v8.5 (2026-08-01) 35 states 7 axes

Substrate entanglement is geopolitical, not universal.

QESIS+ measures what a state's digital sovereignty physically rests on: water, submarine cable, rare earths, foreign platforms, hyperscale operator concentration, cloud risk density and electricity. Seven axes, 35 states, every score reproducible from a primary portal.

0.124coupling, global n=32
0.176coupling, import core n=26
32ranked
3published gaps

Energy-rich states decouple through cheap, reliable, carbon-heavy power. For states that import their energy, their minerals and their hyperscalers, every axis is bound to every other. That is the finding, and the gap between those two numbers is the argument.

Source: derived at build time from the published axes. rho = C/6 trace-normalised, CR = 1 - S(rho)/ln 6. Coverage 32 of 35 global, 26 of 35 core. Vintage v8.5 (2026-08-01).

The gap is the finding

The Binary Integrity Guard never imputes. A state missing a weighted axis is not ranked and not quietly zeroed: it is published as a gap with its coverage stated. Below the declared 0.75 threshold, no composite is emitted at all.

Coverage is the declared weight mass actually present.
StateCoverageAbsent
Hong Kong0.70WSE
Singapore0.70WSE
Taiwan0.70WSE

Ranking

#StateComposite ODIESECov
1Saudi Arabia75.6100.042.31.00
2United States68.236.843.11.00
3Spain60.233.327.71.00
4Belgium59.750.028.41.00
5Bahrain58.1100.034.51.00
6United Arab Emirates56.555.623.21.00
7Qatar50.950.028.41.00
8Poland49.150.059.31.00
9India48.730.663.91.00
10Chile47.750.043.31.00
11Israel46.825.042.51.00
12Netherlands46.1100.033.31.00
13Denmark44.0100.023.11.00
14South Africa43.928.075.11.00
15Australia43.330.950.21.00
16Italy42.825.044.11.00
17France42.528.025.01.00
18United Kingdom42.137.544.31.00
19Mexico41.733.345.41.00
20Indonesia40.433.357.41.00
21Sweden39.750.018.21.00
22Finland38.8100.022.51.00
23Japan38.726.539.61.00
24South Korea33.750.031.41.00
25Germany33.328.040.81.00
26Malaysia32.550.044.51.00
27Austria31.2100.027.11.00
28Norway29.7100.016.31.00
29Brazil28.636.039.31.00
30Canada26.025.029.01.00
31New Zealand20.850.029.11.00
32Switzerland19.637.520.41.00

Composite = 0.3*WSE + 0.3*CSE + 0.17*ODI + 0.08*RGD + 0.15*REE. Derived at build time, never carried. ODI is the continuous Herfindahl measure over operator shares of active cloud regions. Sources: v8_qesis_country_scores.csv, cloud_regions_master.csv. Generated 2026-08-01T12:20:25Z.

Today's focus: Norway

Norway sits at composite 29.7 with coverage 1.00. Its binding constraint is ODI, and its operator concentration reads 100.0 across 1 providers over 2 active cloud regions.

Focus rotates daily so the instrument does not read as a single case study. Asset: qesis_v8.json, coverage 35 of 35, vintage v8.5 (2026-08-01).

What this vintage does not know

Every known limitation, derived from the index at build time rather than kept by hand. An instrument that publishes its own uncertainty is checkable; one that does not is a claim.

IdSeverityLimitation
U-01high3 of 35 states carry no composite: HKG, SGP, TWN. Water stress is absent and coverage falls to 0.70 against the declared 0.75 threshold. The absence has two distinct causes, published per state as withholding_cause.
U-02highThe published fsQCA is not a calibration of the published axes. Five independent violations are recorded as D-103: identical raw values with divergent fuzzy membership, an inverted order between GBR and DEU on WSE, a near-constant condition present in every sufficient path, solution statistics that are internally inconsistent, and an outcome variable that contradicts the index it is meant to explain.
U-03mediumFidelity is computed for 3 of 35 states (DEU, ESP, GBR).
U-04mediumAvailability-zone counts are missing for 27 of 35 states, so the ODI Herfindahl is weighted one unit per active cloud region rather than by capacity.
U-05mediumFPE, ESE are published per state and enter no composite (D-044).
U-06mediumNo semiconductor axis exists. The February 2026 roadmap named semiconductor capacity as the Infrastructure Pillar. Scoped as a declared roadmap item, target vintage v9.0, 2026-12-31, contingent on identifying a 35-state fabrication-capacity source. Until that date the roadmap claim is stated as pending, not delivered.
U-07lowThe v8.0 ordinal ODI carried four manual overrides (NLD, HKG, KOR, SGP), each one step below its own stated rule, recorded in no export.
U-08lowCSE composes EMODnet and SubmarineMap at 0.6/0.4 for EU states and uses SubmarineMap alone elsewhere.

Asset: qesis_v8.json uncertainty_ledger, 8 entries, vintage v8.5 (2026-08-01). Also served by qesis_get_integrity.

Connect it

The index is an MCP server, so any MCP-capable client can call it as a tool. Local, over stdio:

pip install mcp
python server.py

Claude Desktop:

{"mcpServers":{"qesis":{"command":"python",
  "args":["/path/to/qesis-mcp/server.py"]}}}

Without a licence key the server runs in demo tier: rounded scores, limited depth, component audit locked. That is the product working as designed.

Integrity

Every composite is recomputed from its own published axes on each build, and the deploy fails on drift. The v8.0 vintage carried a composite that could not be reproduced from the axes beside it: the United Kingdom scored at or above Switzerland on all seven axes yet carried a lower composite, which no non-negative weighting admits. It is derived now, and qesis_get_integrity answers which generation you are reading.

Formula QESIS_THEORY_v3. Gate: 16 checks plus a mutation test that injects each defect class the gate must catch.

Licence and contact

Demo vintage CC-BY-NC. Derived scores are own work; upstream sources keep their own licences and are cited per axis. TeleGeography-derived cable material is published only as derived aggregates, never raw.

For the institutional vintage, or to ask a question about the method, write to the address on the repository. There is no form on this page, because there is no backend behind it and a form that silently discards what you type is worse than no form. Nothing you do here is logged, and no cookie is set.