Lintel

The instrument, the argument for it, and the road from here to a national system

Lintel

A conformance authority: a compiled determination, signed by a professional of record, insured, and accepted by a jurisdiction in place of first-pass review. Every part of it already exists somewhere — in American state law, in interstate practice, in Singapore’s production system. Nothing has been assembled this way, and the two joints that matter have never been tested.

This is the argument for it, not a claim that the case is won. “Solution” in the nav is the name of the instrument being argued for. Elsewhere this site says it is not a case for the solution, and both are true: the reasoning is published in full, and the objections that would settle it are stated at full strength on About — the dispositive ones are still open, and no gate has been cleared.

The whole thing in sixty seconds

What exists today

Send an address, get an opinion.

Compiled screening is a real product category with commercial entrants across thousands of counties. It produces guidance. A jurisdiction still performs its own review, which is why the approval did not move. How much time compiled screening takes out has never been measured (E-060).

What does not exist

Send a type and a parcel, get a determination.

Every binding provision evaluated, conformance stated, exceptions referred to the jurisdiction, signed by a licensed professional and insured — and accepted in place of the first pass a person performs today.

Why that is the whole difference

Legal weight, not better software.

A screening tool tells you what it thinks. A determination carries a professional obligation and an insurer behind it, so a class of review leaves the official’s desk entirely. Nobody has offered a jurisdiction that trade.

The two things nobody has demonstrated: that a jurisdiction will accept it, and that a carrier will insure it. Both are named below →

Part one

The instrument

What it is, what you can ask it, and what it issues. Everything after this part is the argument for why it takes this shape and sits in this place.

The whole system 01 of 7

Reusability: Nothing carries forward

The same journey, twice. Same standards, same reviewers — the difference is whether anything survives to the next one.

TODAY · EVERY PROJECT STARTS FROM ZERO LANDoptioned blindDESIGNdrawn bespokeREVIEWby hand, serialAPPRAISALneeds compsLENDINGworst-case pricedBUILDstop-startSALEbid on scarcityNEXT ONEstarts from zero NOTHING CARRIES FORWARD — the tenth project costs what the first did REBUILT · THE TYPE AND THE RECORD CARRY QUERYenvelope, freeTYPEfrom the registryCHECKdeterminationJURISDICTIONaccepts + refersLENDINGtype recordFACTORYcommitted orderSALEquoted priceNEXT ONEstarts from the type THE TYPE CARRIES FORWARD — the tenth project is cheaper than the first Same eight steps. Same standards. What changes is that three of them stop being redone from scratch, and one of them — the determination — is new. That is the whole structural difference between the two rows.

Today: land is optioned before anyone knows what is permitted, the design is drawn bespoke, review runs by hand and in series, appraisal needs comparables that do not exist, lending is priced to the worst case, building runs stop-start, and the sale is a bid against scarcity. Then the next project starts from zero. Rebuilt, the same eight steps run against a certified type and a shared record — and the loop returns to the type rather than to nothing.

Investors

That loop-back arrow is the whole thesis in one mark. An industry where the tenth unit costs what the first did has no products, only projects, and no compounding of any kind.

Policymakers

Count what is unchanged: the standards, the reviewers, and every check that gets made. What changes is that the first three steps stop being redone from scratch, one step — the determination — is new, and the loop at the end returns to the type instead of to nothing.

Research

The claim is about marginal cost, and it is not measured here. That the tenth project would be cheaper than the first follows from reuse rather than from evidence — no American jurisdiction has produced the data because no American jurisdiction has the arrangement.

RegisterE-021 E-022 E-002 check these rows →

Deferred 02 of 7

Charter a conformance authority

A conformance authority — designed in full, deferred in full, and revivable by two measurements.

COMPILED HOUSING — THE END STATE Five stages. Each one ordinary in some other industry, or in some other country. 01Rules compilea query, not an opinion02Type certifiesonce, not per project03Gates run at onceone record, one clock04Lender reads itthe type, not the comp05Factory has itbuilt to a real order LINTEL a conformance authority — the part you actually build. Delegated design approval is federal law from 1976 compiled determination + professional of record + insurance = accepted in place of first-pass review STAGES 04 AND 05 ALREADY EXIST Lenders and factories are not waiting on new technology. They are waiting on something trustworthy to underwrite and to build against — which is what stages 01 to 03 produce.

What was designed is a conformance authority called Lintel: five stages, only the first three new — compiled rules, a determination against one parcel, a professional signature with insurance behind it — and the check itself certified rather than each building (E-088). It would answer two questions. One is advisory and already sold by others: send an address, get an opinion (E-040–E-042). The other carries legal weight and nobody sells it: send a certified type and a parcel, get a determination a jurisdiction accepts in place of first-pass review. Seven layers, each with its own kill test; three ordinary building types that are legal somewhere and impossible here (E-036). All of it is deferred, because the two joints that carry the load are unmeasured: whether binding siting provisions are mostly computable (E-039), and whether the product is cheaper like-for-like (E-098). Those two measurements would revive it. Until then this section states the design and does not argue it.

Investors

The distinction that matters: screening tools produce opinions, the jurisdiction still performs its own review, and what that market did to the clock has never been measured (E-060). A determination would carry a professional obligation and an insurer behind it. That trade — advisory to load-bearing — is the product, and it is exactly the part no jurisdiction has accepted.

Policymakers

Nothing here asks a jurisdiction to surrender judgement: discretionary provisions stay with the jurisdiction, referred by construction. What the design removes is re-verification of the already-certified — the trade Arizona already makes for the factory-built shell itself (A.R.S. §41-4006, E-132).

Research

The design documents are HAP-05 and HAP-06, and the sixteen-model autopsy is on /models. The rows this section rests on keep their grades: E-039 and E-098 empty, E-060 unmeasured, E-047–E-050 as graded. Nothing in this section upgrades them.

RegisterE-021 E-022 E-036 E-040 E-041 E-042 E-047 E-048 E-050 E-060 E-088 check these rows →

Part two

Why it sits there

Three locks bind American housing and they bind in a fixed sequence. Every serious attempt picked one and called it the answer — and the pathology underneath all four.

The honest edge 03 of 7

Make the housing code compilable

The chart is schematic. The crossing point has never been measured — and that is the claim, not a weakness in it.

Stated plainly, at the moment it would pay not to

The shape of the curve is the claim.
The axis values are the research question.

Quantifying the crossing point — cost per jurisdiction to encode, at a stated accuracy level — is the single most valuable number this project could produce, and it does not currently exist.

The shape of that curve is the claim; the axis values are the research question. Cost per jurisdiction to encode, at a stated accuracy, is the most valuable number this project could produce, and it does not exist. The method has changed: hand-encoding was the only way to turn prose into executable rules until about 2023 (E-051), and large language models make that step something a machine can attempt. That moves the cost, not the question — four jurisdictions inherited Singapore’s encoding free and none reached production (E-080, E-081).

Investors

An opportunity that is real and unmeasured is the only kind still available to a new entrant. If the number had been published, the position would already be taken.

Policymakers

No state has instrumented any of this. Arizona passed a significant housing reform and built nothing to determine whether it worked. Confirming that absence is itself a finding.

Research

Thirty-two register rows are empty and they are the ones the argument leans on hardest: E-039, the objective-versus-discretionary split, never measured. E-051, the cost to compile one jurisdiction. E-037, Arizona permit volume since preemption. E-038, the idea-to-sale cycle. E-060, what compiled screening does to cycle time. Plus E-070 and E-071, both assumptions.

RegisterE-039 E-051 E-080 E-081 E-037 E-038 check these rows →

The consequence 04 of 7

The locks bind in an order

Politics chooses the ground. Information is the wedge. Capital is the thesis.

Stage
The question it answers
What binds there
ENTRY
Can this be built here at all?
POLITICS — the principal-agent structure decides which segments may exist
THROUGHPUT
Given permission, how fast and how certainly?
INFORMATION — prose rules force per-instance review
SURVIVAL
Given throughput, does it survive at scale?
CAPITAL — comparables-based valuation kills anything without neighbors

Three things block American housing and they bind in a strict order. Entry: can this be built here at all — politics decides. Throughput: given permission, how fast and how certainly does a unit move — information decides. Survival: does the product last at scale — capital decides. The binding constraint is not a fact about housing. It is a function of which stage you are standing in.

Investors

Sequence by politics, wedge with information, win on capital. Type-based valuation is the endgame because it is the lock that kills at scale, and the position with no identified competitors.

Policymakers

Politics first does not mean lobbying first. It means letting the map choose the segment — build where the incumbent homeowner is also the builder, because that is the only ground where entry has already been granted.

Research

Note the asymmetry in evidence strength. The valuation lock has the best evidence because it is the only lock America has ever gotten far enough to fail at. Evidence strength tracks how far each attempt got, not how much each lock matters.

RegisterE-007 E-008 E-021 E-027 E-028 check these rows →

The alternatives 05 of 7

Every other order has been run

Four sequences, four failures, and each one is already in the record.

SequenceWho ran itWhy it failed
Capital firstKaterraMoney waited in a manual queue. No certified type for finance to attach to.
Information firstThe screening marketCompiled rules without political standing are advisory. The rules got compiled and the approval did not move; how much time it took out has never been measured (E-060).
Politics onlyArizonaThe right granted, the plans mandated, and the volume flat. No shared record, no type-based finance.
Politics + capitalThe HUD CodeClosest anyone came. No information layer, so finance defaulted to chattel.

Capital first: Katerra. Money waited in a manual queue. Information first: the screening market — Symbium, Buildability, Canibuild — compiled rules across thousands of counties and did not displace a single approval, because compiled information without political standing is advisory. How much time it took out has never been measured by anyone, including us (E-060). Politics only: Arizona, a granted right with nothing to exercise it. Politics plus capital, skipping information: the HUD Code, the closest anyone came, with no type-based valuation record to lend against.

Investors

The screening market matters most here. It is the information-first attempt at commercial scale, and the observable part is qualitative: the rules got compiled and the approval did not move. Compiling the rules is necessary and insufficient — the wedge has to carry legal weight, not just information. No number belongs in that sentence. This site used to put one there, sourced to an estimate of its own that it described as the market's published answer; the estimate is withdrawn and nothing replaces it (C-019).

Policymakers

The failures are not symmetric. Skipping politics wastes a company. Skipping information wastes a statute. Skipping capital wastes an industry.

Research

This corrects an error in the thesis as published: Red Team 04 claims existing vendors leave the ruleset in prose. Register row E-040 refutes that. The sharper form is the one on this slide.

RegisterE-023 E-027 E-037 E-040 E-041 E-042 check these rows →

The paradigm 06 of 7

Redundancy: Everyone keeps their own copy

The industry is not a line. It is a mesh in which everyone rebuilds the same facts and nobody is authoritative.

TODAY · SIX PARTIES REBUILD THE SAME FACTS ARCHITECTPLANNERPLAN REVIEWERUTILITYAPPRAISERINSURER 15 possible conversations · 6 reconstructions · no shared answer None of them is authoritative, so disagreement is resolved by meeting. REBUILT · SIX PARTIES READ ONE RECORD SHARED RECORD + DETERMINATION ARCHITECTPLANNERPLAN REVIEWERUTILITYAPPRAISERINSURER 6 connections · 1 record · 1 authoritative answer Disagreement is about judgement, not about what the facts are.

An architect, a planner, a plan reviewer, a utility engineer, an appraiser and an insurer all need the same facts about the same parcel — dimensions, easements, utility access, flood status, setbacks, what is permitted. Every one of them reconstructs those facts independently, none of them is authoritative, and disagreement is resolved by meeting. Six parties is fifteen possible conversations. Give them one record and one determination to read, and it is six.

Investors

This is the pathology behind every other one. Six independent reconstructions of the same dataset is not inefficiency at the margin — it is the reason a project cannot be a product, and the reason the tenth costs what the first did.

Policymakers

Note what the right-hand picture does not change: the same six parties, with the same authority, asking the same questions. What changes is that they stop each producing a private answer. Singapore's eight agencies read one national schema, which is why they can answer together inside a twenty-working-day ceiling.

Research

The claim that this is the binding pathology is argued rather than measured. Layer 3's falsification test sits exactly here: if most measured delay turns out to be jurisdiction queue time rather than applicant resubmittal from inconsistent facts, a shared record delivers considerably less than this slide implies.

RegisterE-022 E-021 E-050 check these rows →

The whole process, drawn Detail

Every party, every approval, twice

The same transaction as it runs today and as it would run compiled. First both on one axis, then each drawn in full at the same scale — so the difference is a fact of the picture rather than a claim underneath it.

Both, on one ruler How long each approval takes to land Approval queue only

The same nine approvals, the same axis, drawn twice. Each bar runs from the start of the project to the moment that approval is finally in hand. Nothing is removed — every one still happens and every one can still stop the project.

Read the totals with the correction attached. This site used to head this figure “thirty-two months to seven”. That headline was withdrawn in August 2026: four of the seven months are factory build, so the total set site-built-today against factory-built-tomorrow and credited compiled approval with a saving that partly belongs to a change of construction method. What survives is the approval queue below, which is the only part this architecture claims to touch — and it is an estimate, not a measurement. What changed →

The three longest queues — 14.8 of the 32 months

  1. Planning — G18.1 months holding the file
  2. Plan review — G25.0 months holding the file
  3. Lender / appraiser1.7 months holding the file

None of these is construction. They are the three points where the file sits on a desk waiting for a person, which is exactly what a determination is meant to remove — and together they are more than half the critical path.

Today — each approval begins only when the one above it has finished

048121620242832MONTHS FROM STARTLender / appraiserLender / appraiser: starts month 2.1, answers month 3.8 (1.7 months holding the file)1.7 mo · done 3.8LONGEST QUEUEHOA / covenantHOA / covenant: starts month 3.8, answers month 4.7 (0.9 months holding the file)0.9 mo · done 4.7Planning — G1Planning — G1: starts month 4.7, answers month 12.8 (8.1 months holding the file)8.1 mo · done 12.8LONGEST QUEUEPlan review — G2Plan review — G2: starts month 12.8, answers month 17.8 (5.0 months holding the file)5.0 mo · done 17.8LONGEST QUEUEUtility — G3Utility — G3: starts month 17.8, answers month 19.4 (1.6 months holding the file)1.6 mo · done 19.4Fire district — G4Fire district — G4: starts month 19.4, answers month 20.5 (1.1 months holding the file)1.1 mo · done 20.5Districts — G5Districts — G5: starts month 20.5, answers month 21.4 (0.9 months holding the file)0.9 mo · done 21.4InspectorInspector: starts month 21.4, answers month 29.6 (8.2 months holding the file)8.2 mo · done 29.6Title / insurerTitle / insurer: starts month 29.6, answers month 31.8 (2.2 months holding the file)2.2 mo · done 31.8

Compiled — the seven gate rows share a start

048121620242832MONTHS FROM STARTLender / appraiserLender / appraiser: starts month 0.62, answers month 0.96 (0.3 months holding the file)0.3 mo · done 0.96HOA / covenantHOA / covenant: starts month 0.62, answers month 0.92 (0.3 months holding the file)0.3 mo · done 0.92Planning — G1Planning — G1: starts month 0.62, answers month 0.92 (0.3 months holding the file)0.3 mo · done 0.92Plan review — G2Plan review — G2: starts month 0.62, answers month 0.92 (0.3 months holding the file)0.3 mo · done 0.92Utility — G3Utility — G3: starts month 0.62, answers month 0.92 (0.3 months holding the file)0.3 mo · done 0.92Fire district — G4Fire district — G4: starts month 0.62, answers month 0.92 (0.3 months holding the file)0.3 mo · done 0.92Districts — G5Districts — G5: starts month 0.62, answers month 0.92 (0.3 months holding the file)0.3 mo · done 0.92InspectorInspector: starts month 3.1, answers month 6.8 (3.7 months holding the file)3.7 mo · done 6.8Title / insurerTitle / insurer: starts month 6.8, answers month 7 (0.2 months holding the file)0.2 mo · done 7
ApprovalTodayCompiled SavedReduction
Lender / appraiser1.70.31.480%
HOA / covenant0.90.30.667%
Planning — G18.10.37.896%
Plan review — G25.00.34.794%
Utility — G31.60.31.381%
Fire district — G41.10.30.873%
Districts — G50.90.30.667%
Inspector8.23.74.555%
Title / insurer2.20.22.091%
Total, all nine29.7 6.023.7 80%

Months from the start of the project, from the specimen above — modeled values (E-070, graded Untested), printed to one decimal for internal consistency, measurements of nothing. The two smallest reductions are the two approvals that wait on physical work — inspection and closing — which is the honest floor: an approval can be compiled, a foundation cannot. Not shown: the designer and the standalone appraisal, which the certified type and the type-based valuation record replace; and the producer, which is new.

2026 The process as it runs today 32 months

One row per party; each bar is that party holding the file, from the moment it reaches them until they answer. 9 of the 13 can stop the project outright, marked with an amber rule. The bars form a staircase — no two overlap, because nothing runs concurrently. Elapsed time is therefore the sum of the queues rather than the sum of the work, and the eight months that actually build the house are the single long bar near the bottom.

FEAS2 moENTITLEMENT11 moDESIGN + REVIEW5 moSITE4 moBUILD8 moSALE2 moHomeownerapplicantDesignerengagedEquity lenderHELOC / refiAppraiserparent parcelHOA / covenantprivate lawPlanningG1Plan reviewG2UtilityG3Fire districtG4DistrictsG5 floodBuildertradesInspectorduring buildTitle / insurerat closingAn opinion, after paid hours · 1 monthConditional approval · 1 monthValue — comparables exist · 1 monthNo prohibition found · 4 weeksEntitlement granted · 8 monthsCorrections — resubmit · 2 monthsPlan review cleared · 2 monthsCapacity confirmed · 1 monthConditions attached · 1 monthCleared · 4 weeksBuild — staged inspections · 7 monthsFinal inspection passed · 1 monthOccupancy · 1 month048121620242832MONTHS FROM START
2036 The same parties, compiled Compiled

Same rows, same axis, with the approval queue compiled. Nothing was deleted: 8 parties still hold a veto and every standard is still checked. The staircase becomes a column — the gate bars stack vertically at month one because they read the same record at the same time under a statutory clock, instead of queueing behind each other.

Month one, in order: an address returns the buildable envelope and the certified types that fit it, in seconds rather than weeks of paid interpretation; the applicant chooses a type; the determination is issued, signed and insured; and it goes out once, to every gate at the same time. The utility bar is amber — a capacity condition attached, not a clean yes. Concurrency shortens the wait for an answer; it does not make the answer unconditional.

HomeownerapplicantThe recorddeterminationLenderreads the typeHOA / covenantprivate lawPlanningG1 by-rightPlan reviewG2UtilityG3Fire districtG4DistrictsG5 floodProducercertified typeInspectorduring setTitle / insurerat closingSite prepared · unit in production · 1 monthSet and inspected · 3 monthsSigned off · 1 weekOccupancy · 5 days048121620242832MONTHS FROM START
21months of approval queue, on the estimate
This is the only total this page still states, and it is Untested. Twenty-four months of overhead become three in the model — that is the queue between the parties above, and it is the only part this architecture claims to touch. The rest of the difference between the two panels comes from building in a factory against a certified type rather than bespoke on site, which is a change of construction method and is not claimed here. The aggregate comparison that combined the two was withdrawn in August 2026 as not like for like (E-070, E-071, correction C-013).

Both are specimens. No dataset tracks a single project through these parties in order — that is register row E-038, and it is empty. The durations trace to the month strip, to the modeled compiled cycle (E-070, graded Untested), and to CORENET’s twenty-working-day consolidated response — a service ceiling, not a measured average (E-021); the specific ordering is a reasonable reconstruction, not a measurement.

Note what the compiled panel does not fix. Utility capacity, the fire district and private covenants still hold hard vetoes and this architecture does not reach them — they carry the same amber rule in both panels, and the utility bar in 2036 is amber because a capacity condition is attached. Concurrency changes when they answer, not whether they can refuse.

Part three

How it runs

What is actually executing, and what it is assembled from — each layer carrying its own kill test and its own substitutes.

The artifact 07 of 7

A basic example

The product is a document. This is what it would have to contain.

Determination of conformance Specimen
Parcel301-44-092 · Maricopa County, AZ
TypeADU-2BR-R3 rev 4 · certified 2026-03-11
RulesetMaricopa 2021 IRC + county amendments r7
Evaluated2026-08-06 · engine v2.4.1
Binding provisions evaluated214
Objective — computed, conforming198
Requiring judgement — referred to the jurisdiction16
Professional of recordsigned · license on file
Insurednamed carrier · per-determination limit
EffectAccepted in place of first-pass review under the adopting statute. The sixteen referred provisions are reviewed by the jurisdiction as normal.

Specimen — no determination like this has been issued. It is drawn to show what one would have to contain: what was checked, what was computed, what was referred to a human, who signed it, who insured it, and the statutory basis on which a jurisdiction would accept it. The counts are illustrative. The share of provisions that are objective rather than discretionary is register row E-039, and it is empty — which is exactly what Gate 0 measures.

A determination names the parcel, the certified type and the exact ruleset and edition it was evaluated against. It reports how many binding provisions were examined, how many were objective and computed, and how many required judgement and were referred to the jurisdiction. It carries a professional's signature, a named insurer, and the statutory basis on which it is accepted. It is auditable after the fact, because the ruleset version and the engine version are both on it.

Investors

Everything that makes this defensible is on the face of the document: what was checked, what was not, who is liable, and who pays if it is wrong. That is what makes it insurable, and insurability is what makes it acceptable.

Policymakers

The line that should reassure a building official is the third one. Sixteen provisions referred, reviewed as normal. The determination does not claim the discretionary questions — it clears the ones that are arithmetic so the reviewer's time goes where judgement is actually required.

Research

Specimen only. No determination like this has been issued, and the counts on it are illustrative rather than measured. The ratio it implies — how many binding provisions are objective rather than discretionary — is register row E-039, which is empty, and is precisely what Gate 0 exists to measure. If that ratio is bad, this document is thin and the argument is wrong.

RegisterE-039 E-051 E-050 check these rows →

Underneath the specimen Detail

What one line of an ordinance looks like

The determination above states counts. This is what sits under them — and the difference between the top rows and the bottom rows is the entire thesis.

ProvisionWhat it turns on EvaluationResult
Front setbackDistance, structure face to front lot line 22′-6″ against a 20′-0″ minimumPass
Building heightGrade plane to mean roof height 16′-2″ against a 25′-0″ maximumPass
Lot coverageSum of covered area over parcel area 31.4% against a 40% maximumPass
Unit floor areaConditioned area of the accessory unit 744 sq ft against a 1,000 sq ft capPass
Egress window, bedroomClear opening area, width, height, sill 5.9 sq ft clear against 5.7 requiredPass
Rear setbackDistance, structure face to rear lot line 4′-1″ against a 5′-0″ minimumFail
Fire separation distanceDepends on the assumed lot line location where an easement is unrecorded Computable once the jurisdiction fixes the lineConditional
Neighborhood compatibilityWhether massing and materials are “compatible with surrounding character” Not expressible as a rule over measurable attributesReferred
Design review findingWhether the elevation satisfies the reviewing body’s adopted guidelines A judgement reserved to a person by the ordinance itselfReferred

Specimen. The provision categories are real and universal; the values are illustrative. A failure is not an opinion — it is located, measured and citable, which is why it can be fixed before submission rather than discovered at review. The referred rows are the ones a machine must not decide, and they go to the jurisdiction exactly as they do today.

The ratio between those two groups is the whole question. If most binding provisions look like the top rows, the instrument works. If most look like the bottom two, it cannot be built as specified. Nobody has ever measured which — that is register row E-039, and phase 00 below is the measurement.

The road

From one jurisdiction
to a national system

Seven phases. Each lists the stages inside it and the condition that has to hold before the next one starts — because the failure mode of every prior attempt was starting at a later phase than the one it had earned.

No phase below has begun. Durations are given only where they can honestly be given; the ones that depend on a counterparty saying yes are not invented.

00

Prove it is compilable

weeks · no capital · not started

Before anything is built. Two measurements that either support the diagnosis or kill it, and both are publishable whichever way they come out.

  1. Gate 0 Take one mid-sized jurisdiction’s zoning ordinance and adopted building code. Enumerate every provision governing a defined by-right project class, classify each objective / conditionally objective / discretionary, and encode the objective set. Fills register row E-039.
  2. Gate 1 Records requests across Arizona municipalities and counties for accessory-dwelling permit volumes, synthetic control against comparable metros, and builder interviews establishing which gate actually binds. Fills E-037.

Before the next phaseBinding provisions are predominantly objective, and builders name approval-side gates as the critical path.

01

The first acceptance

not started

One jurisdiction accepts one determination for one project class, in place of a first pass a person would otherwise perform. Nothing else on this road is possible before it, and no amount of capital shortens it.

  1. Scope it narrowly One project class, one code edition, a fixed review scope and an explicit right to terminate. A pilot memorandum, not a contract.
  2. Instrument every determination Predicted result against actual jurisdiction decision, elapsed time, resubmittal count. The accuracy record cannot be manufactured retroactively.
  3. Publish it as research Method, sample, accuracy and failure modes — including the misses. Officials are comparison-sensitive and talk to each other; the measurement is what generates the next conversation.
  4. Repeat it four more times Proof the first yes was a product rather than a relationship.

Before the next phaseFive jurisdictions accepting, an accuracy record a reviewer will state out loud, and a carrier renewing rather than declining.

02

Saturate one state

not started

Arizona, end to end. The objective is not volume — it is a density of acceptance high enough that a legislature can be shown a working system rather than a proposal.

  1. Compile every mandated jurisdiction Not every jurisdiction — every one carrying a statutory obligation. The obligation is both the filter and the argument.
  2. Seed the type registry Certify reference designs as manifests: envelope, load cases, energy path, fire basis, allowable site conditions.
  3. Open the registry to third parties Architects and producers submit designs; the registry certifies and maintains them across code cycles. The marginal cost of a new type approaches zero.
  4. Publish the first cycle-time dataset By named jurisdiction: application to issuance, resubmittals, variance. Register row E-038, closed.

Before the next phaseMarginal encoding cost per jurisdiction has fallen by more than half. If it has not, the compiler is not compounding and national coverage never becomes affordable.

03

Turn practice into law

not started

Convert bilateral agreements into statutory standing, so the arrangement survives an official changing jobs. This is where the instrument stops depending on relationships.

  1. Amend what already exists Recognition of an insured conformance determination under existing third-party-review or industrialized-building provisions. An amendment to a familiar frame, not a new act.
  2. Let the users testify The building officials who have used it for two years, with the published dataset as the exhibit.
  3. Choose the second state by statute The state whose existing law most resembles the one just amended — not the largest housing market. Optimize for resemblance; market size is what later states are for.
  4. Industrialize the compiler Ingestion, classification, encoding and a regression suite that computes which certified types an amendment invalidates. This is what makes the fortieth jurisdiction a budget line rather than a project.

Before the next phaseRecognition law in one state and a bill filed in a second. If the amendment cannot be carried in the friendliest available state, statutory standing is not reachable and the model caps at bilateral agreements — slower and smaller, still viable.

04

Cross the state lines

not started

Multiple states, and the beginning of the thing that actually matters: a valuation record dense enough to lend against.

  1. Use the reciprocity that exists Interstate recognition of streamlined certification is an existing instrument, not one this plan invents (E-048). Bringing certified types under it gives American housing type portability for the first time.
  2. Expand by statutory family Group states by the shape of their existing law and run them in cohorts. The constraint becomes legislative calendars, which are annual and parallel.
  3. Start the valuation record Every certified type accumulates as-built cost, time to completion, performance, defect and warranty claims, and — as units trade — resale prices. It takes years and cannot be accelerated, which is why it starts here.
  4. Get one portfolio lender to underwrite the type Not a GSE. A balance-sheet lender who reads the type record instead of waiting on comparables. The first time in American history a housing type rather than a neighborhood carries the valuation.

Before the next phaseTypes portable under reciprocity, a multi-year valuation record, and one lender underwriting to type. Without that lender, the valuation lock is harder than modelled and the phases below do not follow.

05

Make demand visible

not started

Turn a catalogue into an order book, and an order book into utilization — the one thing every other manufacturing industry has and housing does not: knowledge of what has been bought before it is made.

  1. Publish the catalogue A household sees what can be built on their parcel, what it costs, when it can be delivered and what it will appraise at — every number drawn from the type record rather than quoted by a salesperson.
  2. Convert to committed orders Refundable deposits first. The deposit book is the demand signal and also the political instrument: forty families waiting on an approval is a different sentence from a developer wanting forty units.
  3. Contract existing capacity Underused modular and panelized capacity already exists. Contract it against the book. Buying utilization for someone else’s asset is what makes their economics work and keeps this side variable.
  4. Close the loop Delivery data improves the type record, which improves valuation, which improves financing, which widens the catalogue, which deepens the book. This is the compounding American housing has never had.

Before the next phaseDelivered cost per unit falling year over year on like-for-like types. If it is not, the loop is not closed and this is an excellent approvals business with a catalogue attached.

06

A national system

not started

The end state is not a large company. It is that a compiled determination is simply how American housing is approved and valued — the way an evaluation report is already how a building product is accepted.

  1. Take the record to the secondary market With a decade of type-level cost, performance and resale data, seek type-based valuation treatment. This is the last lock, and it decides whether compiled housing reaches the median household or stays a premium product.
  2. Open the rule graph License the compiled, versioned, diffable national ruleset to anyone, competitors included. By then the moat is the acceptance network, the accuracy record and the liability capacity — none of which a licensee can copy.

Before the next phaseA determination is accepted by default rather than by agreement, a type carries its own valuation, and a household can price and order a home before committing. At that point the thesis is no longer a thesis.

Beneath itWhat is already settled in law

The parts are not hypothetical

Three things this road depends on already operate in American law. They are the reason it is a proposal rather than a wish.

Their register rows are graded Untested, and the grade is about the filing rather than the law — C‑017 found each source field named no document a reader could open. Nothing here upgrades them. Naming the statute and re-reading it is what would.

Untested

State preemption over approved buildings Georgia preempts local authority over approved industrialized buildings, with the Commissioner’s determination final. Equivalents in Maryland, Pennsylvania, Texas, California and New York. Register row E-047.

Untested

Recognition across state lines The Interstate Industrialized Buildings Commission recognizes streamlined certification between states. One certification, many jurisdictions — the mechanism already exists. Register row E-048.

Confirmed

Private review as a statutory right Indiana: seven business days for plan review, three for inspection — and an applicant may use a private provider regardless of whether the unit can meet them. The unit shall accept the report without further inspection, and is immune for the provider’s acts. Register row E-050.

The first questionWhat a city attorney asks

What happens to the citywhen the determination is wrong.

Every liability sentence on this page used to point at the professional side — signed, insured, sealed. None of it answered the question a municipal attorney asks first, which is what happens to them. This is the reason building officials resist third-party review even where state law already allows it, and the site had no answer for it until August 2026.

Confirmed

Two legislatures wrote the answer into the enabling act. Florida immunises the local government, the building official and code enforcement personnel from liability for any act or inaction of a private provider — and that clause has been there since the original 2002 enactment, not bolted on after a loss. Indiana did the same in 2025 and indexed it in the state’s statutory immunity list. Register rows E-160, E-171, E-173.

Confirmed

The immunity is bought with the second look. Florida bars the building official from replicating the review, caps audits at four per provider per year, forbids them delaying the work, and deems the certificate of occupancy granted if the official misses ten business days. A jurisdiction cannot take the immunity and keep the right to check. E-172.

Untested

Neither immunity covers the jurisdiction’s own negligence, and no court has said otherwise. Both statutes immunise the unit for acts of the owner or the provider. Neither reaches the unit’s own failure to verify a provider’s licence or insurance, its own certificate decision, or its own safety judgement. In twenty-four years of Florida operation, no reported appellate decision has construed the clause either way. E-174, E-175.

Confirmed

Where no such statute exists, the jurisdiction has a defence rather than an immunity. The background rule is split and it decides adoptability. Washington holds no duty is owed for negligent building-code enforcement; Colorado, Arizona and Oregon abolished the public duty doctrine outright; and Louisiana held a city-parish liable for issuing on structural plans nobody reviewed, after the building collapsed and killed three workmen. California and Nevada then immunised negligent inspection by statute. E-179, E-181.

Confirmed

The manufactured-housing precedent does not transfer, and this site nearly claimed it did. HUD preempts rather than delegates — no state may require its own inspection of anything the federal standards cover. There is no acceptance decision to get wrong, which is why the regulations allocate no liability to any state or local authority. Citing DAPIA and IPIA as precedent for a jurisdiction accepting an outside determination is a category error. E-176.

So the honest answer is not “the professional carries it.” It is that the legislature has to say so explicitly, and two have. There are three models in American law — immunity (Florida, Indiana), preemption (HUD), and retained duty, where Washington DC expressly provides that third-party review does not relieve the District of its own obligation to review. Which model a state has decides whether any of this is adoptable there, and it is not a question the instrument can answer for itself. E-177.

One thing that cuts against the design: the federal test for discretionary immunity turns on whether an official exercised judgement at all, and it does not apply where a rule specifically prescribes a course of action. A compiled determination is precisely a specifically prescribed course of action. The more mechanical the rule, the less discretionary the act, and the less immunity attaches to getting it wrong. That is stated here rather than left for a reader to find. E-182.

Above itWhere the evidence runs out

Three things nobody has shown

Stated plainly rather than buried at the end. This is the part of the argument with the least evidence behind it and the most weight on it.

Untested

A jurisdiction will accept the determination. No American jurisdiction is currently known to accept a private compiled determination in place of first-pass review. Phase 01 exists to answer it.

Untested

A carrier will insure it. No insurer is known to have priced liability written against a conformance determination. It is the question that has killed certification-reciprocity schemes more often than politics has.

Untested

American codes are compilably objective. The share of binding provisions that are objective rather than discretionary has never been measured, anywhere — register row E-039 is empty. Phase 00 measures it, and a negative result sends the first four layers back to redesign.

Naming a thing does not make it exist. Two of the three are testable within a year and neither needs significant capital — that is phase 00, and every claim on this page is graded in the evidence register.

FalsificationThe ordering is falsifiable, and here is how

What would reorder this

The sequence is argued, not assumed. Five results would rewrite it, each named before the evidence rather than after.

PendingRow ResultEffect on the ordering
Arizona throughputE-037 Permit volume rose sharply after preemption despite no supporting infrastructurePolitics alone suffices for the entry segment. Information demotes to a convenience.
The gate mapGate 1 Builders name utility capacity or fire districts, not approval, as the critical pathThe throughput stage is misdiagnosed and the instrument is aimed at the wrong gate.
CompilabilityE-039 Binding provisions prove predominantly discretionaryThe information lock cannot be engineered. The sequence collapses to politics plus capital — the manufactured-housing path with modern finance.
Acceptance and insurabilityE-159/160/161, E-169 A determination cannot be worded as a certification and still be insured, or no jurisdiction uses the mechanism it already hasThe wedge is permanently advisory and the ceiling is the screening market.
Firm sizeE-016/017 Project scale, not approval mechanics, drives productivityType certification behaves as entry regulation, and the small-builder framing is dropped.

When one of these lands, this page gets a revision rather than a defense.

Three doorsWhere this sits, and what would settle it