G.11 - Decide Whether and How to Refresh SoTA Packs and Related Results (Telemetry and Decay)

Tag. Architectural pattern (architectural; notation-independent)

Status: Stable Normativity. Normative (unless explicitly marked informative)

Stage. run-time and maintenance-time (selective re-computation, republication, and controlled deprecation)

Primary outputs, when their use calls for them. RefreshQueue, RefreshPlan@Context (a local application name for one exact U.WorkPlan, not a new kind), RefreshReport@Context (record of refresh Work or its audit), DeprecationNotice@Context, and EditionBumpLog@Context. Continued reliance on an already applicable result requires none of these solely to prove that refresh was omitted.

Primary hooks. G.Core (RSCR trigger catalogue, alias docking, and Default Governing Definition Index), G.6 (EvidenceGraph; PathId and PathSliceId), G.7 (Bridge Sentinels; CL, Φ, and plane policy pins), G.5 (set-returning selection and dispatch), G.8 (SoS-LOGBundle telemetry hooks), G.9 (parity reruns), G.10 (shipping hooks and pack-level telemetry pins), G.12 (dashboard telemetry pins), B.3.4 (freshness and decay), E.18 (GateCrossing and CrossingBundle visibility), C.18 and C.19 archive, front, and live-pool policy pins, C.23 (SoS-LOG branches and maturity ladders), C.28 (causal-use support results whose SoTA-sensitive fields can change downstream causal-use results).

Non-duplication note. G.11 cites G.Core for RSCR trigger-kind meaning, CN and CG admissibility, tri-state guards, penalties, set-return semantics, shipping or harvesting delegation, RSCRTriggerKindId values, and default governing definitions. Refresh plans and reports cite those governed definitions; they do not create local trigger meanings or default definitions inside the refresh record.

G.11:0 - Use this when

Use this pattern when a shipped pack, evidence set, dashboard, selected set, archive, front, Q-front, term bridge, descriptor set, parity result, or a use that relies on an A.6.RCD predicate definition or derived relation kind may be stale because telemetry, freshness, edition pins, policy pins, evidence, bridge calibration, source currentness, a relied-on base relation definition, the named substrate edition, or derivation applicability changed.

G.11:0.1 - What goes wrong if missed

The team either rebuilds everything after every small change or keeps using a shipped record whose source, descriptor, edition, policy, bridge, or archive currentness has silently drifted. Refresh then becomes an informal maintenance habit rather than a scoped, reviewable work plan and report.

G.11:0.2 - What this buys

The practitioner first decides what the available support permits for the receiving use. When a changed premise requires upkeep, the refresh kit names the affected object and scope, the source and policy basis, and the justified action. Planning and reporting remain separate, and refresh can stay local while preserving comparability and subject-specific result meanings.

When a later or replacement source may change a claim and the actual receiving uses must first be found and revalidated, use A.10.1 for the bounded search frame, discovery coverage and gaps, exact-use test, action-changing reach, application of the direct subject guidance, and the independently obtained subject result. G.11 continues to govern source currentness, decay, refresh planning, and refresh reporting. A practitioner or admitted System may use a separately established currentness result or independently obtained subject result to plan later refresh without changing either result or the governing patterns.

G.11:0.3 - First output

For loop, harness, workflow-store, or DPF seed artifacts, a refresh line names the currentness object directly: source pack, evaluator, benchmark, harness edition, workflow edition, pattern seed, PFAD and PFR dependency, selected set, archive, front, or publication carrier. G.11 records currentness, source decay, edition change, telemetry, scoped refresh action, and report refs; it does not decide whether the artifact improved.

First establish whether the current conditions and available basis already support the receiving use. If they do, retain that result without a new currentness line, WorkPlan, waiver or skip-refresh certificate. When a later recipient needs a changed limit or retained reason, keep the minimum useful content with the existing result or publication; a RefreshCurrentnessLine@Context can express it when a structured line is useful. Produce a RefreshPlan@Context only for selected planned refresh. For an underlying claim about a selected set, archive, culture, bridge, evidence, dashboard or shipping, obtain that subject pattern’s result; currentness does not establish its adequacy.

When currentness is the live question, use G.11 to record framework edition pins, source packs, publication-carrier currentness, deprecation, supersession, and source-decay conditions. In that record, cite E.4 for the affected framework, E.4.PFR for a framework relation, E.4.PFAD for the framework architecture decision, G.2 for source use, and E.11 for discovery. For publication, cite E.17 for a source-backed face and return to source and E.24.PUB for the occurrence, form, carrier, audience, bounded use, and availability. Do not create private refresh vocabulary for these neighboring meanings.

G.11:1 - Problem frame — Keeping shipped SoTA current without global rebuilds

Part G produces shipped, selector-ready publication units and records: packs, bundles, evidence graphs, parity reports, and dashboards. Once shipped, they are exposed to:

• telemetry (illumination and archive changes, parity outcomes, dashboard deltas),
• currentness conditions (a relied-on premise changes, a justified review becomes due, or an actual qualification or use window ends),
• edition drift (descriptor, distance, or transfer rules bump; policy pins evolve),
• bridge evolution (CL or plane penalties or calibrations update).

The kit addresses two recurring refresh failures:

• a brittle set of ad-hoc “full rerun” rituals, or
• an audit-only refresh result that leaves currentness drift unresolved.

G.11 is the Part G governing definition of the refresh orchestration kit: its users turn typed refresh causes into scoped plans and record execution in auditable execution reports. Cause semantics and universal invariants remain delegated to G.Core.

G.11:2 - Problem — Why naive refresh breaks comparability and admissibility

A refresh loop fails (conceptually) when any of the following happens:

1. Full-rerun mania. Minor edits (e.g., a single Bridge calibration) trigger pack-wide rebuilds without a traceable scope rationale.
2. Editionless telemetry. Telemetry signals are recorded without edition pins, making reruns non-comparable and parity-unreplayable.
3. Alias-as-semantics. Local trigger aliases are treated as if they define meaning, fragmenting refresh semantics across patterns.
4. Silent crossings. Refresh actions implicitly change crossing assumptions (UTS, Path, or policy pins) without a visible CrossingBundle.
5. Orchestration smuggles semantics. Refresh introduces new default behaviors (dominance, PortfolioMode, or Γ-fold) or coerces partial orders into scalars “for convenience.”

G.11:3 - Forces — Minimal recomputation under strict invariants

• Minimal scope vs. completeness. Refresh must be as local as possible (slice-scoped), but still include a defensible dependency closure over evidence and crossings.
• Operational urgency vs. auditability. Actual refresh Work must remain inspectable through the needed pins, references and paths. A currentness decision or an unchanged use must not become a fictitious Work occurrence.
• Alias stability vs. semantic unification. Existing trigger labels must remain usable, but their meaning must be one governing definition and id-based.
• Modularity vs. orchestration power. G.11 must coordinate harvesting, parity, and shipping without re-implementing them or importing discipline-specific method semantics into core.
• Policy-bound behavior vs. “smart defaults.” Ordering of refresh, priority heuristics, and budget handling are valuable—but must live as policy-bound extensions, not as hidden universal rules.

G.11:4 - Solution — RSCR-driven refresh as a P2W-scoped orchestration kit

G.11:4.1 - G.Core linkage (normative)

GCoreLinkageManifest (normative; canonical shape per G.Core; Nil‑elision permitted).

GCoreLinkageManifest := ⟨
  CoreConformanceProfileIds := {
    GCoreConformanceProfileId.PartG.AuthoringBase,
    GCoreConformanceProfileId.PartG.TriStateGuard,
    GCoreConformanceProfileId.PartG.UTSWhenPublicIdsMinted,
    GCoreConformanceProfileId.PartG.ShippingBoundary
  },

  RSCRTriggerSetIds := {GCoreTriggerSetId.RefreshOrchestration},

  CorePinSetIds := {
    GCorePinSetId.PartG.AuthoringMinimal,
    GCorePinSetId.PartG.CrossingVisibilityPins
  },

  CorePinsRequired := {
    RSCRTriggerKindId,
    RSCRTriggerAliasId?,
    scope: PathSliceId[] | PatternScopeId,
    payloadPins{…},

    RefreshPlanId?,
    RefreshReportId?,
    DeprecationNoticeId?,
    EditionBumpLogId?,

    WorkPlanRef[]?
  },

  DefaultsConsumed := ∅,
  TriggerAliasMapRef := G.Core.TriggerAliasMap.G11
⟩

By the G.Core Expansion rule, the effective conformance ids, trigger kinds, and pin obligations for G.11 are the manifest expansions (profiles, sets, and pin sets) plus the explicit deltas above.

TriggerAliasIds (visible; labels only). {G.11:T0…T7} (docked via TriggerAliasMapRef; aliases are never semantic authorities).

G.11:4.2 - Refresh orchestration kit (subject-qualified; conceptual artefacts)

G.11 defines a kit of authoring-plane artefacts for actual refresh planning and reporting. RefreshPlanId is required on a plan and RefreshReportId on a report. The linkage manifest applies to the triggers and actions that occur; retaining applicable support creates no trigger, plan or report merely to fill their pins.

1. RefreshQueue (conceptual queue). A queue of refresh candidates keyed by scope (PathSliceId preferred; PatternScopeId permitted). Ordering, prioritization, and batching are policy-bound (and therefore extension-scoped), but every queue item carries canonical trigger kind ids.

2. RefreshPlan@Context (one exact U.WorkPlan). A planned refresh is one U.WorkPlan episteme under A.15.2. It does not execute Work and does not embed gate decisions. RefreshPlan@Context is only this pattern’s application name for the plan; it declares:

   • RefreshPlanId (UTS-published id; editioned)
   • EntityOfConcernRef and ReferencePlane pins (by ref; no implicit widening)
   • TargetScope := PathSliceId[] | PatternScopeId[]
   • PlannedTriggers := RSCRTrigger[] (canonical trigger kind ids, scope, and payload pins)
   • PlannedActions := RefreshAction[] (each action delegates to a subject pattern)
   • RequiredPins: exact affected source and result editions, policy pins and scope for replayability; UTS pins for public identities actually used; graph Path pins when the selected scope is graph-expressed or an independently applicable receiving contract requires them.
   • PlannedFillingRows[]? as ClaimGraph content kept inside the WorkPlan under A.15.3 when a value must be pinned against a declaration member defined by its own pattern. A row is addressed only through the WorkPlan and has no separate reference or identity.

3. RefreshReport@Context (record of refresh Work or its audit). An execution or audit report that records:

   • RefreshReportId (UTS-published id; editioned)
   • ExecutedActions[] with links to cited artefacts governed by cited patterns (e.g., new parity report id, new pack id)
   • ObservedDeltas (telemetry deltas, admissibility changes, evidence-relation or source-relation changes) as refs and pins, not as untyped prose
   • RSCRRefs[] (any RSCR or regression harness artefacts invoked)
   • EmittedNotices[] := DeprecationNoticeId[] and EditionBumpLogId[]
   • the canonical trigger kinds actually applied (not only aliases)

4. DeprecationNotice@Context and EditionBumpLog@Context. Controlled evolution artefacts that preserve ID-continuity:

   • DeprecationNotice explains scope, reason class (canonical trigger kind ids), and successor refs.
   • EditionBumpLog records edition increments and the pins that justify them.

   Note (normative by delegation). ID continuity and alias discipline are governed by G.Core (do not restate as local rules here).

G.11:4.2a - Selected-set, archive, and cultural-variant currentness

Use this line when refresh currentness concerns a selected set, front, Q-front, archive, portfolio lineage, cultural-variant lineage, style or tradition term bridge, path slice, reused A.6.RCD predicate definition, or admitted derived relation kind.

RefreshCurrentnessLine@Context:
  governedObjectRef:
  currentnessObjectKind:
  sourceRecordRef:
  editionOrLineagePins:
  affectedPathSliceOrScope:
  subjectPatternLocator:
  receivingUseAndApplicableConditions:
  currentnessConclusion:
  plannedRefreshAction?:
  refreshReportRef?:

currentnessObjectKind may name, for example, a selected set, Front, Q-front, ExplorationArchive, Archive, a cultural lineage, a term bridge, or a reused predicate definition. Use the line only when its recipient needs this structured currentness result; identify the temporal reference and relevant window within its conditions and pins. plannedRefreshAction? is absent when no refresh is selected, and refreshReportRef? is absent when no such report exists. The line states applicability for a use, not that the subject claim is true or adequately supported. Use G.5 for selected-set declaration, E.17 and E.24.PUB for publication, C.18 for archive and front relations, C.19 for pool treatment, C.36 for cultural-evolution claims, F.17 for exact local SchemeSenseCells, F.18 for name settlement, F.9 for obtaining Bridges, and A.6.RCD for a derived relation kind.

Use the existing result or publication for a needed currentness conclusion or limitation. Use RefreshPlan@Context, RefreshReport@Context, DeprecationNotice@Context or EditionBumpLog@Context when the corresponding plan, performed work, deprecation or edition change actually occurs; do not add an empty ticket or notice for unchanged applicability.

When the governed object is a reusable A.6.RCD predicate definition or an admitted derived relation kind, the currentness line pins the exact base definitions, named substrate and edition, authorized derivation operation, and applicability scope. A change to any of them reopens the affected derivation and its dependent uses under A.6.RCD; G.11 schedules the bounded refresh but does not redefine the relation or derivation.

G.11:4.3 - Orchestration semantics (conceptual; delegating to governing definitions)

Use G.11 to plan scoped actions from typed causes; action semantics remain with their subject patterns.

4.3.1 Ingestion. Consume RSCR triggers from:

• telemetry hooks (e.g., G.8, G.10, G.12),
• freshness and decay events (B.3.4),
• evidence, bridge, policy, edition, relied-on base-definition, named-substrate-edition, or derivation-applicability edits (from the respective subject patterns’ publication faces, forms, or units).

Every ingested signal is normalized into an RSCRTrigger (canonical id, scope, payload pins), with optional alias labels.

4.3.2 Scope closure over the actual dependencies. Compute the minimal dependency closure over:

• cited evidence and source relations, with G.6 PathId and PathSliceId refs when a graph path slice is the current math-lens expression,
• declared crossings (G.7 sentinels; CrossingBundle visibility),
• and pinned references (editions and policies).

G.6 supplies graph expression and citation when used; a graph does not establish the source or dependency relation. For a nongraph result, name the exact source, receiving result/use and dependency under PatternScopeId; use PathSliceId when that dependency scope is actually graph-expressed or independently required by the receiving contract. The closure is a planning-time claim about affected slices, distinct from execution of the planned refresh actions. Interpret a B.3.4 trigger for the receiving claim and use: available information may establish continued applicability, a narrower use, an obtainable refresh need or a necessary suspension. An age-only signal does not determine that disposition. If support remains sufficient, stop with the usable result; retain only the limitation or reason a later recipient needs.

4.3.3 Planning (P2W boundary). When the selected response requires planned refresh, use C.11 and C.19.2 for its marginal contribution, cost, delay and displaced work. Produce RefreshPlan@Context for the actions actually selected; possible action forms include:

• RerunHarvest (delegates to the selected harvesting or SoTA method, such as G.2; use G.1 additionally only when its generator-kit cards or wiring must change)
• RerunParity (delegates to G.9)
• RecomputeSelectionOrSetResult (delegates to G.5)
• RebindBridgeOrCrossing (delegates changes to the obtaining Bridge to F.9, calibration-record changes to G.7, and crossing visibility to E.18 and the applicable visibility harnesses)
• UpdateEvidenceBindings (delegates to G.6)
• ReshipPack (delegates to G.10)
• UpdateBundle (delegates to G.8)
• UpdateDashboardSlice (delegates to G.12)
• EmitDeprecationNotice or EmitEditionBumpLog (publication units governed by this pattern)

4.3.4 Execution and audit. When selected actions are performed as Work or Work-bound audit, publish the corresponding RefreshReport@Context. A scoped applicability judgement can reuse available information without a new experiment; a plan alone establishes neither performance nor a new observation. Gating outcomes (admit, degrade, or abstain) follow G.Core tri-state semantics and are recorded through policy ids and cited evidence or source relations, rather than as local bespoke outcomes.

G.11:4.3a - Causal-use refresh sentinels

When a shipped result consumes C.28, refresh planning watches the causes that can change a supported use, unsupported use, support-result verdict, limits, or downstream decision basis:

These are payload distinctions under existing G.Core trigger kinds, not new trigger kinds. Reopen only the affected result and downstream uses that consumed it.

G.11:4.4 - Extensions (pattern-scoped; non-core)

Discipline-specific refresh strategies and generator-specific wiring live as GPatternExtension blocks. Scheduling, ordering, priority, and budget policy for the refresh queue are not separate extension semantics: G.11 defines the required policy pins on RefreshQueue and RefreshPlan@Context, while A.15.2 and A.15.3 keep the WorkPlan and its local content separate from dated Work.

G.11:Ext.TriggerAliases

PatternScopeId: G.11:Ext.TriggerAliases GPatternExtensionId: TriggerAliases GPatternExtensionKind: InteropSpecific (alias docking) GoverningPatternId: G.Core Uses: {G.Core} (cites G.Core.TriggerAliasMap.G11) ⊑ and ⊑⁺: ∅ Required pins, edition pins, and policy pins (minimum):

• RSCRTriggerKindId[] (canonical ids recorded on triggers)
• RSCRTriggerAliasId? (e.g., G.11:T0…T7 as labels only)
• scope: PathSliceId[] | PatternScopeId

RSCRTriggerKindIds: {RSCRTriggerKindId.EditionPinChange, RSCRTriggerKindId.PolicyPinChange, RSCRTriggerKindId.TelemetryDelta, RSCRTriggerKindId.FreshnessOrDecayEvent, RSCRTriggerKindId.CrossingBundleEdit, RSCRTriggerKindId.PenaltyPolicyEdit, RSCRTriggerKindId.MaturityRungChange, RSCRTriggerKindId.EvidenceSurfaceEdit} Notes (wiring-only): This block does not define what T0…T7 mean; it only preserves the labels and requires docking via G.Core.TriggerAliasMap.G11.

G.11:Ext.DecayAndDebt

PatternScopeId: G.11:Ext.DecayAndDebt GPatternExtensionId: DecayAndDebt GPatternExtensionKind: DisciplineSpecific GoverningPatternId: B.3.4 (use-qualified currentness and interpreted planning debt) Uses: {B.3.4, G.6} ⊑ and ⊑⁺: ∅ Required pins, edition pins, and policy pins (minimum):

• The receiving claim/use and the changed premise or applicable review condition, with source references where published.
• FreshnessWindowDeclRef?, DecayPolicyIdRef? or EpistemicDebtBudgetRef? only when the adopted window, deterioration model or planning measure is used. Their source supplies the meaning; no default expiry or debt budget is required.
• Exact dependent claims and uses actually affected, expressed as PatternScopeId for nongraph scope or PathSliceId[] for graph scope. Retain graph pins when an independently applicable profile requires them; carrier age alone does not select every use of that carrier.

RSCRTriggerKindIds: {RSCRTriggerKindId.FreshnessOrDecayEvent, RSCRTriggerKindId.EvidenceSurfaceEdit, RSCRTriggerKindId.BaselineBindingEdit} Notes (wiring-only): B.3.4 determines what the trigger means for the use. Continue, narrow, refresh, suspend or an authorized exception remain available where warranted; no Refresh/Deprecate/Waive triad or automatic downgrade is introduced here. Currentness is not assurance of the underlying claim. Budget and priority logic apply only when their interpreted policies are used.

G.11:Ext.QDRefreshWiring

PatternScopeId: G.11:Ext.QDRefreshWiring GPatternExtensionId: QDRefreshWiring GPatternExtensionKind: MethodSpecific GoverningPatternId: C.18 (QD semantics; descriptor, distance, and insertion) Uses: {C.18, C.19, G.5, G.8} ⊑ and ⊑⁺: ∅ Required pins, edition pins, and policy pins (minimum):

• DescriptorMapRef.edition, DistanceDefRef.edition
• CharacteristicSpaceRef.edition? (required when a domain-family coordinate is declared by the QD governing definition)
• InsertionPolicyRef, EmitterPolicyRef (policy-bound)
• Exact archive or illumination scope and policy-id for emitted telemetry triggers: PatternScopeId for nongraph scope; PathSliceId when graph-expressed or required by an independently applicable parity, evidence or shipping contract.

RSCRTriggerKindIds: {RSCRTriggerKindId.TelemetryDelta, RSCRTriggerKindId.EditionPinChange, RSCRTriggerKindId.PolicyPinChange} Notes (wiring-only): G.11 does not restate QD semantics; it ensures pins are present so reruns are comparable.

G.11:Ext.OEERefreshWiring

PatternScopeId: G.11:Ext.OEERefreshWiring GPatternExtensionId: OEERefreshWiring GPatternExtensionKind: MethodSpecific GoverningPatternId: C.19 (open-ended exploration and exploration-exploitation logistics) Uses: {C.19, G.5, G.8, G.9} ⊑ and ⊑⁺: ∅ Required pins, edition pins, and policy pins (minimum):

• TransferRulesRef.edition, EnvironmentValidityRegion (when OEE is declared by the subject patterns)
• GeneratorFamilyRowRef = <GeneratorFamilyId, rowEdition> and TransferRulesRef wiring pins (as published by G.5 and the governing definitions); resolve the exact row edition used by the affected result
• exact telemetry scope and policy-id: PatternScopeId for nongraph scope; PathSliceId when graph-expressed or required by an independently applicable parity, evidence or shipping contract.

RSCRTriggerKindIds: {RSCRTriggerKindId.EditionPinChange, RSCRTriggerKindId.TelemetryDelta, RSCRTriggerKindId.PolicyPinChange} Notes (wiring-only): Any OEE method semantics live with the governing definition; this module only wires refresh triggers to comparable reruns.

G.11:4.4a - Scheduling and priority policy pins

Scheduling strategies (bandit-style allocation, queueing, cadence policies, early stopping, or manual priority rules) may influence the order and budget of refresh work, but they do not define trigger meaning, action semantics, parity semantics, shipping semantics, or Part-G-wide defaults.

G.11 therefore treats scheduling as policy-bound refresh planning:

• RefreshPriorityPolicyIdRef names the policy used to order or prioritize queue items.
• BudgetDeclRef names the time, compute, cost, risk, or cadence boundary for the planned refresh.
• RSCRTriggerKindId[] still comes from G.Core; scheduling policy does not mint trigger kinds.
• planned refresh remains the exact U.WorkPlan locally called RefreshPlan@Context; executed refresh is recorded in RefreshReport@Context or Work-bound audit.

If no priority or budget policy is declared, no scheduling heuristic is admissible by appearance; the plan must either use the ordinary queue order or state the missing policy pin as a blocker.

G.11:5 - Archetypal Grounding — System and Episteme (informative; Tell–Show–Show)

U.System illustration — Safety-critical maintenance loop (pump and calibration). A centrifugal pump is serviced under a documented procedure (method description). Sensors report vibration drift (telemetry), and a calibration standard is updated (edition bump). The maintenance team uses G.11 to produce a refresh plan scoped to the affected inspection slices and publishes a refresh report of the executed actions with pins to the updated standard edition and the evidence or source relations. Deprecation notices are issued for obsolete thresholds in the procedure’s acceptance clauses (by subject pattern), preserving ID continuity.

U.Episteme illustration — Living review and benchmark pack (claims and parity). A claim sheet behind a shipped SoTA pack changes (new evidence, retraction, or revised measurement definition). Calibration evidence changes, potentially affecting a bounded-use claim or a justified loss consequence. The maintainers identify which receiving uses depend on the changed row, retain matching results that remain supported, and use the canonical triggers to plan any needed targeted parity rerun. Re-shipping follows only when the publication needs the resulting change; a CL revision alone grants or withdraws no use.

Paired currentness case. A pack’s export-date label changes, but its relied-on claims, source editions, qualification conditions and receiving use remain unchanged and adequately supported by available information. Retain the result; no refresh plan, waiver or no-refresh notice is needed. In the paired case, a dependency changes so that the shipped benchmark comparison no longer supports use beyond its stated window. Restrict that comparison and retain the warning with the shipped result so a later receiver cannot infer continued comparability. Plan the targeted check or update when it is justified and obtainable; currentness reporting alone does not repair the comparison.

Three bounded currentness cases. These are constructed applications of the same rule.

• A pump shortlist still consumes the same two immutable method rows, eligibility conditions and source edition. An age alert supplies no changed premise or expired use condition, and the available support remains sufficient for the same triage use. Retain the shortlist; no plan or omission certificate is needed.
• A local selected-set result consumes PumpReviewBudget-E1. Its replacement PumpReviewBudget-E2 changes the allowed review time from 30 to 20 minutes. The existing result explicitly cites that budget; no evidence graph is in use. Scope PumpTriageSelection names that exact result, changed budget and dependent eligibility comparison. If re-selection is chosen, plan only that comparison through G.5 under PatternScopeId=PumpTriageSelection, carrying both budget editions and the current task/row refs. Record performed refresh and any required targeted check in the resulting report before republication.
• A shipped QD archive expresses its affected dependencies in PathSliceId=ArchiveCell-Q7. A distance-definition change reopens the comparison using that definition. Retain the slice, old/new DistanceDefRef.edition, descriptor, insertion/emitter and policy pins, and the applicable G.9/G.10 evidence and shipping requirements. An OEE change to TransferRulesRef.edition similarly retains its affected graph slice, exact generator row edition and environment-validity scope. Nongraph support elsewhere does not relax these contracts.

G.11:6 - Bias-Annotation (informative)

Bias lenses: Gov, Arch, Onto and Epist, Prag, Did.

• Arch bias (toward explicit wiring). Risk: authors feel “over-pinned.” Mitigation: keep the minimum pin set small; push scheduling sophistication into extensions and policies.
• Gov bias (toward audit over speed). Risk: refresh becomes bureaucratic. Mitigation: create queue, plan and report content only for the corresponding need; retain applicable support without an omission certificate, while preserving warnings a later receiver needs.
• Onto and Epist bias (toward one governing definition semantics). Risk: teams try to localize trigger meaning for convenience. Mitigation: alias docking is allowed, but semantics stay in G.Core.
• Prag bias (toward minimal recomputation). Risk: under-refresh if closure is too narrow. Mitigation: require closure rationale and allow explicit “scope wideners” as policy-bound pins.
• Did bias (toward readable, reusable artefacts). Risk: oversimplified examples. Mitigation: maintain System and Episteme grounding and keep SoTA-echoing explicit.

G.11:7 - Conformance Checklist (normative)

G.11:8 - Common Anti-Patterns and How to Avoid Them (informative)

G.11:9 - Consequences (informative)

• Selective, replayable upkeep. Refresh becomes a controlled planning and execution loop rather than an implicit “maintenance vibe.”
• Stable semantics with flexible operations. Trigger meaning is centralized (G.Core), while scheduling sophistication can evolve as policy-bound extensions.
• Clear governing-definition assignment boundaries. Orchestration coordinates actions under their governing definitions; it does not redefine their semantics (shipping remains G.10, selection remains G.5, etc.).
• Cost: pin discipline overhead. Authors must carry enough ids, editions, and policies to make refresh comparable. This is intentional: it replaces hidden drift with explicit wiring.

G.11:10 - Rationale (informative)

G.11 is intentionally a thin orchestration governing definition:

• The refresh loop coordinates reruns and republishing; trigger semantics, invariants, and defaults are delegated to G.Core.
• The kit is split across the P2W planning-to-work boundary so that the exact U.WorkPlan and its declaration-local planned-filling rows remain planning content while dated Work remains independently established.
• Alias stability is maintained by allowing trigger aliases (T0…T7) while prohibiting them from becoming semantic authorities.

G.11:11 - SoTA-Echoing — Post‑2015 practices aligned (informative)

Each entry follows: claim → practice → source → alignment → adoption status.

0. QD currentness requires visible survey support. Practice: current QD work is surveyed as approaches, applications, and challenges, with archive, diversity, descriptor, and evaluator-currentness concerns still live. Source: A survey on Quality-Diversity optimization: Approaches, applications, and challenges, Swarm and Evolutionary Computation 2026, DOI 10.1016/j.swevo.2025.102240. Alignment: RefreshCurrentnessLine@Context may name selected set, Front, Q-front, ExplorationArchive, Archive, portfolio lineage, descriptor or distance edition, and path-slice scope, while C.18, C.19, and G.5 keep archive, pool, and selected-set meanings. Adoption: Adopt and bound (survey support changes refresh currentness fields and boundaries; it is not the governing ontology source).

0a. Open-ended engineering outputs need source and evaluator currentness. Practice: self-improving-agent, AlphaEvolve-style, and DeepEvolve-style lines use generated variants, external knowledge, evaluators, tests, archives, and empirical validation. Source: Darwin Godel Machine arXiv:2505.22954, AlphaEvolve arXiv:2506.13131, and DeepEvolve-style deep-research augmentation arXiv:2510.06056. Alignment: G.11 refresh records carry source, evaluator, descriptor, policy, edition, lineage, and report refs; generated method text, evaluator success, and archive update keep their subject patterns. Adoption: Adopt and adapt (refresh tracks currentness and smallest affected scope; it does not accept generated text as proof, gate passage, or performed work).

1. Continuous refresh is necessary in deployed evaluation pipelines. Practice: production ML systems use monitoring, retraining, and reevaluation triggers and insist on reproducibility hooks. Source: Breck et al., The ML Test Score (arXiv:1706.04599, 2017); Amershi et al., Software Engineering for Machine Learning (ICSE-SEIP 2019). Alignment: G.11 formalizes triggers as typed causes and forces edition and policy pins for replay. Adoption: Adopt and adapt (adapted to id-based, PathSlice-scoped refresh rather than “retrain everything”).

2. Non-stationarity requires explicit drift and decay handling, not ad-hoc updates. Practice: continual learning emphasizes non-stationarity as a first-class maintenance condition. Source: Parisi et al., Continual Lifelong Learning with Neural Networks (arXiv:1802.07569, 2019); De Lange et al., A Continual Learning Survey (arXiv:1909.08383, 2021). Alignment: B.3.4 supplies use-qualified currentness. G.11 interprets changed conditions and justified review signals before planning an affected-scope refresh; elapsed time alone implies neither truth decay nor deprecation. Adoption: Adapt (refresh of conceptual artefacts and evidence closures, not untracked model mutation).

3. Quality-Diversity requires archive semantics and comparability under descriptor and distance evolution. Practice: QD methods treat the archive as the primary result and track changes under policy and edition conditions. Source: contemporary QD families such as CMA-MAE (arXiv:2205.10752) and differentiable QD (arXiv:2106.03894). Alignment: QD-specific meaning lives with the subject patterns; G.11:Ext.QDRefreshWiring ensures edition pins and scope pins exist so targeted archive refresh is admissible. Adoption: Adopt (set and archive preservation; no covert scalarization).

4. Open-endedness co-evolves environments and agents; transfer rules must be versioned. Practice: POET-class open-ended systems require explicit transfer rules and environment validity constraints. Source: Wang et al., POET (arXiv:1901.01753, 2019); later generator-family claims require a named G.2 SoTA pack or exact current source. Alignment: G.11:Ext.OEERefreshWiring requires TransferRulesRef.edition and scope pins so refresh reruns remain comparable and auditable. Adoption: Adopt and adapt (adapted to Part G pin and UTS publication discipline).

5. Efficient orchestration benefits from bandit and early-stopping scheduling, but scheduling must not redefine trigger, action, parity, shipping, or Part-G-wide default semantics. Practice: modern hyperparameter and experiment scheduling uses bandit-style resource allocation and asynchronous early stopping. Source: ASHA (arXiv:1810.05934) and BOHB (arXiv:1807.01774) as representative post-2015 scheduling practice. Alignment: scheduling is expressed as RefreshQueue and RefreshPlan@Context policy pins (RefreshPriorityPolicyIdRef, BudgetDeclRef) so core semantics remain stable and the exact U.WorkPlan stays separate from dated Work. Adoption: Adapt (useful practice, but quarantined outside core norms).

G.11:12 - Relations

Builds on: G.Core (Part‑G invariants; RSCR trigger catalogue; alias docking; Default Governing Definition Index), G.6 (EvidenceGraph, PathId and PathSliceId), G.7 (Bridge sentinels; CL, Φ, and plane pins), G.5 (selector and set-return), G.8 (bundle telemetry hooks), G.9 (parity), G.10 (shipping hooks), B.3.4 (freshness and decay), E.18 (GateCrossing visibility). Coordinates with: G.12 (dashboard telemetry pins), A.6.RCD for reopening reused predicate definitions and derived relation kinds when their base definitions, named substrate edition, authorized derivation operation, or applicability changes, C.18 and C.19 archive, front, and live-pool policy pins, C.32.P2S when telemetry, decay, or freshness reopens architecture problem-to-structure carry-through, C.23 (SoS-LOG branches and maturity ladders), C.28 (causal-use support results, verdicts, supported-use values, unsupported-use values, and SoTA-sensitive causal-use sentinel payloads), F.15 (RSCR harness publications, when present). Publishes to: UTS (refresh plan, refresh report, deprecations, edition bumps), and to the relevant subject patterns’ publication faces, forms, or units through delegated actions.

G.11:End