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 glossary collects every coined term in Orthemology. All terms listed here are candidates — none has been adopted. Authoritative definitions live in the manuscript’s glossary and the formal core; this page is a plain-language index to those definitions, with formal referents and status notes included where they matter for correct use. The verdict label and notation conventions referenced here are governed by docs/verdict-registry.yaml (Decision 0004) and docs/notation-registry.yaml (Decision 0005).
No terminology has been adopted. The pilot-0 v2 terminology instrument is READY_FOR_HUMAN_MATCHING_REVIEW (Decision 0018) but has not been run. Do not treat any term here as settled vocabulary.
Plain-language definition. The concrete situated case: the specific token occurrence under evaluation, with its own identity key and version. The token pole of the type/token distinction.Formal referent. An orthemma m is identified by a key κ and version v. It is the object to which a profile is attributed: O*(m; A) is the true profile of orthemma m under analysis A.Status note. The word “orthemma” is a candidate coinage. The conceptual object — the concrete, versioned, situated occurrence — is settled and used throughout the formal core, schemas, and validators regardless of whether the term is adopted.
Plain-language definition. A repeatable, consequence-bearing operational state-type that a case may instantiate. The type pole of the type/token distinction. An ortheme is never a local judgment token, a diagnosis, or a placed inferred profile; those are placements (p̂) or placement claims.Formal referent. Orthemes populate the analysis-active repertoire O_A ⊆ 𝓞, where 𝓞 is the universal ortheme universe. The instantiation relation Inst_A ⊆ M_A × O_A records which occurrences instantiate which orthemes under analysis A. Preferred status language: well-individuated; analysis-relevant; over/under-segmented. Avoid: “concrete ortheme” for a judgment token.Status note. Candidate coinage.
Plain-language definition. The set of orthemes that a given case truly instantiates, relative to a declared, versioned analysis. This is the ground truth against which placement claims and result verdicts are judged.Formal referent. O*(m; A) is the true profile of orthemma m under the declared, versioned analysis A. The analysis A specifies boundary, task, repertoires, loss, hard constraints, horizon, tolerance, representation family, merger family, and governance boundary where relevant. O*_T(m) is scoped shorthand, licensed only after a single analysis with task(A) = T is explicitly fixed (Decision 0001).Important caveat. The occurrence and its worldly facts are analysis-independent; only the state-type description is indexed. No task-to-analysis bridging law exists — two analyses can share a task while differing in tolerance or representation. Episode records carry the analysis identity and version; result correctness (V1 / RESULT_CORRECT) is judged against O*(m; A(e)).
Plain-language definition. A repeatable higher-order governing distinction, standard, or configuration, paired with a meta-policy. The type pole of the governing type/token distinction. A metaortheme specifies how the orthing process for a case should be configured and conducted.Formal referent. μ is a metaortheme type with structure ⟨g, S_μ, select_μ, prov, ver⟩, where select_μ is the state-selecting evidence procedure. It is paired with a meta-policy π_μ. Preferred status language: adequate/inadequate; fitting/misfitting; current/obsolete (as type/version). Avoid: “concrete metaortheme” for a case-bound token (that is a metaorthemma).Status note. Candidate coinage.
Plain-language definition. The concrete, case-bound configuration token of a metaortheme inside an orthing episode. It records the case-specific binding of the governing standard to this particular occurrence. It is not the whole episode and not the execution event.Formal referent. μ̄_{e,j} is the j-th metaorthemma in episode e. It is typed by MetaInst(μ̄_{e,j}, μ) and must satisfy Compatible(μ̄, A(e)) — the token binds case-specific values within the declared analysis without creating a hidden second analysis. A metaorthemma records: its identity and lineage; metaortheme type and version; episode analysis and version; target occurrence (κ, v); governed component; the case-specific binding map (reference frame, tolerance value, fixture, success surface, etc.); scope with dependent claims; instrument/calibration provenance; the binder with binding warrant kept distinct from the executor; binding time; and validity.Important caveat (zero-burden rule). Configurations with no material case-specific binding get no token; the associated verdict (GOV_TOKEN_ADEQUATE / V3c) is then not-applicable with a recorded reason. A metaorthemma isolates the failure mode: correct governing standard + correct policy + faithful execution + defective case-specific binding (Decision 0002).Status note. Candidate coinage. The word “metaorthemma” remains benchmark-gated; Decision 0002 adopts the conceptual object, not the terminology.
Plain-language definition. The rule-governed process of apprehending, profiling, routing, validating, and revising a case. The full governed workflow from encountering an occurrence to closing out the episode record.Formal referent. Orthing is the process whose complete, dated, auditable run constitutes an orthing episode e. All verdicts are defined over episodes, not over the process in the abstract.Status note. Candidate coinage.
Plain-language definition. One dated, concrete run of the orthing process — the auditable record on which all verdicts are defined. An episode may contain several metaorthemmata and covers a single bounded execution of the process.Formal referent. An episode e carries: a claim ledger 𝒬_e; a residual-disposition map δ_e; a per-claim evidence-status map estatus_e; a set of metaorthemmata {μ̄_{e,j}}; the analysis identity A(e); and pathway and result verdict records. Episodes may be composed into a DAG Γ_E = (E, ⇝) with a composed boundary-level episode e_Γ = comp(Γ_E) when composition conditions hold.Important caveat. Preferred status language for an episode: pathway-adequate/defective/undetermined; ex-ante justified/unjustified. Making pathway adequacy factive is prohibited.
Plain-language definition. The episode’s asserted profile: the proposition that a given case instantiates a particular ortheme or profile at the governed level.Formal referent. A placement claim is judged by the result verdict V1 (RESULT_CORRECT) against O*(m; A(e)). A placement is not an ortheme; it is the episode’s assertion about what ortheme(s) the orthemma instantiates under the declared analysis. Preferred status language: true/false; supported/unsupported; current/stale. Avoid: “pure ortheme.”
Plain-language definition. A result-free judgment over the governed pathway verdict core. Pathway adequacy is independent of whether the answer happened to be correct; a correctly answered case can be pathway-defective, and an incorrectly answered case can be pathway-adequate.Formal referent. PathwayAdequate(e) holds when every verdict in ReqPath(e) ∩ CorePath passes. The core is {EVIDENCE_SUPPORT, PROCEDURE_RELIABLE, EVIDENCE_CURRENT, GOV_CONFIG_ADEQUATE, GOV_POLICY_ADEQUATE, GOV_TOKEN_ADEQUATE, EXECUTION_FAITHFUL, EX_ANTE_JUSTIFIED, ROUTE_ADMISSIBLE, CLOSURE_TRUTHFUL, ROBUST_NEIGHBORHOOD}. Applicability is governance-derived through ReqPath(e), never discretionary. Decision 0003 established the result-free core; verdict labels were normalized by Decision 0004; symbols were normalized by Decision 0005.Important caveat. AdequatePathError — correct pathway, incorrect result — is the justified rare miss of a reliable process. It is explicitly not a moral, legal, or theological “blamelessness” predicate.
Plain-language definition. The space of all complete placements under an analysis (Π_A), and the space of all partial placements where some axes are left open (Π_A^∂). Candidate families range over Π_A; inferred profiles live in Π_A^∂.Formal referent. Π_A is the complete-profile space under analysis A; every element is axis-consistent under A. Π_A^∂ is the partial-profile space. The inferred profile p̂_{A,α,t}(m) ∈ Π_A^∂ is belief, never ground truth. The candidate set Ĉ_{A,α,t}(m) ranges over complete profiles in Π_A. Defined in manuscript Definition 10 and Decision 0007. Π_T is scoped shorthand.
Plain-language definition. A candidate property term: apt for orthing relative to an analysis and repertoire. An occurrence is orthable if it is amenable to determinate, correct/incorrect resolution under the declared (or a possible) analysis.Formal referent. Three senses are distinguished (Decision 0010): Orthability-L (local, analysis-relative aptness: Orthable(m; A)); Orthability-O (objective conditions of evaluability, not practice-relative); Orthability-R (created representation/application). Every occurrence of the term in normative prose must be readable under exactly one sense.Important caveat. “Orthable” is explicitly kept out of the operational confirmatory terminology core. It is exploratory in the philosophical companion lane only, under R2 disposition. It must not be used as operational vocabulary until benchmark results support adoption.

Build docs developers (and LLMs) love

Get started for freeTalk to us