Nishi Family › Compare › Water: Best-Tasting, Healthiest, From the Air

Nishi Compare · capabilities, evidence and delivery

Water: Best-Tasting, Healthiest, From the Air

Compare capabilities and delivery evidence. Source discovery, qualification, deployment and accepted outcomes are separate; an open contract names work still to verify.

The roadmap-binding board behind /compare/water. Rival codes copied verbatim from water.sota; Nishi cells are MEASURED -- the ranker requires the implementing symbol as a top-level func in the real organ source on disk.

Overview

The purpose, declared position and evidence coverage of this domain. Source presence and completed acceptance are different measures.

Where we are. Measured 2026-08-21 from water.sota and the organ sources on disk. STRONG and gate-proven: three organs (nx_water_awg 14/14, nx_water_quality 15/15, nx_water_design 15/15) that model five harvesting technology classes rather than one appliance, recommend across them for a climate and return NONE when none can feed, refuse over-limit lead arsenic nitrate PFAS bacteria THM with a per-contaminant reason, and compose end to end from raw condensate through treatment to remineralisation with the three stages proven DISTINGUISHABLE. Every assertion in those 44 teeth is an exact integer equality, which is the checkable form of the determinism claim and is not available to a float model at all. WEAK, and the sota page already says so in seven rows: no measured empirical data of any kind on either half -- the yield model has never been compared against a published field figure and the quality engine scores only values a caller hands it -- the sorbent envelope is a single min-RH threshold per class where the physics is an isotherm, there is no map from our refusal constants to the clauses of the standard we cite, and there is no admission contract for a live reading. THE HONEST BOUNDARY OF THIS WHOLE DOMAIN, stated here rather than smuggled into a rung: four of the seven absences -- physically produces water, installed appliance or hardware, filtration or membrane media, field-proven yield at scale as a deployed fleet -- are HARDWARE and this lane will never close them. They are deliberately absent from water.matrix, because a rung this lane cannot build is a rung that never closes and a backlog that cannot drain teaches everyone to ignore the board.

Where we need to go. Move this domain from A MODEL NOBODY HAS CHECKED to a model that answers to measured data and to the standard it cites. Concretely: the modeled yield is compared against each vendor's own published rated figure with the error printed, the quality engine scores a real published water-quality record instead of only supplied values, the sorbent envelope is a curve rather than a threshold, every refusal constant maps to a named clause with the uncovered clauses listed, and a stale or gapped reading reads UNOBSERVABLE rather than SAFE. The exceed this domain already holds is that its recommender REFUSES rather than guessing; every rung below extends that shape to a new axis instead of adding a number.

The unit. 1 u = one measured session-leg (organ plus gate plus the live flip). Local calibration: the three water organs and their 44 teeth were built and gated inside one working day on 2026-08-14, so a rung that EXTENDS one of them with a new function and its teeth is priced at 1 u and a rung that must also ingest and parse an external record is priced at 1.5 u.

Where we are: 5 open rungs. Three landed and gate-proven, five open. Every open rung is a row water.sota already publishes as an absence or a partial. Counts measured at emit below this line.

Cost to a model that answers to measured data: 2.5 u. WA1 the published-yield comparison and WA2 the measured-record ingest. These two are the whole of the honest answer to the sota's largest self-declared gap.

Cost to depth and the standard: 5 u. Adds WA3 the sorbent isotherm and WA4 the clause-coverage map.

Cost to live input: 6 u. Adds WA5, the admission contract for a reading stream. No hardware is required to build it and none is claimed by it.

Latest recorded release

No valid dated release entry is recorded for this domain.

Release entries describe recorded changes; they do not establish that every capability passed evaluation.

3 of 8 capability source symbols found|2 of them authored stronger-claims|5 open|coverage 375/1000|adoption 3 full / 0 partial

Evidence profile — what the gaps on this board actually are

Measured by nx_swcompare_evidence, read back by nx_evprofile_lib. Every figure is a count with its denominator — there is deliberately no score, no grade and no percentage anywhere in this band, because a stored scalar is a field a seat can edit and a counted partition is not.

proven able to fail counts the gates that have a RECORDED RED — nx_gate_bite mutated the gate subject, rebuilt it, watched the gate go red, and that record is inside the shared TTL. never bitten is its complement over the same denominator: those gates ran and were green, and nothing has ever shown them able to detect anything, so their green is a statement about this run and not about the gate. green at 0/0 is a separate and much weaker observation — the gate printed GREEN on a zero denominator, so its own tooth counter says it examined nothing. A gate can be green, non-zero, and still never bitten; that is the common case and it is now visible instead of implied.

partition: grounded + unsupported = 3 vs present 3 · named + unnamed + withheld = 5 vs open 5 · both reconcile

liar-kill conj=GpqN · first failing conjunct: present

graded document: BUILDROOT tree, 5131 bytes · gates map: NONE · stamped 37d 5h ago · STALE against the shared TTL of 604800s — this is evidence about a build nobody can vouch for now · source ../knowledge/status/evstamp_water.verdict

Gap class	What it is, and the work it names
CLAIM-ONLY	No declared gate ran for this board, so every capability mark here rests on a symbol existing in source and on nothing executing. The remedy is to WRITE a `.gates` map naming gates that already run — authoring, not research.
STALE	This profile is older than the shared TTL, or its clock ran backwards. It is reported as unusable rather than as fresh, because being wrong in the direction of refusing is the safe direction.

This band reports the referee counts. The per-axis worklist rows — which axis is unsupported, which watch is ready to flip, which gap is unnamed — are printed by nx_swcompare_evidence water itself and are not carried on the stamp, so this page names the classes and the producer names the rows. That split is stated rather than hidden: a count without a worklist is not actionable, and this band is honest about which half of that it is.

Production map

Follow the dependencies, declared acceptance criteria and recorded priorities. Inspect source binding before treating a rank as executable work.

Ranking source binding: PLAN_MATRIX_BOUND_ONLY. Recorded priorities require current acceptance evidence and resource checks before execution.

Ranking matches the captured plan and matrix only. Latest execution outcome, research freshness, accepted delivery and investment return are unverified.

Recorded priority estimates

Order from nx_compare_rank (nx_dr_ocm: (deficit + cost-of-delay + option + enables) x sponsor x self-sufficiency x momentum / cost). FINISH rows are rungs whose symbol is present but whose organ is short of full adoption: listed before new work by this heuristic. Priority is not measured delivery cost or execution readiness. Stamp: # asof=1790081635 domain=water target_version=0.1 rungs=8 done=3 open=5 finish=0 ranker=nx_dr_ocm

#	Stage	Rung	Priority	Derivation
#1	0.1	Measured water-quality record ingest (WA2) wq_ingest_report dossier	1066	v=8 m=2 c=15
#2	0.1	Modeled yield validated against published field figures (WA1) wa_validate_published dossier	1000	v=5 m=2 c=10
#3	later	Refusal constants mapped to named standard clauses (WA4) wq_std_clause_map dossier	800	v=6 m=2 c=15
#4	later	Sorbent uptake as a curve not a threshold (WA3) wa_sorbent_uptake dossier	400	v=2 m=2 c=10
#5	later	Admission contract for a live reading stream (WA5) wq_stream_admit dossier	200	v=2 m=1 c=10

Declared roadmap — contract, acceptance, executor, effort

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
Climate-to-harvester recommendation (WA0)	wa_recommend	LANDED and gate-proven 14/14. Four recommender teeth one per climate regime, and the load-bearing one is the NEGATIVE CONTROL -- a climate too dry to feed any class returns NONE rather than the least-bad guess. Listed so dependents carry its momentum	Organ	0 u
Never-poison safety gate (WA0B)	wq_safety_admit	LANDED and gate-proven 15/15. Five named negative controls one per contaminant class each asserting its OWN refusal reason, plus the positive control that makes them mean anything -- a clean sample must be ADMITTED, because a function that refused every input would otherwise score full marks on every refusal test [epa-npdwr]. Listed so dependents carry its momentum	Organ	0 u
Design capstone that composes the suite (WA0C)	wd_design_best_water	LANDED and gate-proven 15/15. One gate imports the harvesting model the quality engine and the design organ together and walks raw condensate through treatment to remineralisation, asserting the three stages are DISTINGUISHABLE -- an implementation that returned the same sample at every stage would pass a suite that only checked the final number. Listed so dependents carry its momentum	Organ	0 u
Modeled yield validated against published field figures (WA1) after WA0	wa_validate_published	The sota grades field-proven yield at scale as ABSENT and it will stay absent, because we deploy no fleet. The honest form of the gap is VALIDATION. ACCEPT RULE PRE-DECLARED: for each vendor whose rated conditions are stated in a mirrored and content-pinned reference [watergen] [source-hydropanel], the modeled litres per day at THOSE conditions is computed and the SIGNED error against the vendor's own published figure is printed per row -- a signed error, never an absolute one, so a model that is consistently optimistic is visible as a bias rather than averaged away. A vendor whose reference states no rated temperature and humidity reports UNWITNESSED as its own third state and does NOT score as agreement. ANTI-VACUITY TOOTH REQUIRED: a deliberately wrong coil temperature must push at least one row outside the declared band, or the comparison has not been shown to be capable of disagreeing	Organ	1 u
Measured water-quality record ingest (WA2) after WA0B	wq_ingest_report	Today every score is of numbers the caller supplied, which is why the sota grades lab-measured analysis ABSENT and the standard Best. ACCEPT RULE PRE-DECLARED: a mirrored public water-quality record parses into the NxWaterSample the engine already takes and scores through the EXISTING wq_taste_score and wq_safety_admit -- one ruler, never a second scorer for ingested data. An analyte the record does not carry is UNKNOWN, and a sample carrying an UNKNOWN regulated analyte makes the safety verdict ABSTAIN rather than ADMIT, because an axis that cannot see must abstain and not acquit [who-gdwq]. The partition must be printed and must sum: records read equals ingested plus refused plus unparseable. NEG-CONTROL: a record with an over-limit analyte is refused NAMING that analyte, and a clean record is admitted	Organ	1.5 u
Sorbent uptake as a curve not a threshold (WA3) after WA0	wa_sorbent_uptake	wa_class_min_rh is one integer per class, which can express PERMITTED but cannot express HOW MUCH, and that is why the sota grades sorbent low-humidity Part against SOURCE Hydropanel Best. ACCEPT RULE PRE-DECLARED: uptake is monotone non-decreasing in relative humidity across the whole tested range, it DISCRIMINATES two named sorbent materials at the SAME humidity -- a curve that returns one shape for every material has measured nothing [kim2025-amphiphilic] -- and the existing wa_class_works and wa_recommend teeth stay byte-identical afterwards, so the depth is proven ADDITIVE rather than a rewrite that moved the answers	Organ	1 u
Refusal constants mapped to named standard clauses (WA4) after WA0B	wq_std_clause_map	A certificate is not a thing software can ship, so the decidable half of the certification gap is CLAUSE COVERAGE. ACCEPT RULE PRE-DECLARED: every WQ_REFUSED constant in the organ maps to a named clause of a mirrored and content-pinned standard [epa-pfas-2024], a constant with no mapping is REPORTED as UNMAPPED rather than silently omitted, and a clause of the standard with no corresponding gate is listed as UNCOVERED -- the direction nobody audits, because an omission flatters. The partition of constants must be printed and must sum. This rung claims coverage and NEVER claims certification	Organ	1.5 u
Admission contract for a live reading stream (WA5) after WA2	wq_stream_admit	The sota grades real-time telemetry ABSENT and reads it as a hardware gap. Half of it is not: the ADMISSION CONTRACT for a timestamped reading needs no sensor to define, and defining it before hardware exists is the only order in which it can be got right. ACCEPT RULE PRE-DECLARED: a reading older than a declared max-age reads UNOBSERVABLE and never SAFE, a stream with a gap longer than the declared cadence REPORTS the gap with its duration rather than interpolating across it, and a negative-age reading -- a future stamp from clock skew -- is UNOBSERVABLE too and not silently clamped. The max-age is DERIVED from the declared cadence and is never a typed number. This rung ships no sensor and claims none	Organ	1 u

Milestones

Milestone	Rungs	Cumulative
W0 · The model answers to measured data	WA1,WA2	2.5 u
W1 · Depth and the standard	WA3,WA4	5 u
W2 · Live input	WA5	6 u

Ladder verdict. NOT DECLARED. This board names no dated best-in-class or frontier target and no rung roles (sotatarget and rungrole rows on its plan); the ranker labels it NO-LADDER until it does, and until then its rungs climb toward a target nobody has written down. Bars read NO-BAR against the current month 2026-10.

Bar freshness, counted
Class	Count	Note
bars	0	sotabar rows on the plan
fresh	0	the bar month is the current month
attested	0	an older bar with a this-month barscan reporting no newer entry
stale	0	a this-month scan saw a newer month the bar ignores
unattested	0	nobody looked this month
malformed	0	a sotabar row missing a field
duplicate	0	bar ids an earlier row already carries
bars, summed	0	RECONCILED: the parts sum to the total 0
current month	2026-10	the month every bar is judged against

Treatment verdict. NOT DECLARED. No rung on this board declares its three-step treatment (a treat row naming the MCP row, the API route, the primitive it composes, the workflow that composes it and the agent beat that runs it); the ranker labels every rung TREATMENT-UNDECLARED until it does.

Inspect a rung and its prerequisites

Declared nodes 8. Rank input binding: PLAN_MATRIX_BOUND_ONLY. Dependency order is authored. Implementation, acceptance evidence, authority and resource readiness are unverified. No action is recommended or dispatched here.

Plan SHA-256 04065e7d718ff6320febff1118ba8d1733e3f71952ec117384ed4278a6670db2. Target rows 0; role rows 0. Existing risks and release worklog retain their own scope; no node completion is inferred.

Runtime observation coverage: 0 of 8 rungs have current, identity-bound execution observations. Observation may report RED. Accepted completion remains unknown; source presence, qualification and deployment do not establish it.

Use Enter or Space on a rung to inspect its contract. Prerequisite links locate another rung in this list; open its summary to inspect it. Estimates are authored effort, not forecasts.

WA0 — Climate-to-harvester recommendation
Prerequisites: None declared; this does not establish execution eligibility.
Contract: wa_recommend
Acceptance: LANDED and gate-proven 14/14. Four recommender teeth one per climate regime, and the load-bearing one is the NEGATIVE CONTROL -- a climate too dry to feed any class returns NONE rather than the least-bad guess. Listed so dependents carry its momentum
Authored effort: 0. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA0B — Never-poison safety gate
Prerequisites: None declared; this does not establish execution eligibility.
Contract: wq_safety_admit
Acceptance: LANDED and gate-proven 15/15. Five named negative controls one per contaminant class each asserting its OWN refusal reason, plus the positive control that makes them mean anything -- a clean sample must be ADMITTED, because a function that refused every input would otherwise score full marks on every refusal test [@epa-npdwr]. Listed so dependents carry its momentum
Authored effort: 0. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA0C — Design capstone that composes the suite
Prerequisites: None declared; this does not establish execution eligibility.
Contract: wd_design_best_water
Acceptance: LANDED and gate-proven 15/15. One gate imports the harvesting model the quality engine and the design organ together and walks raw condensate through treatment to remineralisation, asserting the three stages are DISTINGUISHABLE -- an implementation that returned the same sample at every stage would pass a suite that only checked the final number. Listed so dependents carry its momentum
Authored effort: 0. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA1 — Modeled yield validated against published field figures
Prerequisites: WA0 (acceptance unverified)
Contract: wa_validate_published
Acceptance: The sota grades field-proven yield at scale as ABSENT and it will stay absent, because we deploy no fleet. The honest form of the gap is VALIDATION. ACCEPT RULE PRE-DECLARED: for each vendor whose rated conditions are stated in a mirrored and content-pinned reference [@watergen] [@source-hydropanel], the modeled litres per day at THOSE conditions is computed and the SIGNED error against the vendor's own published figure is printed per row -- a signed error, never an absolute one, so a model that is consistently optimistic is visible as a bias rather than averaged away. A vendor whose reference states no rated temperature and humidity reports UNWITNESSED as its own third state and does NOT score as agreement. ANTI-VACUITY TOOTH REQUIRED: a deliberately wrong coil temperature must push at least one row outside the declared band, or the comparison has not been shown to be capable of disagreeing
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA2 — Measured water-quality record ingest
Prerequisites: WA0B (acceptance unverified)
Contract: wq_ingest_report
Acceptance: Today every score is of numbers the caller supplied, which is why the sota grades lab-measured analysis ABSENT and the standard Best. ACCEPT RULE PRE-DECLARED: a mirrored public water-quality record parses into the NxWaterSample the engine already takes and scores through the EXISTING wq_taste_score and wq_safety_admit -- one ruler, never a second scorer for ingested data. An analyte the record does not carry is UNKNOWN, and a sample carrying an UNKNOWN regulated analyte makes the safety verdict ABSTAIN rather than ADMIT, because an axis that cannot see must abstain and not acquit [@who-gdwq]. The partition must be printed and must sum: records read equals ingested plus refused plus unparseable. NEG-CONTROL: a record with an over-limit analyte is refused NAMING that analyte, and a clean record is admitted
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA3 — Sorbent uptake as a curve not a threshold
Prerequisites: WA0 (acceptance unverified)
Contract: wa_sorbent_uptake
Acceptance: wa_class_min_rh is one integer per class, which can express PERMITTED but cannot express HOW MUCH, and that is why the sota grades sorbent low-humidity Part against SOURCE Hydropanel Best. ACCEPT RULE PRE-DECLARED: uptake is monotone non-decreasing in relative humidity across the whole tested range, it DISCRIMINATES two named sorbent materials at the SAME humidity -- a curve that returns one shape for every material has measured nothing [@kim2025-amphiphilic] -- and the existing wa_class_works and wa_recommend teeth stay byte-identical afterwards, so the depth is proven ADDITIVE rather than a rewrite that moved the answers
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA4 — Refusal constants mapped to named standard clauses
Prerequisites: WA0B (acceptance unverified)
Contract: wq_std_clause_map
Acceptance: A certificate is not a thing software can ship, so the decidable half of the certification gap is CLAUSE COVERAGE. ACCEPT RULE PRE-DECLARED: every WQ_REFUSED constant in the organ maps to a named clause of a mirrored and content-pinned standard [@epa-pfas-2024], a constant with no mapping is REPORTED as UNMAPPED rather than silently omitted, and a clause of the standard with no corresponding gate is listed as UNCOVERED -- the direction nobody audits, because an omission flatters. The partition of constants must be printed and must sum. This rung claims coverage and NEVER claims certification
Authored effort: 1.5. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.
WA5 — Admission contract for a live reading stream
Prerequisites: WA2 (acceptance unverified)
Contract: wq_stream_admit
Acceptance: The sota grades real-time telemetry ABSENT and reads it as a hardware gap. Half of it is not: the ADMISSION CONTRACT for a timestamped reading needs no sensor to define, and defining it before hardware exists is the only order in which it can be got right. ACCEPT RULE PRE-DECLARED: a reading older than a declared max-age reads UNOBSERVABLE and never SAFE, a stream with a gap longer than the declared cadence REPORTS the gap with its duration rather than interpolating across it, and a negative-age reading -- a future stamp from clock skew -- is UNOBSERVABLE too and not silently clamped. The max-age is DERIVED from the declared cadence and is never a typed number. This rung ships no sensor and claims none
Authored effort: 1. Executor kind: Organ. Responsible, accountable and verifier not established. Inspect retained worklog.
Runtime observation: UNKNOWN; outcome UNKNOWN; observed epoch -1. Declared owner: . Receipt , SHA-256 . Identity-bound observation is not accepted completion. Qualification and deployment remain unverified by this projection.

Learning and practice paths

No structured learning path is declared for this plan. Existing research, roadmap and worklog remain available above.

Rung dossiers — every record that bears on a rung, in one place

Joined by symbol, never typed. Each dossier is assembled on every publish from the rows this board already carries: the plan's rung row, the matrix row that carries its contract symbol (its status re-measured by the one symbol ruler), the gate the organ declares and the verdict the evidence referee recorded on its last run, the reference marks in the rung text, the design rows that name the rung and the work log filed against it. The capability matrix links each cell to its dossier and each dossier links back to its cell and forward to the rungs it stands on; a dependency naming no rung on this plan is written as UNRESOLVED rather than linked. Absences are named: NO-MATRIX-ROW, NO-GATE-DECLARED, UNRECORDED.

Gate verdicts: UNSTAMPED — no per-gate verdict artifact is published for this domain yet (nx_swcompare_evidence water writes it beside its stamp); every gate below reads UNRECORDED until it runs.

Gate declarations: knowledge/organ_gate.conf read from this working directory (1 rows) and from the estate root (196 rows); rows in both counted once: 0; declarations read: 197. Every declaration in either copy counts, so a stub beside the publisher can never hide the root.

WA0 Climate-to-harvester recommendation · cell PRESENT · gate NO-GATE-DECLARED

Contract	`wa_recommend` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED and gate-proven 14/14. Four recommender teeth one per climate regime, and the load-bearing one is the NEGATIVE CONTROL -- a climate too dry to feed any class returns NONE rather than the least-bad guess. Listed so dependents carry its momentum
Matrix cell	Climate-to-harvester recommendation · `runtime/nx_water_awg.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_awg`; declare `nx_water_awg TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA0B Never-poison safety gate · cell PRESENT · gate NO-GATE-DECLARED

Contract	`wq_safety_admit` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED and gate-proven 15/15. Five named negative controls one per contaminant class each asserting its OWN refusal reason, plus the positive control that makes them mean anything -- a clean sample must be ADMITTED, because a function that refused every input would otherwise score full marks on every refusal test [epa-npdwr]. Listed so dependents carry its momentum
Matrix cell	Never-poison safety gate · `runtime/nx_water_quality.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_quality`; declare `nx_water_quality TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA0C Design capstone that composes the suite · cell PRESENT · gate NO-GATE-DECLARED

Contract	`wd_design_best_water` · executor Organ · 0 u · stands on nothing
Definition of done	LANDED and gate-proven 15/15. One gate imports the harvesting model the quality engine and the design organ together and walks raw condensate through treatment to remineralisation, asserting the three stages are DISTINGUISHABLE -- an implementation that returned the same sample at every stage would pass a suite that only checked the final number. Listed so dependents carry its momentum
Matrix cell	Composable with the food-science suite · `runtime/nx_water_design.nx` · PRESENT decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_design`; declare `nx_water_design TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA1 Modeled yield validated against published field figures · cell WATCHING · gate NO-GATE-DECLARED

Contract	`wa_validate_published` · executor Organ · 1 u · stands on WA0
Definition of done	The sota grades field-proven yield at scale as ABSENT and it will stay absent, because we deploy no fleet. The honest form of the gap is VALIDATION. ACCEPT RULE PRE-DECLARED: for each vendor whose rated conditions are stated in a mirrored and content-pinned reference [watergen] [source-hydropanel], the modeled litres per day at THOSE conditions is computed and the SIGNED error against the vendor's own published figure is printed per row -- a signed error, never an absolute one, so a model that is consistently optimistic is visible as a bias rather than averaged away. A vendor whose reference states no rated temperature and humidity reports UNWITNESSED as its own third state and does NOT score as agreement. ANTI-VACUITY TOOTH REQUIRED: a deliberately wrong coil temperature must push at least one row outside the declared band, or the comparison has not been shown to be capable of disagreeing
Matrix cell	Field-proven yield at scale · `runtime/nx_water_awg.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_awg`; declare `nx_water_awg TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA2 Measured water-quality record ingest · cell WATCHING · gate NO-GATE-DECLARED

Contract	`wq_ingest_report` · executor Organ · 1.5 u · stands on WA0B
Definition of done	Today every score is of numbers the caller supplied, which is why the sota grades lab-measured analysis ABSENT and the standard Best. ACCEPT RULE PRE-DECLARED: a mirrored public water-quality record parses into the NxWaterSample the engine already takes and scores through the EXISTING wq_taste_score and wq_safety_admit -- one ruler, never a second scorer for ingested data. An analyte the record does not carry is UNKNOWN, and a sample carrying an UNKNOWN regulated analyte makes the safety verdict ABSTAIN rather than ADMIT, because an axis that cannot see must abstain and not acquit [who-gdwq]. The partition must be printed and must sum: records read equals ingested plus refused plus unparseable. NEG-CONTROL: a record with an over-limit analyte is refused NAMING that analyte, and a clean record is admitted
Matrix cell	Lab-measured water analysis · `runtime/nx_water_quality.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_quality`; declare `nx_water_quality TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA3 Sorbent uptake as a curve not a threshold · cell WATCHING · gate NO-GATE-DECLARED

Contract	`wa_sorbent_uptake` · executor Organ · 1 u · stands on WA0
Definition of done	wa_class_min_rh is one integer per class, which can express PERMITTED but cannot express HOW MUCH, and that is why the sota grades sorbent low-humidity Part against SOURCE Hydropanel Best. ACCEPT RULE PRE-DECLARED: uptake is monotone non-decreasing in relative humidity across the whole tested range, it DISCRIMINATES two named sorbent materials at the SAME humidity -- a curve that returns one shape for every material has measured nothing [kim2025-amphiphilic] -- and the existing wa_class_works and wa_recommend teeth stay byte-identical afterwards, so the depth is proven ADDITIVE rather than a rewrite that moved the answers
Matrix cell	Sorbent low-humidity capability · `runtime/nx_water_awg.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_awg`; declare `nx_water_awg TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA4 Refusal constants mapped to named standard clauses · cell WATCHING · gate NO-GATE-DECLARED

Contract	`wq_std_clause_map` · executor Organ · 1.5 u · stands on WA0B
Definition of done	A certificate is not a thing software can ship, so the decidable half of the certification gap is CLAUSE COVERAGE. ACCEPT RULE PRE-DECLARED: every WQ_REFUSED constant in the organ maps to a named clause of a mirrored and content-pinned standard [epa-pfas-2024], a constant with no mapping is REPORTED as UNMAPPED rather than silently omitted, and a clause of the standard with no corresponding gate is listed as UNCOVERED -- the direction nobody audits, because an omission flatters. The partition of constants must be printed and must sum. This rung claims coverage and NEVER claims certification
Matrix cell	NSF or ANSI certification · `runtime/nx_water_quality.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_quality`; declare `nx_water_quality TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

WA5 Admission contract for a live reading stream · cell WATCHING · gate NO-GATE-DECLARED

Contract	`wq_stream_admit` · executor Organ · 1 u · stands on WA2
Definition of done	The sota grades real-time telemetry ABSENT and reads it as a hardware gap. Half of it is not: the ADMISSION CONTRACT for a timestamped reading needs no sensor to define, and defining it before hardware exists is the only order in which it can be got right. ACCEPT RULE PRE-DECLARED: a reading older than a declared max-age reads UNOBSERVABLE and never SAFE, a stream with a gap longer than the declared cadence REPORTS the gap with its duration rather than interpolating across it, and a negative-age reading -- a future stamp from clock skew -- is UNOBSERVABLE too and not silently clamped. The max-age is DERIVED from the declared cadence and is never a typed number. This rung ships no sensor and claims none
Matrix cell	Real-time sensor telemetry · `runtime/nx_water_quality.nx` · WATCHING decl
Gate	NO-GATE-DECLARED — `knowledge/organ_gate.conf` carries no row for `nx_water_quality`; declare `nx_water_quality TAB <gate>` and the ship loop proves it and the verdict lands here.
Work log	no log row names this rung yet

dossiers=8 cells_joined=8 no_matrix_row=0 (sum 8) gates_recorded=0 gates_unrecorded=0 no_gate_declared=8 (sum 8 of cells_joined) evgates=UNSTAMPED

Capability comparisons

Compare the field, search individual capabilities and open their source and adoption evidence. Documented presence does not establish comparative quality.

Position map — centrality and distinctiveness

The four-quadrant map the field uses for brand strategy (Dawar and Bagga, HBR June 2015), re-derived from this matrix on every publish. Centrality is the share of the category's feature mass a player covers, each feature weighted by how many hold it; distinctiveness is the average lead over each rival on the rows the player holds; breadth is the depth-weighted share of the whole matrix (the bubble); depth is how deeply the rows held are held; momentum is the day-over-day move off the spine (green rising, red falling, grey until day two); the dashed path runs first day → previous day → today. Dividers are the category means. Axes are fitted to the field of play, so read the tick numerals, not the frame. Rival marks are documented presence, so a rival's position reads the record, never its quality. The picture grades its own readability below; the table beside it is the same data for a screen reader or a second method.

Rows. 8 feature rows mapped of 8 matrix rows; 0 substrate rows (kernels, bounds and instruments the plan's rungrole marks substrate) are the estate's scaffolding, not category features, and stay off the map

Views. 2D cut: centrality, distinctiveness · 3D cube: centrality, distinctiveness, breadth · axes registered: centrality, distinctiveness, breadth, depth, momentum · a board picks its own in knowledge/compare/water.cdmap (cut|x|y, cube|x|y|z); absent = the HBR defaults

bubble area = breadth · ring = momentum (green rising, red falling, grey until day two) · dashed = category means · axes fitted to the field of play: centrality 200–1000, distinctiveness 100–800 of 0–1000 permil (the full range put every player in one corner)
readability of the 2D cut, self-graded by the layout ruler: label overlaps 0 · labels over marks 0 · off-canvas 0 · unresolved labels 0 · mark overlaps 0 (a fact of the data: two players that close are that close) · data spread 754 permil of the plot · quadrant words unseated 0
text contrast, measured with wcag2-ratio (floors from contrast.conf), light theme: labels 16.24 (floor 4.50) · notes 16.24 (floor 4.50) · quadrant words 5.89 (floor 4.50) · tick numerals 5.89 (floor 4.50) · axis titles 5.59 (floor 4.50) · dark theme: labels 13.78 (floor 4.50) · quadrant words 5.55 (floor 4.50) · axis titles 5.97 (floor 4.50) · dark classes under their floor 0 (one figure serves both themes: a dark shortfall is a token to fix, never a class to hide) · classes refused under their floor 0 (a refused class is not drawn; the scale classes are measured, never hidden) · export: SVG PNG (receipt, rendered by the estate's own rasteriser from this page)

axes fitted: centrality 200–1000 · distinctiveness 100–800 · breadth 300–500 permil · farther bubbles are painted first, nearer ones over them · the floor shadow is each bubble's (x, y) at height 0
readability of the 3D cube, self-graded by the layout ruler: label overlaps 0 · labels over marks 0 · off-canvas 2 · unresolved labels 0 · mark overlaps 0 (a fact of the data: two players that close are that close) · data spread 495 permil of the plot

Nishi Watergen SOURCE-Hydropanel APEC-RO WHO-EPA-standard
10 panels over 5 registered axes, every pair once (the lower axis on x, the higher on y) · each panel fitted to its own field of play, first and last tick numerals shown · no labels in a small cut, the legend names the colours; the grade below reports the mark terms only (MM summed over the panels, spread averaged)
readability of the small multiples, self-graded by the layout ruler: label overlaps 0 · labels over marks 0 · off-canvas 0 · unresolved labels 0 · mark overlaps 7 (a fact of the data: two players that close are that close) · data spread 545 permil of the plot

Player	Quadrant	centrality	distinctiveness	breadth	depth	momentum	Rows held	Days on spine	First seen
Nishi	Unconventional	318	722	333	888	0 since 2026-09-24	3	11	2026-09-15
Watergen	Mainstream	863	266	375	600	0 since 2026-09-24	5	11	2026-09-15
SOURCE-Hydropanel	Mainstream	909	333	458	611	0 since 2026-09-24	6	11	2026-09-15
APEC-RO	Mainstream	863	183	333	533	0 since 2026-09-24	5	11	2026-09-15
WHO-EPA-standard	Unconventional	590	472	333	888	0 since 2026-09-24	3	11	2026-09-15

Day	Matrix rows reviewed	Players recorded
2026-09-15	8	5
2026-09-16	8	5
2026-09-17	8	5
2026-09-18	8	5
2026-09-19	8	5
2026-09-20	8	5
2026-09-21	8	5
2026-09-22	8	5
2026-09-23	8	5
2026-09-24	8	5
2026-10-08	8	5

How to read this evidence. A Nishi source mark records whether the named symbol is declared in the selected source file. It does not establish working behavior or accepted delivery. A watching tag names an expected source entry; discovering it updates source coverage only. A dark tag means the file exists but the named symbol was not found: inspect the source and contract before assuming the capability is missing or renaming code. Peer codes are authored assessments; unassessed means no comparison is established, not absence. Qualification, deployment identity, runtime observations and acceptance must carry their own evidence. Adoption records the catalog's observed path through source, build, promotion, registration and invocation; libraries use importer evidence and gates use declared execution surfaces. A full catalog path does not establish quality, freshness or a successful user outcome. Partial and unknown states retain their stated next action. Census stamps: importers asof 1790637268, gate census asof 1790104131 (unix seconds; -1 = census absent).

Capability matrix — measured against source

◉ authored stronger claim; source present● present◐ partial○ absent? unassessed; excluded from relative comparison · click any capability for its evidence

Peer cell coverage: authored assessed 32, unassessed 0. Authored assessments are not verified outcomes.

Capability	Nishi	Watergen	SOURCE-Hydropanel	APEC-RO	WHO-EPA-standard
Climate-to-harvester recommendation Source present; authored stronger claim: `wa_recommend` in `runtime/nx_water_awg.nx`, verified at emit. LANDED. Picks the harvesting class for a climate and returns NONE when no class can feed -- refusal over the least-bad guess. No product in the field chooses ACROSS technologies so every rival is absent. Listed so dependents carry its momentum Adoption: `LIB-WIRED importers=5 nonval=2` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Watergen, SOURCE-Hydropanel, APEC-RO, WHO-EPA-standard; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
Never-poison safety gate Measured: `wq_safety_admit` exists in `runtime/nx_water_quality.nx`, verified at emit. LANDED. Structurally refuses over-limit lead arsenic nitrate PFAS bacteria THM with a per-contaminant reason [epa-npdwr]. Filters remove physically and the standard defines the limits so the rivals are present here Adoption: `LIB-WIRED importers=5 nonval=2` — fully adopted (top of its ladder). dossier Pros Implementing source symbol found; runtime acceptance separate; ahead of Watergen, SOURCE-Hydropanel; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	●	◐	◐	●	●
Composable with the food-science suite Source present; authored stronger claim: `wd_design_best_water` in `runtime/nx_water_design.nx`, verified at emit. LANDED. One design capstone composes the harvesting model the quality gate and the chemistry organs; separately-licensed appliances cannot be composed into one gate at all Adoption: `LIB-WIRED importers=3 nonval=1` — fully adopted (top of its ladder). dossier Pros Source symbol found; stronger capability claim is authored; ahead of Watergen, SOURCE-Hydropanel, APEC-RO, WHO-EPA-standard; fully adopted on the estate ladderCons none on the measured axes (rival marks are documented presence, not depth)	◉	○	○	○	○
Field-proven yield at scale DARK — `runtime/nx_water_awg.nx` EXISTS but does not declare `wa_validate_published`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. OPEN -- rung WA1. We will never deploy a fleet so the honest form of this gap is VALIDATION not deployment. The contract is a comparison of the modeled daily yield against each vendor's own published rated figure [watergen] [source-hydropanel] with the signed error printed per row dossierwatching wa_validate_published Pros none measured yetCons behind Watergen (leads), SOURCE-Hydropanel, APEC-RO; open contract; expected source declaration not established	○	◉	●	●	○
Lab-measured water analysis DARK — `runtime/nx_water_quality.nx` EXISTS but does not declare `wq_ingest_report`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. OPEN -- rung WA2. Today the engine scores values a caller supplies. The contract is ingesting a real published water-quality record so the scores are of MEASURED water; accredited labs and the standard remain Best because we still run no assay [who-gdwq] dossierwatching wq_ingest_report Pros none measured yetCons behind Watergen, SOURCE-Hydropanel, APEC-RO, WHO-EPA-standard (leads); open contract; expected source declaration not established	○	◐	◐	◐	◉
Sorbent low-humidity capability DARK — `runtime/nx_water_awg.nx` EXISTS but does not declare `wa_sorbent_uptake`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. OPEN -- rung WA3. wa_class_min_rh is a single threshold per class which cannot express an isotherm. The contract is a real uptake curve so the desert case is modeled rather than merely permitted [kim2025-amphiphilic] dossierwatching wa_sorbent_uptake Pros none measured yetCons behind SOURCE-Hydropanel (leads); open contract; expected source declaration not established	○	○	◉	○	○
NSF or ANSI certification DARK — `runtime/nx_water_quality.nx` EXISTS but does not declare `wq_std_clause_map`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. OPEN -- rung WA4. A certificate is not something software can ship so the decidable half is CLAUSE COVERAGE: every refusal constant mapped to a named clause of a mirrored standard with the uncovered clauses listed rather than omitted [epa-pfas-2024] dossierwatching wq_std_clause_map Pros none measured yetCons behind Watergen, SOURCE-Hydropanel, APEC-RO, WHO-EPA-standard (leads); open contract; expected source declaration not established	○	●	●	●	◉
Real-time sensor telemetry DARK — `runtime/nx_water_quality.nx` EXISTS but does not declare `wq_stream_admit`: something shipped at this path under another name, and this mark would read open forever. Repoint the contract to the real entry point (the plan rung AND this row) or rename the function; the flip follows on the next beat, and the `comparewatch-` plane row reads DARK until then. OPEN -- rung WA5. No hardware is required to define the ADMISSION contract for a reading stream. The load-bearing half is that a stale or gapped stream must read UNOBSERVABLE rather than SAFE rival marks uncited — the Yes, Best or Part codes on this row are an observation read with no reference mark behind them dossierwatching wq_stream_admit Pros none measured yetCons behind Watergen, SOURCE-Hydropanel, APEC-RO; open contract; expected source declaration not established	○	●	●	◐	○

Rival-claim provenance. Every Yes, Best or Part mark in a rival column is a claim about someone else's product. A cited row carries a reference mark that resolves to a pinned mirror (the refs gate measures that); an uncited row is an observation read and is badged in its evidence until a reference lands. The badge is the receipt, never the proof: a mark proves a mirror exists, not that the mirror supports the code. The rows are counted below.

Rival claims, counted
Class	Count	Note
rows with rival marks	6	rows carrying a Yes, Best or Part in a rival column
cited	5	a reference mark resolves to a pinned mirror
uncited	1	an observation read, badged until a reference lands
rows, summed	6	RECONCILED: the parts sum to the total 6

Delivery and evidence

Inspect risks, technical debt, rendered observations, experiments and references. Read scope and limitations alongside every result.

Risk register

Risk	Likelihood x impact	Mitigation
A validation rung that compares against vendor marketing figures dresses a marketing number as evidence	likely x medium	WA1 accepts a figure only from a mirrored and content-pinned reference, prints the SIGNED error rather than a pass or fail, and reports UNWITNESSED for any vendor whose rated conditions are not stated. It publishes a comparison, never a verdict that we match the field.
Ingesting measured data makes the engine look validated when only its inputs changed	possible x high	WA2 changes where the numbers come from and nothing about the scoring; the row it closes is lab-measured ANALYSIS and the sota keeps accredited labs at Best because we still run no assay. The claim the rung licenses is stated in the rung and is narrower than the row it touches.
A clause map is read as a certification claim	possible x high	WA4 says coverage and refuses the word certification anywhere in its output, and it must list UNCOVERED clauses -- a coverage map that could only ever report good news is not a map.

On these two registers. Rows are declared in the domain's plan file and carry the debt id, which is the join key back to the sovereign debt plane — that plane, not this page, is the authority on state. Reconciling them automatically (the regen reading the plane and refreshing these rows) is a named, owed rung; until it lands, treat an id here as a pointer to look up, not a status to trust.

SOTA field — 8 peers, 26 axes, measured bindings

Nishi water-system design engine vs Watergen / SOURCE Hydropanel / Kara Water / ZeroWater / Berkey / Brita / APEC RO / the WHO-EPA standard. Quantitative axes measured or modeled (Nishi by gates; competitors researcher-sourced); grade axes B=Best Y=Yes ~=Part n=No. Honest: Nishi DESIGNS + SCORES the system sovereignly but builds no hardware and runs no lab assay.

How the field is scored. A state-of-the-art comparison across the FULL competitor field: quantitative axes carry measured or published numbers (Nishi’s column is measured on this estate, competitor columns are researcher-sourced and cited in the note); grade axes carry Best / Yes / Part / No for every peer. A Nishi grade may be bound to an organ and a symbol and is then re-measured on every publish by the one symbol ruler the capability matrix above uses; an over-claim refuses to publish.

field, strip order:WatergenSOURCE HydropanelKara WaterZeroWaterBerkeyBritaAPEC ROWHO-EPA standard·BestYesPartNo

Measured bindings. 3 of the Nishi grades on this page are bound to an organ and a symbol and re-measured on every publish by the one symbol ruler; the drift classes are counted below and their sum is re-derived beside the bound total.

Measured bindings, counted
Class	Count	Note
consistent	0	the grade and the symbol agree
stale-under	0	the grade says No while the organ carries the symbol; published with the badge, named by the gate until raised
unbacked	0	a claim over an absent symbol refuses to publish
unresolved	0
unreadable	0
unserved	3	the organ carries the symbol in source while its binary is not on the serving surface: built nowhere, invisible to symbol presence; published with the badge, never refused
partition sum	3	RECONCILED: the parts sum to the total 3

Scope (quantitative)

Harvester technology classes modeled

Nishi models condensation + sorbent-MOF + passive-capillary + fog + radiative and picks per climate; each product embodies one; filters do not harvest air

Watergen 1 condensationSOURCE Hydropanel 1 sorbent-solarKara Water 1 condensationZeroWater 0 filter onlyBerkey 0 filter onlyBrita 0 filter onlyAPEC RO 0 RO filterWHO-EPA standard n/a standard

5 classes

Nishi, measured

Modeled specific energy Wh per L

Nishi first-order psychrometric model; Watergen published about 0.3 to 0.4 kWh/L [watergen]; SOURCE is solar-driven passive

Watergen 300 to 400SOURCE Hydropanel solar passiveKara Water poweredZeroWater n/a filterBerkey n/a gravityBrita n/a pitcherAPEC RO n/a RO pumpWHO-EPA standard n/a

about 208 humid

Nishi, measured

Regulated contaminants gated

Nishi refuses over-limit lead arsenic nitrate PFAS bacteria THM; products physically remove; WHO-EPA define the limits [epa-npdwr]

Watergen filter setSOURCE Hydropanel filter setKara Water UV plus filterZeroWater ion exchangeBerkey carbon ceramicBrita carbonAPEC RO RO membraneWHO-EPA standard the standard

6 limits

Nishi, measured

Taste palatability bands scored

Nishi scores WHO TDS palatability plus mineral balance plus pH; a TDS meter shows one number

Watergen displaySOURCE Hydropanel displayKara Water displayZeroWater reads near 0 TDSBerkey noneBrita noneAPEC RO optionalWHO-EPA standard the source

6 WHO bands

Nishi, measured

Atmospheric water harvesting

Water-from-air harvesting model

Nishi models the full harvesting landscape and its physics; Watergen/SOURCE/Kara each build one class; filters do not harvest air at all. EVIDENCE (nx_water_awg_test, 14/14 GREEN, MEASURED 2026-08-14): saturation vapour density, absolute humidity, extractable water at a given coil temperature, daily yield and specific energy are each asserted to an exact integer, and the technology envelopes are tested in PAIRS -- sorbent works at 20 percent RH while refrigeration at the SAME humidity does not -- so the model is shown to discriminate between classes rather than to score them all alike. HONEST SCOPE: this gate proves the MODEL, not a built appliance; the three rivals ship hardware and we do not

BEST

Multi-technology envelope

Nishi covers condensation AND sorbent-MOF AND the UPenn passive-capillary material [kim2025-amphiphilic] AND fog in one model; each product is single-class

YES

Psychrometric yield from climate

Nishi computes extractable g/m3 and L/day from temperature and RH; AWG vendors quote yield at rated conditions

YES

Climate-to-harvester recommendation

Nishi picks the best harvesting technology for a climate (desert to sorbent, moderate to passive capillary, humid to condensation) - no product chooses across technologies. EVIDENCE (nx_water_awg_test, 14/14 GREEN, MEASURED 2026-08-14): four recommender teeth, one per regime, and the load-bearing one is the NEGATIVE CONTROL -- a climate too dry to feed any class returns NONE rather than the least-bad guess. A recommender that always answered would pass the other three; refusal-over-fabrication is what makes the answer worth reading

bound to runtime/nx_water_awg.nx:wa_recommend · measured PRESENT (decl) · UNSERVED — the hand grade claims the capability and the organ carries the symbol in SOURCE, but its binary is not on the serving surface: built nowhere -- this cell over-reports what RUNS and stays flagged until the organ is promoted

BEST

Sorbent low-humidity capability

SOURCE hydropanels and MOF sorbents are Best in dry desert air [source-hydropanel]; Nishi models the envelope but builds no sorbent

PART

Taste, health, and safety

TDS taste band scoring

Nishi scores the WHO palatability bands [who-gdwq] and a composite; products display a TDS number at most

YES

Mineral health scoring Ca Mg

Nishi scores the WHO desirable calcium and magnesium; alkaline and mineral AWGs add minerals; ZeroWater strips them out [zerowater]

YES

Never-poison safety gate

Nishi STRUCTURALLY refuses over-limit lead arsenic nitrate PFAS bacteria THM; filters remove physically; the standard defines. EVIDENCE (nx_water_quality_test, 15/15 GREEN, MEASURED 2026-08-14): five named negative controls, one per contaminant class -- lead above 15 ppb, PFAS above 4 ppt per EPA 2024, any bacteria, arsenic and nitrate above limit -- each asserting its OWN refusal reason rather than merely that something was refused. The tooth that makes those five mean anything is the POSITIVE control: a clean sample must be ADMITTED, because an admit function that refused every input would otherwise score full marks on every refusal test

bound to runtime/nx_water_quality.nx:wq_safety_admit · measured PRESENT (decl) · UNSERVED — the hand grade claims the capability and the organ carries the symbol in SOURCE, but its binary is not on the serving surface: built nowhere -- this cell over-reports what RUNS and stays flagged until the organ is promoted

BEST

PFAS 2024 limit aware

Nishi gates the EPA 2024 PFOA and PFOS 4 ppt limit; RO and some media reduce PFAS; WHO-EPA set the number [epa-pfas-2024]

YES

Remineralization design generative

Nishi doses calcium magnesium and bicarbonate to hit the taste and health optimum by design; mineral AWGs add a fixed blend [kara-water]

YES

Condensate-to-potable treatment chain

Nishi models UV disinfection plus filtration plus remineralization [apec-ro]; AWG appliances integrate it; the standard requires it

YES

Sovereignty and method

Sovereign zero-license design tool

Nishi is a free sovereign design and scoring engine; the rest are paid appliances or filters. EVIDENCE (MEASURED 2026-08-14): the three organs behind this domain -- nx_water_awg, nx_water_quality, nx_water_design -- were compiled and their gates RUN on our own toolchain (nx_cc to nxasm to ELF, no LLVM, GCC or libc beneath) and returned GREEN 14/14, 15/15 and 15/15. HONEST NOTE ON THIS ROW'S HISTORY: until today those organs had never been BUILT AT ALL. The domain was claiming Best on the strength of source nobody had compiled, which is exactly the state the evidence-counted liar-kill exists to refuse, and it refused it

BEST

Integer-deterministic model

Nishi no-float deterministic scoring reproduces bit-for-bit; unique in this field. EVIDENCE (MEASURED 2026-08-14): the determinism is not asserted, it is what makes the gates expressible -- 44 teeth across the three organs assert EXACT integer equalities (saturation density 304 and 173, absolute humidity 212, extractable 144, daily yield 34560 ml, specific energy 208 Wh/L, taste scores exactly 100 and exactly 15). A float model cannot be tested this way at all; the fact that these are equality assertions rather than tolerance bands IS the evidence

BEST

Runs fully offline

Nishi is a local model; passive filters need no power; connected AWGs can phone home

YES

Composable with the food-science suite

Nishi water design composes the same steam psychrometrics and chemistry organs as the other 42 food-science capabilities. EVIDENCE (nx_water_design_test, 15/15 GREEN, MEASURED 2026-08-14): the composition is exercised, not described -- one gate imports nx_water_awg, nx_water_quality and nx_water_design together and walks raw condensate through treatment to remineralisation, asserting the three stages are DISTINGUISHABLE (raw refused for bacteria, treated potable but NOT healthful and still flat, designed both healthful and excellent) and that the harvester choice and daily yield agree with the awg model. An implementation that returned the same sample at every stage would fail here while passing a suite that only checked the final number

bound to runtime/nx_water_design.nx:wd_design_best_water · measured PRESENT (decl) · UNSERVED — the hand grade claims the capability and the organ carries the symbol in SOURCE, but its binary is not on the serving surface: built nowhere -- this cell over-reports what RUNS and stays flagged until the organ is promoted

BEST

Where Nishi is honestly behind

Physically produces water

Nishi ABSENT: it DESIGNS and scores, it builds no hardware and makes no water

Installed appliance or hardware

Nishi ABSENT: there is no physical product

Filtration or membrane media

Nishi ABSENT: it models treatment but owns no filter or membrane

Lab-measured water analysis

Nishi ABSENT: it computes from supplied values and runs no assay; accredited labs are Best

NSF or ANSI certification

Nishi ABSENT: it is not a certified product

Real-time sensor telemetry

Nishi ABSENT: no live sensors or monitoring

Field-proven yield at scale

Nishi ABSENT: it is a model, not a deployed fleet

8 competitors26 axes4 quantitativeNishi best 6 · absent 7bound 3 · stale-under 0 · unbacked 0 · unserved 3

Honest verdict. Measured across the full field. Where Nishi is under SOTA, that is filed work with an owner, never ‘by design’ (operator law: less-than-SOTA is never design).

Honest verdict. The coverage above records source-symbol presence. Stronger comparisons are authored claims, not measured runtime superiority or accepted outcomes. A watch contract names an expected source symbol; finding it changes source discovery only. Qualification, deployment and runtime acceptance require their own evidence.

Person · product · place — not yet measured for this domain

Every compare carries this layer. Declare knowledge/compare/water.ppp (rows surface|nishi or c1..c4|label|url|connect naming OUR live surface and each rival's front door), run nx_ppp_probe domain water, and this section fills itself on the next beat: the same ruler on both sides — privacy and CX (third-party hosts, tracker classes, cookies, security headers), design and longevity (design hygiene, computed WCAG contrast, render-blocking resources, unsized media, script weight, theme and motion queries), findability (landmarks, skip link, on-site search, breadcrumb, headings, internal links).

The field — discovered, not chosen

Rows written by nx_field_discover from water.seeds: the industry's own lists (Wikipedia wikitext, GitHub topics, awesome lists) read mechanically, every candidate counted across seeds. The matrix columns above are a SEAT'S pick; this band is the population they were picked from, and the stats line measures one against the other. A rival here is a lead, never a verdict — it earns a column when its capabilities are read and pinned.

Discovery receipt. One row per counter the run wrote, in the receipt's own order; the rival rows follow in their own table.

Field discovery, counted
Class	Count	Note
seeds	4
scanned	4
fetched	4
reused	0
failed	0
named	4
candidates	4
mentions	4
capped	0

Rank	Rival	Seeds	Mentions	First seed	Kind	Link
1	Watergen	1	1	col0-watergen	named	https://www.watergen.com/
2	SOURCE-Hydropanel	1	1	col1-source	named	https://arena.gov.au/projects/source-hydropanel-demonstration/
3	APEC-RO	1	1	col2-apec	named	https://www.apecwater.com/
4	WHO-EPA-standard	1	1	col3-who	named	https://www.who.int/publications/i/item/9789240045064

Exclusive and shared capabilities

One capability is graded on one board; every other board references it. The fleet duplicate census (nx_rivalcensus dupes, on the fleetdupes clock row) reads every board's matrix and names each capability graded as a row on two or more boards, with a home derived from evidence: rung when exactly one board's plan contracts the symbol (that board is starred), gates when no plan contracts it and exactly one board's executed gate map runs the organ's gate (that board is plussed), ambiguous when several boards carry the deciding evidence (a decision for their seats), and none when no evidence decides it, with the reason beside it: unmapped (no board of the duplicate declares a gates map; declare the maps) or unproven (a map exists and runs none of the organ's gates; gate the organ on its proving board, or declare it a shared library by policy). A home elsewhere means this board should carry an @xrow reference to the home board's row rather than a graded row of its own; a home here means the other boards should. The table is this board's slice of one estate-wide census: the shape is shared, the rows are the board's.

This board grades 8 capabilities: 8 exclusive to it and 0 shared with other boards. Shared rows on this page: 0 = home here 0 + home elsewhere 0 + ambiguous 0 + undecided 0 + refused 0 (partition 0); undecided = unmapped 0 + unproven 0 + unevidenced 0; homonyms 0; since the previous census: new 0, resolved 0. Census FRESH age_s=6182.

Gaps from the record — what the estate does that no board carries

The record census (nx_goalmap record) reads the invoked-tool population and every plan queue row and files each organ or directive that NO matrix, plan or gates row names. A row here is a callout the boards missed: adjudicate it onto a board or declare it infrastructure. Census state STALE (age 1395287 s), sources read 4 of 7 declared — a BLIND census is a FLOOR: unread sources can only add rows.

kind	name	board	source	evidence

rows shown 0|this board's directives 0|estate-wide un-boarded organs 452 (listed in full on /compare/ecosystem)|census rows 639|malformed 0 (counted, never rendered)

References

Beyond a link list. Every reference below resolves twice — the publisher's copy and, where banked, the estate's own non-rottable library mirror with a content pin — and carries its evidence class plus the exact claim on this page it grounds. Keyed marks like [key] in the matrix notes jump here. A dash means honestly absent, never assumed.

[who-gdwq] World Health Organization. Guidelines for drinking-water quality: fourth edition incorporating the first and second addenda. Geneva, WHO, 2022 (ISBN 9789240045064). Publication page mirrored; the TDS palatability bands, calcium and magnesium desirability and the chemical guideline values are in the volume it links. publisher · read in our library knowledge/fetched/cmp_water_who-gdwq.html · pin he64eba4d78abab61ba9c3b33e26994f512391dfc711258ec0e5eb774bbd8f4af · accessed 2026-08-18 · published-standardGrounds: The WHO half of the WHO-EPA standard column: TDS taste band scoring (the 6 WHO palatability bands the note names), Mineral health scoring Ca Mg (WHO desirable calcium and magnesium) and Taste palatability bands scored -- Nishi's scoring is graded against the WHO guideline text, and the standard is coded Best on Lab-measured water analysis and NSF or ANSI certification.
[epa-npdwr] US Environmental Protection Agency. National Primary Drinking Water Regulations -- the legally enforceable MCLs, MCLGs and treatment techniques for microorganisms, disinfection byproducts, inorganic and organic chemicals (lead action level 15 ppb, arsenic 10 ppb, nitrate 10 mg/L as N, total trihalomethanes 80 ppb). Accessed August 2026. publisher · read in our library knowledge/fetched/cmp_water_epa-npdwr.html · pin h801f35663ab7bbcba7bb3adeb96abee8134f97220c5b0df1b907539df4d2a80b · accessed 2026-08-18 · published-standardGrounds: Regulated contaminants gated (6 limits) and Never-poison safety gate -- the lead, arsenic, nitrate, bacteria and THM refusals in nx_water_quality carry these EPA limits, and the row note says the WHO-EPA standard defines the limits the products physically remove against.
[epa-pfas-2024] US Environmental Protection Agency. Per- and Polyfluoroalkyl Substances (PFAS) -- the April 2024 PFAS National Primary Drinking Water Regulation setting enforceable MCLs of 4.0 ppt for PFOA and PFOS. Accessed August 2026. publisher · read in our library knowledge/fetched/cmp_water_epa-pfas-2024.html · pin h1f50f73a74db2d65b86c738e0428aaa8221f63d471fdffc8e61f1caaf598a59d · accessed 2026-08-18 · published-standardGrounds: PFAS 2024 limit aware -- the note says Nishi gates the EPA 2024 PFOA and PFOS 4 ppt limit and that WHO-EPA set the number; this is the regulation that set it, and the negative control in nx_water_quality_test (PFAS above 4 ppt per EPA 2024) is pinned to it.
[kim2025-amphiphilic] Kim, B.Q., Vicars, Z., Furedi, M., Escobedo, L.F., Venkatesh, R.B., Guldin, S., Patel, A.J., Lee, D. Amphiphilic nanopores that condense undersaturated water vapor and exude water droplets. Science Advances 11(21), eadu8349, 21 May 2025 (University of Pennsylvania). Open-access copy at PMC12094244. publisher · read in our library knowledge/fetched/cmp_water_kim2025-amphiphilic.html · pin hff52267bfc38384205015b034682ca5c025d5cd82fe744523859fdd7ca903f00 · accessed 2026-08-18 · published-paperGrounds: Multi-technology envelope and Climate-to-harvester recommendation -- the UPenn passive-capillary material the note names (moderate climate to passive capillary) is this paper: capillary condensation in amphiphilic nanopores at undersaturated humidity with spontaneous droplet exudation, no external energy -- the passive class our nx_water_awg model carries beside condensation, sorbent-MOF and fog.
[watergen] Watergen. Watergen -- atmospheric water generators (GENNY, GEN-M, GEN-L), condensation-class water-from-air appliances with published energy and yield figures. Accessed August 2026. publisher · read in our library knowledge/fetched/cmp_water_watergen.html · pin h57446c6db9885297292c7b9cf0f36f16a50c79717da1b19ba299f52e2e7b134b · accessed 2026-08-18 · vendor-docGrounds: The Watergen column: Water-from-air harvesting model (condensation class, Best on Physically produces water and Installed appliance), Modeled specific energy Wh per L (the note quotes Watergen's published about 0.3 to 0.4 kWh/L against Nishi's about 208 Wh/L humid first-order model) and Field-proven yield at scale.
[source-hydropanel] Australian Renewable Energy Agency (ARENA). SOURCE Hydropanel Demonstration -- project page for the SOURCE Global solar-thermal sorbent hydropanel trials (project report updated 2026). Cited in place of source.co, which answered Store unavailable (HTTP 402) to both the sovereign fetcher and an independent read on 2026-08-18. publisher · read in our library knowledge/fetched/cmp_water_source-hydropanel-arena.html · pin h4e71ccf1f239513262757bea2f29f4fc2e4911942c8b3282cd701b838008733f · accessed 2026-08-18 · vendor-docGrounds: The SOURCE Hydropanel column: Sorbent low-humidity capability (SOURCE hydropanels and MOF sorbents coded Best in dry desert air) and Harvester technology classes modeled (1 sorbent-solar) -- an independent government demonstration record of the sorbent-solar class that Nishi models but does not build.
[kara-water] Kara Water. Kara Pure 2 and Kara Pod -- atmospheric water generation appliances that condense water from air and remineralise it to alkaline drinking water. Accessed August 2026. publisher · read in our library knowledge/fetched/cmp_water_kara-water.html · pin h8f9a2d81a5d6e7a5532739515a4fc14342ca5071289d3424ec1693167165f456 · accessed 2026-08-18 · vendor-docGrounds: The Kara Water column: Water-from-air harvesting model (condensation class), Remineralization design generative (Kara adds a fixed mineral blend where Nishi doses calcium, magnesium and bicarbonate by design) and Mineral health scoring Ca Mg.
[zerowater] ZeroWater. ZeroWater 5-stage ion-exchange filter pitchers and dispensers with the included TDS meter reading 000. Accessed August 2026. publisher · read in our library knowledge/fetched/cmp_water_zerowater.html · pin h2bb12736fb3bbb005e036d84015f7dd14a2232e4acd9e2b42631d56881802d5c · accessed 2026-08-18 · vendor-docGrounds: The ZeroWater column: TDS taste band scoring (the note says ZeroWater reads near 0 TDS -- a TDS meter shows one number where Nishi scores the WHO bands) and Mineral health scoring Ca Mg (ZeroWater strips minerals out, coded No), plus Filtration or membrane media coded Best.
[apec-ro] APEC Water Systems. Reverse osmosis drinking-water systems (freedrinkingwater.com now redirects to apecwater.com). Accessed August 2026. publisher · read in our library knowledge/fetched/cmp_water_apec-ro.html · pin hb7fd657200520a5d268995378da00f588a47a74604bb20b990c603374398bfb1 · accessed 2026-08-18 · vendor-docGrounds: The APEC RO column: Harvester technology classes modeled (0 RO filter -- filters do not harvest air), PFAS 2024 limit aware (RO membranes reduce PFAS, coded Yes) and Filtration or membrane media coded Best; the RO stage is also the treatment class Nishi models in Condensate-to-potable treatment chain.

generated by nx_swcompare_matrix (sovereign NishiLang organ) from knowledge/compare/water.matrix · source checks show implementation presence; runtime and user-outcome evidence are reported separately · JavaScript supports page controls