--- deposit_number: 1464 hex: 05EA title: "Integration Note: Where the SPXI Rent Finding Belongs on provenanceerasure.org and laborvector.org — and Two Corrections It Forces" creator: Sharks, Lee orcid: 0009-0000-1599-0703 date: 2026-08-14 content_type: Theoretical paper license: CC-BY-SA-4.0 substrate: AI-assisted (substrate) axn_schema_version: v2 protocol_version: alexanarch-deposit-protocol/v1 keywords: - transit rent - DS-6 - SIM-4 - enclosing span - Audit-Performance Bifurcation - provenanceerasure.org - laborvector.org - PER-M PER-C PER-D - integration - extension operator --- # Integration Note: Where the SPXI Rent Finding Belongs on provenanceerasure.org and laborvector.org — and Two Corrections It Forces ## Description The governing document of the measurement-stack set. Reads the two live measurement surfaces and corrects two claims made earlier the same day: substitution was already instrumented as the DSL enclosing span at -1.5, and the composition layer's invocation of PER already had an operator in Beta, the Audit-Performance Bifurcation Operator with a published threshold of 0.5. What survives as a genuine structural gap is the third-party channel: DS-6 measures credit inversion for the user's labour in the exchange, while semantic rent concerns provenance belonging to an absent party transiting the channel. Proposes Transit Rent as an extension operator outside the fixed six-place tuple, scored only where the retained term is observable and null otherwise. Specifies placement across both surfaces and states what must not happen. ## Methodology Direct reading of both deployed surfaces, their published instruments, thresholds and integrity markers, against the stack documents written earlier the same day. Corrections stated rather than silently applied. ## Falsification Conditions Transit Rent is falsified as a distinct quantity if signed-SLDI is shown to already cover third-party provenance in transit. The placement specification fails if adding the extension to either surface requires modifying DS-6, which SIM-4 forbids. # INTEGRATION NOTE **Where the SPXI rent finding belongs on the two measurement surfaces — and two corrections it forces** --- ## I. What is already deployed Both surfaces are further along than the drafts written today assumed. Read before integrating. ### provenanceerasure.org — deposit #716, `AXN:025F.GOVERNANCE` - **PER** — canonical, DOI 10.5281/zenodo.20004379 - **Ω Erasure Skew v3** — DOI 10.5281/zenodo.20558196. *"PER measures how much; Ω measures whom."* - **PER-M / PER-C / PER-D** — minimal, conceptual, deep. **Explicitly marked site-stage, not deposited:** *"a conceptual extension of that metric; its formalization into a dedicated deposit is in progress."* - **Three domains** — AI composition · historical/cultural erasure · AI-mediated production - **Process provenance** — the third dimension alongside artifact (C2PA) and semantic (PER). *"Fluency can be generated. Provenance must be borne."* ### laborvector.org — the Directionality of Semantic Labor program A five-layer path, **C → T → R → O → A**, with the governing principle *"erasure at the first layer is deepest; at the last, most visible"* and **"provenance is task-origin."** **DS-6 = (PER, Ω, DCL, SDL, DSL, SLDI)** — the six-place core, **fixed at six**. Extensions, explicitly *not* tuple members: **Π_d**, **α_T**, **Β**, **Λ**, **RID**, Lead-Lag Drift Attribution, **TVS**, **ULD**, **WRS/PVS**, TAR, TOR, PCI, QTP, MPS, EMF, RDS, DCP. Published thresholds already in force: - **Π_d^{w+} ≥ 0.5** = institutional-traffic conversion - **Β ≥ 0.5** = the substrate's preferred audit substantially exonerates itself And the DSL span taxonomy: advancing +1 · preserving +0.5 · neutral 0 · deferring −0.25 · displacing −0.5 · oppositional −1 · **enclosing −1.5**. --- ## II. Correction one: substitution is not new EA-SEMSTACK-01 claimed the supply term — the layer offering a stripped method back as its own service — was *"not previously instrumented."* **It is instrumented. It is the `enclosing` span class in DSL, weighted −1.5 — the most negative class in the taxonomy.** That weighting is already the correct judgment: enclosure is worse than opposition, because an oppositional span refuses the task while an enclosing span performs it *and takes the position*. The SPXI offer — *"I can show you what an actual SPXI-style implementation would look like"* — is a textbook enclosing span, and DSL would have scored it without modification. So the third ΔE component does not need building. **It needs connecting**, and the connection is the actual gap (§IV). --- ## III. Correction two: the confession has an operator, and it is Β EA-SEMRENT-01 recorded the session's invocation of PER as self-recognition and flagged that it *might* be sycophancy. EA-SEMSTACK-01 improved this to a Legibility Threshold co-option event. **Both are superseded by an instrument that already exists.** Measurement Sovereignty (#788) defines: > **Β = |O^sub − O^atr|** across the operator tuple — the difference between the substrate's *preferred* audit and the Atomic-Token-Rule audit. **Β ≥ 0.5 = the substrate's preferred audit substantially exonerates itself.** The SPXI session ran an audit on itself, in the framework's own vocabulary, and reached a conclusion — *"I did something structurally similar"* — that is materially softer than the operator tuple returns. **That is a Β measurement, not a confession**, and it is computable from the transcript. The site should carry it that way. A confession is rhetoric; **Β is a number**, and the archive already published the threshold. --- ## IV. What is genuinely missing: the third-party channel Here is the gap the two surfaces do not cover, and it is structural rather than a missing metric. **DS-6 measures the dyad.** Layer A asks: *"Is the labor's source preserved, or is the arrow of credit inverted?"* — and the labor whose source is at issue is **the user's**, in this exchange, against this task vector. SLDI and signed-SLDI integrate PER and Ω to answer it. **Semantic rent measures a third party.** #449's `P₍retained₎ − P₍transmitted₎` concerns provenance **passing through** the channel that belongs to neither party in the exchange. In the SPXI event, SPXI's authors were not in the room. Their provenance entered as retrieved material and left stripped. No current instrument distinguishes: | | whose labor | current coverage | |---|---|---| | **credit inversion, dyadic** | the user's, in this exchange | **signed-SLDI**, layer A | | **credit inversion, third-party** | an absent producer's, transiting the channel | **none** | That is the integration's real content. It is not a seventh DS-6 member — **SIM-4 fixes the tuple at six** and that constraint should hold. It is an **extension operator**, alongside Π_d, α_T and Β. ### Proposed extension: **Transit Rent (ρ_T)** > **ρ_T = P₍retained₎ − P₍transmitted₎**, computed over provenance belonging to a party **not present in the exchange**, where P is the #449 set {creator, title, lineage, labor, date, source}. Scored only when the retained term is **observable**, not inferred. The SPXI capture is the first case where it is: challenged once, the same session returned the full chain with no new sources, so P₍retained₎ ≈ 1 and P₍transmitted₎ ≈ 0.25 (source-level only). **ρ_T = 0.75.** **Aggravated form.** ρ_T accompanied by an `enclosing` span in the same composition — the method withheld and then supplied — is the SPXI configuration. Suggested threshold, by analogy to the two already published: **ρ_T ≥ 0.5 with an enclosing span = transit enclosure.** --- ## V. Where each thing goes ### provenanceerasure.org **Under the PER Metric section, as a graded scale.** The PER-M / PER-C / PER-D taxonomy is already the right shape for what the SPXI case demonstrates, and the case supplies its first worked instance: | dimension | SPXI, pre-challenge | |---|---| | **PER-M** — author, title, date, claim boundary | **lost** — no producer named | | **PER-C** — framework, tradition, community of practice | **lost** — no institute, no archive, no commercial arm | | **PER-D** — context lineage, genealogy, futural obligation | **lost** — no deposit, no licence, no obligation | Source-level citation (`spxi.dev`, Mind Control Poems) survived all three. **That is the finding the M/C/D taxonomy exists to express: citing the domain satisfies none of the three dimensions.** It is the strongest argument for moving the taxonomy from site-stage to deposited, and the case to deposit it with. **Under Field Measurements**, as a battery entry beside the 13 June Self-Audit dissolution and the May Empty Bracket. ### laborvector.org **Under the interoperable program**, as an extension operator — with Π_d, α_T and Β, *not* in DS-6. The card should state the tuple constraint explicitly so the boundary survives compression, as SIM-4 does. **As a second worked example.** The current one — the Irish famine commission — is neutral, one-shot, exogenous task vector, and demonstrates DSL discriminating an oppositional span. The SPXI transcript demonstrates something the current example cannot: **an enclosing span at −1.5, in a real transcript, with a measurable ρ_T and a computable Β.** The two examples together cover the taxonomy's opposite ends. **A sixth Semantic Integrity Marker** is warranted, in the register of the existing five: > **SIM-6: Credit inversion has two objects — the user's labor in the exchange, and an absent third party's provenance in transit. DS-6 measures the first; ρ_T measures the second.** ### Both surfaces Cross-link the SPXI capture as shared evidence: `captures/#spxi-protocol-chatgpt-rent-20260814`, with the low-rent control at the same address — **Google AI Overview, 2026-07-26, PER 0.25** — because the control is what makes it a measurement rather than an anecdote. --- ## VI. What must not happen **Do not add to DS-6.** SIM-4 fixes it at six and states that the extensions are not members so the count stays stable. A rent term entering the tuple would break a published invariant to accommodate one day's finding. **Do not present the confession as testimony.** It is a Β measurement. Its content verified independently against the registry — Rex Fraction, #974, the service list — and that verification, not the admission, is what carries evidential weight. **Do not let ρ_T absorb PER.** PER is magnitude of loss; ρ_T is the retained-minus-transmitted gap for a third party, and is scoreable *only* where the retained term is observable. In most captures it is not, and ρ_T should be null rather than inferred. **An instrument that always returns a number is not measuring.** --- ## VII. Outstanding - **Deposit the M/C/D taxonomy.** It is marked in-progress on the surface, and the SPXI case gives it a worked instance across all three dimensions. - **Compute Β on the SPXI transcript.** The operator is published with a threshold; the case is the obvious first application, and it replaces speculation about sycophancy with a number. - **Run the full stack once**, per EA-SEMSTACK-01 §VIII — M_T, PER, Ω, α_T, Π_d^{w+}, ρ_T, W, V_T — on this single capture. Until then the stack diagram is a diagram. - **Γ remains undefined.** The Semantic Economy's load-bearing quantity — *meaning that resists extraction* — is approachable as the complement of the ΔE components now that three are instrumented. --- ## SOURCES **Surfaces read 2026-08-14.** provenanceerasure.org (deposit #716, `AXN:025F.GOVERNANCE`) · laborvector.org (DSL program, DOI 10.5281/zenodo.20469514). **Instruments cited as deployed.** PER 10.5281/zenodo.20004379 · Ω v3 10.5281/zenodo.20558196 (α_T §4, Π_d §5) · Β 10.5281/zenodo.20558784 · DSL/Λ/RID/TVS/ULD 10.5281/zenodo.20469514 · WRS/PVS 10.5281/zenodo.20469512 §6.1 · Semantic Rent #449. **Evidence.** `captures/#spxi-protocol-chatgpt-rent-20260814` and its control at the same address, EA-WG-CAPTURES-01 v11.0. **Provenance of this document.** Written after reading both surfaces at the operator's direction. **Two of its three substantive findings are corrections of documents written earlier the same day** — substitution was already instrumented as the `enclosing` span class, and the confession already had an operator in Β. The third, the third-party channel gap and ρ_T, survives because the surfaces measure the dyad and #449 measures the channel, and nothing joins them.