Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/theislampill/orthemology/llms.txt

Use this file to discover all available pages before exploring further.

This page documents the current state of the Orthemology project with precision. The authoritative machine-readable statement is docs/current-state.yaml; this file, VERSION, README, the primary document headers, and OPEN-DECISIONS.md are all checked against it in CI. The current revision is R6 (2026-07-20), described as: experiment-readiness, sourcing-state, and reference-apparatus tie-off.
This project is not peer reviewed and not empirically validated. The R6 fresh-session review was a formal, schema, source-record, and build-integrity review of the repository — it is not external human peer review, not empirical validation, and not legal or publication closure. No designed study has been run. No term has been adopted. Draft PDFs are DRAFT-stamped artifacts, not publications.

Current review state

The R4 revision was produced as a mixed-provenance candidate under a mid-run model-provenance dispute, recorded and unadjudicated in docs/project-closure/r4/AUTONOMOUS-R4-STATE.json and docs/project-closure/r4/MODEL-SUBSTITUTION-INTERRUPTION-PR3.md. A fresh-session review then:
  • Reproduced the substitution boundary
  • Re-reviewed the pre- and post-substitution ranges hunk by hunk
  • Selectively recovered the quarantined work
  • Repaired findings from the review
  • Merged PR #3 through protected CI
Sign-off: docs/project-closure/r4-fresh-fable-review/FABLE-REVIEW-SIGNOFF.md. Merged-state verification: docs/project-closure/r6/FINAL-MERGED-VERIFICATION.md. That review’s scope is the repository’s formal, schema, source-record, and build integrity only.

What is and is not done

ItemStatus
Formal specificationInternally conformance-checked; open parameters acknowledged; no proof of consistency, completeness, or utility
Machine validators27 validator scripts; green CI
External peer reviewNOT done
Empirical validationNOT done — no study has been run
Terminology adoptionNOT done — no term adopted
Companion papersComplete drafts; conclusions conditional on labeled premises
Legal licenseOPEN DECISION — default copyright applies
Citation metadataOPEN DECISION — cite by commit (see Citing)

Correction carried by R4: SELF-1

The R2 and R3 closeout prose reported the pilot-0 v1 freeze hash as ece0412f…. The committed terminology/pilot0/FREEZE-HASH.txt has recorded 988a6522498df73ad1c7b0f73961a054ff20862d50fff6d644d0274877412772 since it was created, and freeze_pilot0.py --check has always passed against it. The packet was never wrong; three R3 documents and two closeout messages mis-stated the value. The historical documents’ bodies are preserved; the correction is recorded in the R4 correction ledger and is now machine-checked by validate_current_state.py.

Claim status by lane

Internally conformance-checked current specification over the declared definitions, schemas, positive and adversarial fixtures, and source-status contract; acknowledged open parameters; no proof of consistency, completeness, or utility.
Complete research draft, source-verified to the stated threshold; not peer reviewed; not empirically validated.
Complete conditional philosophical draft; no claim of universal proof.
Complete, source-verified-to-threshold, explicitly Atharī/Taymiyyan school-internal draft; comparative positions accurately sourced; not a neutral theological conclusion.
v2 instrument READY_FOR_HUMAN_MATCHING_REVIEW (Decision 0018), not run; no term adopted.
Not run / open. No designed study has been run. Nothing here reports an experimental result. Deterministic validators check consistency only.
Reproducibly built and visually inspected draft artifacts. Byte-reproducible, DRAFT-stamped, commit-pinned. No numbered release, DOI, or archival claim exists.

Open decisions (owner-only)

These six burdens require owner resources or legal acts and are not routine research tasks.
Real-world author name, affiliation, ORCID, and preferred bibliographic identity. Only then can CITATION.cff or a numbered release exist honestly. Interim guidance: docs/CITING.md (cite by commit).
Executing any designed study requires owner resources and responsibility: model spend, ≥3 human raters, blind human matching review of the pilot-0 v2 packet, human-subjects requirements, and any external preregistration/registry submission. Packet preparation is not owner work and is done: FCSP-1 and ER-1 are READY_TO_RUN; the terminology v2 instrument is READY_FOR_HUMAN_MATCHING_REVIEW. No run has occurred; nothing is externally preregistered.
Preprint server, journal, DOI archive, or any submission under the owner’s identity.
The internal casebook, longitudinal record, screenshots, and transcripts remain unpublished and are not rights-cleared. No public claim depends on them. Publishing any sanitized form is an owner call.
None is currently identified — the R3 and R4 sourcing passes found accessible scans, databases, or open-access editions for every needed work. This entry stays open as the standing category.

Research residuals with triggers

These are ordinary research tasks — not owner-blocked — that have defined triggers for completion.
IDTaskTrigger
RR-1Edition-level pagination for classical lociLibrary/edition access before any external submission
RR-2Direct full-text verification of the Evans 1998 concrete/ideal-reason page locusFull-text access; currently cited as reported by Doko & Turner 2023
RR-3Primary-locus verification queue for Taymiyyan Dar’ attributionsContinued research; all such attributions labeled secondary reconstruction meanwhile

Open formal parameters

The formal core has acknowledged open parameters that do not block current use but are not yet resolved:
  • RequiredBy beyond the shipped machine-readable governance instance
  • Evidence-class exhaustiveness (working hypothesis)
  • Fusion-mapping non-uniqueness
  • Δ_A idealizations
  • Many-to-many MetaInst typing (explicitly unimplemented in R4)
  • The ProfileOf_A bridge between latent-state models and orthemic profiles is partial and analysis-declared

Sourcing state

The machine registry (references/source-status.yaml) is authoritative for its declared claim families (CIR-*, ELT-*, ATH-*, LAT-*, EXT-*); paper prose must agree with it bidirectionally. The Qurʾānic registry is references/quran-loci.yaml. The current consolidated sourcing view is docs/sourcing/CURRENT-SOURCING-LEDGER.md; the R2-era ledgers are historical baselines with banners. A green offline validator establishes record shape and internal agreement — never source truth.
Historical candidate reports from R1–R5 remain intact and are classified in docs/project-closure/HISTORICAL-STATUS-INDEX.yaml. They are never rewritten; supersession is recorded by dated notice.

Build docs developers (and LLMs) love