AQ-LANG-001 — Language discipline standard
PURPOSE AND SCOPE
This standard governs the wording of every artefact that leaves the programme record: charter parts, design notes, review packs, demonstration scripts and captions, interface strings, run reports, correspondence and any briefing derived from them. It restricts what may be claimed. It places no restriction on what may be built or attempted; ambition belongs in the capability statement, discipline belongs in the evidence statement, and the two are written in the same sentence.
Architecture tier CA. Evidence level of this document: Theory — it is a defined process, not a validated one. It applies from DM-0 onward and is binding on all ten layers L0–L9.
Unhackable. Impossible to intercept. Universally quantum-secure. Information-theoretically secure at system level. Satellite-ready. Space-proven. Operational. These are not degraded into hedged forms; they are removed. No qualifier rescues them, because each names a property the programme has no measurement path to at its current rung.
secure, resilient, autonomous, high-precision, low-latency, quantum-secure, robust, real-time, scalable, reliable, intelligent, proven, validated. None may stand as a bare adjective. Each may appear only in a sentence that also names the measured quantity (or its explicit absence), the conditions under which it applies, and the evidence level.
TENSE RULE
Nothing has been built and no element has reached Lab validation. Past tense is therefore permitted only for process events — a review was held, a document was issued, a rule table was fixed. Past tense about system performance is prohibited outright, including in figure captions, commit messages and internal drafts, because internal drafts are the source text for external artefacts and the tense survives the copy.
WORKED PAIRS
| Topic | Unsupported form (prohibited) | Evidence-aligned form |
|---|---|---|
| LP-01 Autonomy | "The node operates autonomously." | "The node runs a defined decision procedure without operator input for the command classes listed in the L5 catalogue. Every consequential transition remains hardware-gated at TB-4, and Q4 can only reduce what Q1–Q3 allowed. No unattended run has been performed. Evidence level: Concept." |
| LP-02 Intelligence | "AI-driven detection identifies attacks in real time." | "L8 produces advisory annotations on evidence records. An annotation cannot set a trust state, raise an uncertainty band or authorise a command; a human decides. Detection performance is unmeasured and no detection metric is yet defined." |
| LP-03 Real-time performance | "Low-latency response to trust events." | "TB-2 acquisition is specified as a bounded-interval path in FPGA logic. The metric is the acquisition-to-decision interval; its bound is a design parameter to be fixed from bench measurement at DM-1. No interval has been measured." |
| LP-04 Robustness | "Resilient to interference and jamming." | "For the interference cases enumerated in F-3 and F-7 the defined behaviour is a transition to RESTRICTED at band UB-3 with an abstention record. Whether the implementation exhibits that behaviour is untested; no interference tolerance is characterised." |
| LP-05 Reliability | "Highly reliable; targeting 99.9% availability." | "No availability figure exists and none is asserted. Availability will be reported only as observed uptime over a stated window on a named article, with every failure mode excluded from the count listed beside it." |
| LP-06 Scalability | "Scales seamlessly to a full constellation." | "The reach ladder defines six rungs. The architecture is written so that rungs 4–6 reuse the L3–L6 interfaces without change. Nothing above rung 1 has been built; this is a stated design intent, not a measured property." |
| LP-07 Live validation | "Validated live in the field." | "Run AQ-R-nnnn executed the scripted sequence on AQ-NODE-01 at DM-2 under the conditions recorded in the run file. It exercised the decision path only; link performance was not measured. Evidence level: Lab validation for the decision path, Concept elsewhere." |
| LP-08 Breakthrough | "A breakthrough in quantum communications." | "The programme states a research question and the measurement that would answer it. No result has been produced. No claim of novelty is made until a result is reproduced under an independent run identity by a person who did not perform the original run." |
| LP-09 Safety | "Fail-safe by design." | "The defined behaviour on loss of a required input is safe-hold with a recorded abstention (DC-2, DC-3). Whether the implementation exhibits that behaviour across F-1–F-14 has not been tested. Safety here is an argued property with an open evidence set, not a demonstrated one." |
| LP-10 Repeatability | "Consistently reproduces the result." | "n runs under the declared conditions produced the recorded values; dispersion is reported with each run identity. Below the run count fixed in the test plan, repeatability is reported as 'not established' rather than as a weaker positive." |
| LP-11 Quantum claims | "Quantum-secure by physics; interception is impossible." | "The key-exchange element targets a security argument stated against a named adversary model and a named set of device assumptions. No element has reached Lab validation, no security property is measured, and the device assumptions are unverified. A component-level argument is never restated as a system-level property." |
| LP-12 Readiness | "Satellite-ready, space-proven, operational." | "The article sits at reach-ladder rung 1 at demonstration maturity DM-0. Environmental qualification has not begun. There is no flight heritage and no deployment." |
| LP-13 Comparison | "Outperforms existing systems." | "No comparative measurement has been made. A comparison will be stated only when the same metric is measured on both articles under conditions recorded in one run plan, with every configuration difference listed alongside the numbers." |
| LP-14 Cost efficiency | "Dramatically cheaper than the alternatives." | "Unit cost is undefined because no article has been built. Cost per node will be reported as a bill of materials plus recorded build hours, for the configuration actually built and for no other." |
| LP-15 Precision | "High-precision timing across the link." | "L2 defines the timing-integrity quantity as offset against the disciplined reference over a stated interval. Its value is unmeasured; its target will be set from the first bench distribution, not chosen in advance." |
| LP-16 Provenance | "Fully traceable and auditable end to end." | "DC-5 requires every consequential transition to be reconstructable from the Evidence Fabric alone. The reconstruction test is defined in AQ-NODE-01 and has not been executed; traceability is therefore a requirement, not a finding." |
| LP-17 Funding | "Backed by partners and government collaboration." | "The programme is funded from EduRankAI revenue and disciplined reinvestment. External capital, partnership or collaboration would accelerate defined work packages; none is a precondition for the programme existing, and no counterparty is named in any artefact before a signed instrument exists." |
| LP-18 Programme relationship | "Akasha-Q is the payload of the Visvambhara satellite." | "Visvambhara is a separate aerospace programme and the first platform designed to carry an Akasha-Q terminal. Akasha-Q is a communications programme in its own right, spanning five layers across the six-rung reach ladder; the two share one restricted-access record." |
THE GENERAL PATTERN
Every acceptable form above satisfies the same four-slot structure. A statement missing any slot is not a weaker claim — it is an unreviewable one, because a reader cannot tell what would falsify it.
Figure 1 - Anatomy of an acceptable capability statement
+--------------------------------------------------------------+ | slot 1 CONDITIONS | | configuration, environment, assumption set, | | adversary or failure model where relevant | +--------------------------------------------------------------+ | slot 2 RUN IDENTITY | | which run(s), on which article, at which DM level; | | 'no run has been performed' is a valid filling | +--------------------------------------------------------------+ | slot 3 QUANTITY | | the named metric with its value and DC-1 tag | | (MEASURED / DERIVED / MODELLED) | | OR the explicit absence: 'the metric is undefined' | +--------------------------------------------------------------+ | slot 4 NOT ESTABLISHED | | what the statement does not support, named | | positively, in the same sentence or the next | +--------------------------------------------------------------+ slot 3 absent -> the sentence is marketing, not a claim slot 4 absent -> the reader will supply the strongest reading slot 2 absent -> the claim cannot be re-run and cannot be retired
Two consequences follow directly. First, a capability may be described as ambitiously as the design warrants, provided slot 4 is filled: "the terminal is designed to operate across all six rungs; nothing above rung 1 exists" is compliant, while "a six-rung terminal" is not. Second, an absence is written as a value, mirroring DC-2 at the level of prose: "the metric is undefined and will be established from bench measurement" is an acceptable slot-3 filling; silence is not.
REVIEW PROCEDURE
- Pass 1 — reserved-word sweep. Locate every reserved descriptor and every past-tense verb attached to system behaviour. Each occurrence is either rewritten to the four-slot form or deleted.
- Pass 2 — slot audit. For each remaining capability sentence, mark which of the four slots it fills. Any sentence below four is returned to its author with the missing slot named, not silently edited.
- Pass 3 — evidence-level check. Every capability statement carries an evidence-level chip drawn from the Part 12 ladder, and no chip may exceed the level of the run record it cites. A statement citing no run cannot exceed Concept or Theory.
- Derived artefacts inherit. A slide, caption or summary derived from a compliant document must carry the same evidence level and the same slot-4 statement. Compression is not an exemption; if the four slots do not fit, the claim is cut, not the qualification.
WHAT THIS STANDARD DOES NOT DO
It does not decide whether a claim is true, and it does not substitute for review by the layer that owns the subject. A sentence can satisfy all four slots and still be wrong about the physics. It also carries no geopolitical, sovereign-capability or defence framing of any kind, and artefacts introducing such framing are out of scope of the programme record entirely rather than being rewritten into compliance. Conflict CF-4 — the tension between describing intended capability and the reader's tendency to read intent as achievement — is managed by slot 4 and is not considered closed.