently skips non-matching lines. This lab makes both counts visible.
Compare the lossless dictionary route with the compact text route.
to_dict/from_dict preserves metadata. format/parse reconstructs the four axes and description but returns default empty metadata.Render parsed units as a header followed by four-space-indented annotation lines.
The bridge converts only the hermeneutic axis between uppercase grammar tags and lowercase session-status values.
Inspect the parser, formatters, dictionary route, session bridge, DIALECTIC consumption and context rendering.
Design, initial Phase 5 implementation, the intended TIMELESS extension, tests, generated reference and current source pin.
Vocabulary cardinality and source-reported repository receipts are kept separate from efficacy claims.
The 1,000 tuples are derived as 10 × 5 × 4 × 5 before description and metadata.
Receipts describe larger commits with different scopes and were not rerun for publication.
Past, present, future and timeless.
FACT through UNKNOWN.
FOUNDATIONAL through ADVANCED.
DECISION through QUESTION.
Parser, formatters and bridge.
Dictionary serialization route.
Before free text and metadata.
Identified in inspected sources.
Representation behavior, integration evidence and evidence-based claims remain separate.
| Claim | Evidence class | Current status |
|---|---|---|
| Ten temporal markers and the other three vocabularies exist | Source implementation | Supported at pinned commit |
| Typed object, parsing, formatting and bridges exist | Source implementation | Supported at pinned commit |
| DIALECTIC consumes parsed knowledge units | Integration source | Supported in the inspected implementation |
| Dictionary round trip preserves metadata | Program behavior and committed tests | Supported; tests not rerun here |
| Text format round trip preserves metadata | Program behavior | Not supported; metadata is omitted |
| The grammar improves reasoning quality | Outcome efficacy | Not established |
| A FACT or TIMELESS tag establishes truth | Epistemic claim | Not established |
The grammar makes status distinctions explicit and machine-readable. Whether that improves downstream decisions must be measured separately.
Project memory and model output can mix historical decisions, active state, hypotheses, unresolved questions and learning dependencies in undifferentiated prose.
Independent temporal, hermeneutic, didactic and content axes should reduce accidental status collapse and make filtering, routing and review more precise. This remains a mechanism hypothesis.
Matched-task studies could measure status-preservation errors, incorrect promotion of hypotheses, superseded-decision reuse, unresolved-question recall, parser loss, reviewer agreement and downstream correction cost.
Producers can assign wrong labels, consumers can over-trust FACT, non-matching lines can disappear silently, metadata can be lost through text rendering and TIMELESS can be overused.
FACT is a classification supplied by the producer.TIMELESS does not enforce a decay or retention policy in this module.Claims are bounded to the named source artefacts and commits.
fbratten/spine @ 8fa62c1ebfc0c7b65680ef2d89b9953a519e4138spine/grammar/ebnf_rv.pyspine/grammar/README.md26f13c191769a658138bd41c1fbb8b5a2f1c225b8fb9a55fea67809795813b846dd22e580c9ad2dcscripts/tests/test_grammar.pyscripts/tests/test_phase5.pyspine/patterns/hermeneutic_loop.pyfbratten.github.io/methods/rigvedan/This page performs no model call, source retrieval, repository write, memory write, context execution or SPINE runtime operation. All examples are synthetic and browser-local.