FORKC_ANALYSIS_CANON_TRAVEL_2026-07-27.md

pipelines/FORKC_ANALYSIS_CANON_TRAVEL_2026-07-27.md

FORK C — AXIS: CANON TRAVEL AUDIT

What each ACTUAL slice seam is in canon; whether canon stages content on those voyages or elides

them; what the walked-geography ethos demands; and what a transition may not show pre-Ch-38.

Read-only analysis. Every claim carries its artifact. Absence-claims are POSITIVE-CONTROLLED (the

search is shown to match known-present content before its zero is trusted), per the standing

a-search-that-cannot-match-reports-zero rule.

---

0. THE ONE-LINE ANSWER

**Canon does not elide these crossings — it elides only their MIDDLES, and it has already authored

their heads and tails as playable beats.** The boarding, the route, and the landfall are typed beats

today; the open water is unauthored. So the question "cover or playable leg?" is answered by canon

as *both, at different points of the same seam* — and at these two specific seams the canon-correct

playable leg and the doctrine's strongest cover class are the same object: a deck the player

walks while the world streams.

The cover is the LESS honest option where canon has authored content (the leg heads/tails and the

Kra traverse), and the MORE honest option for the open-water middle, which canon deliberately

declines to describe.

---

1. WHAT EACH ACTUAL SLICE SEAM IS IN CANON

The governing artifact is docs/spine/transitions/TRANSITIONS_CH02_21.md — the Layer-2 inter-chapter

pass authored at the RR-22 window, 21 seam entries, gate-enrolled, DOC_MAP-indexed. Its own §"What

this document is and is not" (L9) declares its consumer: **"the 3D build (each seam maps to a

discrete arrival or departure SEQUENCE ASSET)"** — canon's declared intent for a seam is a *built

sequence*, not a screen.

Distances below are my derivation by great-circle from the WGM's own published coordinates (WGM

§§3-4 Ch 3/4/5 entries), flagged as derived, not canon text.

1.1 SEAM CH_03 → CH_04 — Flores → Bali

TRANSITIONS_CH02_21.md §"Ch 3 to Ch 4 — Manggarai highlands to Bali (down to the coast, then by sea)":

Flores by sea toward Bali, the architecture-and-unity weaving crossing; Mama Klara stays behind and

no continuous companion departs alongside."

the range's first water leg."

then down into the villages below."

Shell, now bonded, travels the water alongside for the first time as the chosen-bond companion."

strategic (ACTIVE_CONCEALMENT); healer's-child watching-mystery carried as unease; no human

companion travels on.

Canon-side beat coverage:

player_objective ends: *"Say what needs saying to the ones who will hear it, and **take the road

down to the coast**."* The walk to the coast is issued to the player as a task.

from Besakih**, the mother-temple high on Gunung Agung's flank, into the villages below…"*

up Gunung Agung's flank to the mother-temple. All of it unhosted by any beat, site, or scene row.

1.2 SEAM CH_04 → CH_05 — Bali → Sumatra and Java

TRANSITIONS_CH02_21.md §"Ch 4 to Ch 5 — Bali to Sumatra and Java (by ship)":

Captain I Wayan Dwija**, sailing toward Sumatra and Java."

monument-country, coming in by river and road; continuous ground-and-water travel, no flight."

Sumatra."

Bali Starling, bonded at Ch 4, rides the passage as the roster begins to accumulate**."

turning outward into reading others.

Here canon puts the departure INSIDE a beat, and types it explicitly. CH04_B12:

CH_04,CH04_B12,12,exit,traversal,uluwatu_seacliff_temple,none,none,
  "Carry the balance out past the Uluwatu sea-cliff temple — the cliff-vertical agility course run
   above the surf, the wind-step and gust-push learned on the exposed rock — and take ship from Bali
   across the water toward Sumatra and Java…"
  actors: protagonist;CHAR_0029
  player_objective: "Wayan Dwija's boat goes west on the turning tide, and he will take
                     a passenger who can hold a rope. Take your leave of people properly first."

Four load-bearing facts in that one row:

1. role=exit, mode=traversal — traversal is a first-class beat mode (42 traversal beats across

20 chapters in build/spine_rows/*/beats.csv), not an exception carved for this seam.

2. The captain is an actor (CHAR_0029), so the seam already has a person in it.

3. The player_objective — already written in the natural-voice register Josh ruled on 2026-07-26 —

casts the player as crew, not cargo: *"he will take a passenger who can hold a rope."*

4. Its out-state advances real ws: local.departed_bali=true; ws.identity_hiding_status=+active_concealment_pressure_med.

Derived distance: ~554 km Uluwatu → Borobudur.

FLAG (game/canon mismatch). UHumanityQuestSliceDriver::MapForChapter("CH_05") returns
/Game/Maps/SamosirToba, and TODO_CANON.md calls it "the CH_05 ARRIVAL map." Canon's Ch-5
arrival is Central Java / BorobudurCH05_B01 is seq 1, site borobudur, *"Come into Java's
monument-country by river and road and climb Borobudur's terraced mandala."* Samosir is beat 5
(CH05_B05). The seam currently lands the player at the chapter's fifth site and calls it arrival.

1.3 THE CH_05 INTERNAL BELT — not four maps in a line

The slice ships the belt as samosir → palembang → borobudur → angkor (TODO_CANON: *"The further

intra-chapter map travel (Samosir → Palembang → Borobudur → Angkor) is not a runtime seam yet"*).

Canon's route is a different shape, and it is not a line — it doubles back. From CH_05.md §10

beat graph + §"Key sites":

legfrom → tocanon carrierderived gc km
1borobudur (B01-B03) → palembang_port (B04)"Come down from the high stone to the Srivijayan river-port"791
2palembang → samosir_highlands (B05-B06)"Climb into the Batak highlands to the great caldera-lake"910
3samosir → angkor_wat (B07) via kra_traverseMalacca Strait + Kra portage + Gulf + Tonle Sap1328 (gc; canon route far longer)
4angkor → bayon_taprohm (B08) → west_baray (B09)local
5myson_champa (B10)"Cross to the coastal Champa sanctuaries"527
6THE DOUBLE-BACK myson → samosir (B11, the climax)"then double back down the whole long route to the sea"; out-state local.doubled_back: truefull route reversed
7samosir → palembang (B12, exit)"Come back down to the coast… take ship across the Bay of Bengal"910

The decisive belt fact: canon has already typed one of these legs as a first-class SITE.

kra_traverse: terrain coastal · functions [traversal, dialogue] · ambient_threat low ·
  vril_density baseline · protected false
  env: the Malacca Strait, the Kra isthmian portage, and the gulf, with their port-audiences and
  tribute politics (Tambralinga, Grahi/Chaiya and the 1183 Old-Khmer Grahi Buddha, the
  Mon-Dvaravati Theravada substrate), scope-controlled against sprawl; a traversal connector for
  the excursion, traversed at the top of CH05_B07 (hosts no primary beat); L2.

A dialogue function. Port audiences. A readable L2 inscription (the 1183 Grahi Buddha) found

*en route*. That is canon authoring a travel leg as a place the player is in, with people to talk

to and lore to read — with the explicit note that it hosts no *primary* beat, i.e. it is connective

tissue that is nonetheless a site. Canon's own answer for the belt's longest hop is not a card.

And CH05_B07's natural-voice player objective already narrates the whole route as the player's own

plan: *"The ledger does not stop at the strait. It runs north — over the isthmus, up a lake that

floods backwards — to a mountain the Khmer built out of stone."*

---

2. DOES CANON STAGE CONTENT ON THESE VOYAGES, OR ELIDE THEM?

Split verdict — and the split is the finding. Canon elides the open-water MIDDLE and authors the

BOARDING, the ROUTE, and the LANDFALL.

2.1 What is genuinely ELIDED (four positive-controlled zeros)

(a) No at-sea SITE at CH_03/04/05.

combat;learning;transformation) and CH_38/deck_at_the_midpoint (water_body;

discovery;character;puzzle;transformation).

(both terrain coastal, i.e. shore, not water) and palembang_port.

(b) No voyage cell in the 22-thread grid at Ch 3-5.

sea|ship|sail|voyage|boat|crossing|strait|monsoon|navig|thalasso.

body and WGM, not in a scored thread cell — the strait is *context*, not a threaded anchor.)

anchor per open-ocean chapter-type"* — canon has an explicit *chapter type* for open ocean, and

it does not apply here.

(c) No voyage seed in the 400-row seed-housing ledger.

shelved by REGION (flores-indonesia, bali, sumatra-batak, angkor-khmer, malay-world,

vietnam, fairy-literary). **There is no maritime/voyage shelf and no chapter_ref keyed to a

seam.** The only sea-adjacent rows are region-internal (The Ship That Carries the Teeth,

sumatra-batak) or fairy-literary. Nothing is housed *on a crossing*.

(d) No chapter_bridging_cinematic scene row anywhere in the arc.

meaning. Story load belongs to the authored bridging scene (chapter_bridging_cinematic)."* That

designated home does not exist as data. Today the seam's meaning has no home at all: not a

cover (DT-3 forbids it), not a bridging scene (zero rows), not a beat (the transitions file is

prose only).

(e) No elapsed time to montage. The transitions file's internal-age ledger (L23): *"about 7

rising to 9 across Ch 2, 9 through Ch 3 and Ch 4, 10 across Ch 5."* Neither slice crossing

advances the age clock. Canon dates both as days, not seasons — consistent with ~550 km coastal

sails. A card has nothing to date, and L17 forbids dating it anyway (*"the protagonist's internal age

is designer-only and never surfaced"*).

2.2 What is NOT elided (canon authors the heads, tails, and one middle)

1. The walk to the coastCH03_B12 player_objective: *"take the road down to the coast."*

2. The boardingCH04_B12, a typed traversal exit beat at the departure site, with the

captain as an actor and the player cast as working crew.

3. The landfall and approachCH05_B01: *"Come into Java's monument-country **by river and

road**."*

4. A whole route as a beat objectiveCH05_B07: the Kra portage, the gulf, the reversing-flood

Tonle Sap, and an inscription found *on the peninsula*.

5. A travel leg typed as a SITE with a dialogue functionkra_traverse.

6. A specific, named, non-fungible thing on each slice crossing — the transitions file's

"Transition sequence" field:

as a *chosen* bond, in her own element.

These are not decoration. They are the first two payments on the "familiar accretion" through-line

the transitions file tracks at range scale (L24). **And they are precisely what a card cannot

show.** A still image plus one line cannot render a softshell turtle keeping pace with a hull, or

a starling working the mast on a west wind. A player standing at the rail sees both for free.

2.3 The structural rule this reveals

In this canon, **a voyage is playable content when it is a CHAPTER, and unauthored when it is a

SEAM.** The seam is the unit of narrative economy under the locked 77.

The proof that crossings are not considered boring is overwhelming — canon spends **six consecutive

nodes** on one:

nodesites (all water_body unless noted)functions
CH_33caravel_and_open_ocean_edge (+2 land)combat; learning; transformation
CH_35cape_verde_rise (+3 land/coastal)discovery; traversal
CH_36horse_latitude_calms, storm_passage, sargasso_edgelearning; character; investigation; puzzle / combat; transformation / discovery; vril_management
CH_37empty_quarter, convergence_approach, drowned_coasts, sargasso_gyre_centrelearning; character; vril_management / combat / puzzle; discovery
CH_38deck_at_the_midpoint, drowned_world_ridge (+1 sacred)discovery; character; puzzle; transformation

WGM §4 types Ch 36 *"Mid-Atlantic (The Sailor's Knowledge)"* — biome open ocean, key content

*"horse latitudes, trade winds, doldrums, navigation trade reaches its ancient-world maximum."* Ch 37

is *"No encounter, solitary sail, convergence visibility, last quiet before midpoint."* Both are bound

by Pillar 11 — no chapter coasts on one mode (CVD §4.5), so canon has already committed to making

weeks of open water carry combat, puzzle, learning, character and transformation beats.

Canon's answer to "can a sea passage be played?" is a six-chapter yes. The slice's crossings are

bare because they are seams, not because water is filler.

---

3. THE ETHOS QUESTION — IS A COVER MORE HONEST TO WALKED-GEOGRAPHY THAN A PLAYABLE LEG?

Less honest, on five independent grounds. Then one ground on which it is more honest.

3.1 A cover fails the CVD pillar the seam exists to serve

CVD §"Layer 1.5 (everyone)" — constitutional authority, wins every conflict:

*"Every player absorbs each culture's daily life, history, and landscape through immersion —
lived, walked, and played, never lectured. The world teaches by being inhabited, not by
exposition."*

A travel card is exposition about a passage. A deck is inhabitation of one. The seam is exactly where

"landscape" is the subject; a card is the one presentation form that converts the pillar's verb from

*inhabit* to *be told*.

3.2 A cover spends the Ch-55 transformation 52 chapters early

WGM §10 "The Three Major Geographic Transformations" names only three moments in a 79-node arc

where the game's geographic relationship fundamentally changes. One of them is:

"Ch 55 Atacama (Physical Transformation): Vimana activated at 33.33°S. Geographic scope shifts
from sequential to simultaneous. **Protagonist no longer constrained to walking or sailing — every
previously visited location becomes accessible."**

And WGM §7 gates the free-travel clause to Ch 59-65: *"Vimana enables free travel between all

locations visited."*

Canon has made **"you no longer have to cross the distance" one of three structural turning points in

the entire arc.** A cover at the Ch 3→4 seam hands the player the *felt experience* of instant arrival

in the tutorial band — the exact sensation Ch 55 exists to grant as a reward. That is not a

presentation nit. It is spending the arc's second-biggest structural payoff before the player has

walked far enough to want it. The walked-geography ethos is not decoration around the seams; **the

seams are where the ethos is either paid or defaulted on.**

PRESENTATION_DOCTRINE §4.1 already states the build-side consequence and reaches the same place:

*"A player who walked every kilometre should not be handed a progress bar at the one moment the

walking pays off."*

3.3 A cover must invent an authorial voice the canon does not have — and the doctrine bans it

The seam's content is already written, and it is written **in the first person, as a person's

intention**, in the player_objective column of beats.csv:

rope. Take your leave of people properly first."* (CH04_B12)

out of stone."* (CH05_B07)

A card converts a first-person intention into third-person narration over an image. That is precisely

the failure the Natural Voice Doctrine (Josh, 2026-07-26) rules against — *"Quest text is what

someone SAID or what the protagonist THINKS — never a system's summary… The quest log is a memory of

conversations and intentions, not an oracle."* A playable leg introduces no new voice at all; a

cover introduces a narrator the game otherwise does not have, at the seam, where the doctrine is

newest and least tested.

3.4 A cover is the seam's only presentation, and DT-3 forbids it from carrying meaning

DT-3 sends the story load to chapter_bridging_cinematic. Zero such rows exist (§2.1d). So a

cover-only seam is a seam whose authored content — Shell in the water, Kira on the passage, the first

water leg *as* the first water leg — has nowhere to live. A playable leg has a home that already

exists, is already gate-enrolled, and already carries the natural-voice column: beats.csv.

3.5 A cover is the highest-risk reveal surface in the seam; a playable leg is reveal-safe by construction

See §4 in full. The short form: HL_0043 bans the thesis "not in narration… not in UI," and

HL_0044 extends the pre-Ch-76 ban to "environmental storytelling, art direction, music." A card

is UI + narration + art direction + score — four named surfaces at once. A playable deck has no

narrator, so it can only ever show what a nine-year-old on a boat can see. **The playable leg is

reveal-safe by construction; the cover is the one form that has to be actively policed.**

3.6 Where the cover IS the honest option

The open-water middle. Nothing is staged there (§2.1a-e): no site, no thread cell, no seed, no

scene row, no age advance. Canon *declines to say* what happened in the middle of a multi-day coastal

sail. A rung-2 cover — the crossing continuing under a single held natural-voice line drawn from the

exit beat's own register — is a truthful presentation of a middle canon has left unwritten. Faking

elision is dishonest only where canon wrote something; here canon wrote nothing, deliberately, and

saying nothing back is faithful.

Verdict on the ethos question: a cover that fakes elision of *the whole seam* is LESS honest than

a playable leg, because canon did not elide the whole seam — it authored the boarding, the route, the

landfall, and a named companion beat on each crossing. A cover that presents only *the middle* is

MORE honest than inventing content canon withheld.

---

4. DOES THE NATURAL-VOICE TRAVEL CARD SATISFY THE ETHOS WHERE A PLAYABLE LEG IS NOT AFFORDABLE?

**Yes — and the affordability question is much narrower than it looks, because at these two seams the

canon-correct playable leg and the doctrine's strongest cover class are the same object.**

4.1 The canon-correct leg needs no vehicle system

T0_Vehicle_Registry [DRAFT v0.1] has three rows. The player's first watercraft is **VH_002

Coastal Skiff, acquisition_chapter_id = CH_08** (Swahili, deploys from the mothership line). And

T1_Trades_Master §3.10: Trade 10 Fishing and Navigation — ancient home base Ch 08 Swahili Coast;

its infrastructure is *"trade master NPCs at home base chapters, navigable boats and ocean access, and

field mechanics for ocean-substrate perception"* — none of which the player has at Ch 3-5.

So at both slice crossings the player is canonically a passenger/crew on an NPC's boat — which

CH04_B12 says outright. The leg therefore requires:

No sailing sim. No boat handling. No navigation trade. No vehicle system. That is exactly

PRESENTATION_DOCTRINE §4.2's strongest cover class:

*"THE CROSSING. A passage the player is genuinely making — the ship's deck, the river boat, the
mountain pass, the descent. Loopable by construction (the crossing simply continues), which is
what makes it the strongest class."*

And it satisfies the binding cover rules without strain: DT-1 (loops by construction — the

crossing simply continues), DT-2 (*"where the player can keep control during a cover (standing on

the deck…) the player keeps it"*), DT-5 (the cover cannot end until the destination is ready — a

deck can hold indefinitely).

At this point the cover/playable-leg distinction collapses. A walkable deck IS the rung-1 diegetic

cover AND the playable leg. Josh's challenge — *"why does it need a map change cover instead of a

playable loading transition?"* — is answered by canon in the affirmative, and the doctrine's own

taxonomy already names the answer; the doctrine simply had not noticed that its best cover class is

not a cover in the pejorative sense at all.

4.2 Where the card still earns its place — and what it costs

The card (rung 2 / rung 3) remains honest and necessary for:

returns SamosirToba only) and will ship uncovered before they ship covered;

recommendation of Option 1 with a quality floor.

Its authoring cost is zero new voice. Rung 2 already specifies *"a single natural-voice line…

drawn from the exit beat's own diegetic register."* Those lines already exist in

beats.csv:player_objective for every seam in the slice (§3.3). The card should lift, never

compose. That both satisfies the Natural Voice Doctrine by reuse and removes the §3.3 objection: a

line the protagonist already thought is not a narrator.

Its ethos cost is real but bounded, provided it (a) never covers a leg canon authored, (b) never

dates the passage, (c) never carries the seam's named content (Shell's first water leg, Kira's

passage) — those belong to a beat.

---

5. THE REVEAL SURFACE — WHAT A TRANSITION MAY NOT SHOW PRE-CH-38

Binding at these specific seams. Sources: TRANSITIONS_CH02_21.md §"Governing constraints" +

T0_Hard_Lines [ACTIVE v0.2] + the per-beat reveal notes.

5.1 The family / pattern line (the headline constraint — and it is slack HERE)

*"Reveal discipline binds every seam. **No House-of-Velheim name, family, or organized pattern
surfaces to the player anywhere in this range** — the family first appears only as an individual
vessel at Ch 8, and the pattern itself is not surfaced to the player until the mid-game's end and
the modern chapters (about Ch 38). The endgame reveal-figure (Ch 76) and the mythological-realm
names stay dark. Antagonist recognition is carried as unexplained unease, never as a named pattern."*

At the slice's two seams there is no payload yet to leak. Ch 3's land-pressure broker predates the

family entirely (CH_03.md §"Boss structure" reveal binding); Ch 4's Dalang apparatus is

institutional-proxy only. The first reveal-carrying seam is Ch 5→6, whose Continuity line reads:

*"the first family-briefed recognizer's unease (Datu Siahaan) carried forward as an unexplained

exposure vector, never a named pattern."* So the slice's transitions are the *easiest* reveal case in

the range — which is exactly why a habit set here (a narrator, a montage, a lore rail) will be

invisible until it becomes dangerous at Ch 5→6 and lethal by Ch 8.

5.2 HL_0043 — the surface the card IS

*"The thesis is never declared explicitly anywhere — **not in dialogue, not in narration, not in
marketing copy, not in tutorial text, not in UI**. The Grand Sage Ch 76-77 reveal is the single
canonical direct surfacing of the thesis in the 77-chapter arc."* — HL_0043 (CVD §17.9), scope
universal, consumers include UE5 / Quest / Region / Spine / Cinematic.

A loading screen is UI. A held card is narration. This is the one hard line whose *named surfaces*

the cover form occupies by definition.

5.3 HL_0044 — art direction and music are named

*"Pre-Ch 76 Grand Sage identity hints are forbidden… Reveal is not stated or foreshadowed in any
in-game content before Ch 76-77 across dialogue, journal entries, inscription content, NPC
exposition, environmental storytelling, art direction, music."* — HL_0044, scope Ch1-75.

A cover's still, its palette, and its cue are three of those surfaces at once.

The specific live hazard at the Ch 4→5 seam: CH05_B01 (the very next beat after that crossing)

establishes the environmental waterfall cue at Borobudur — *"beneath the great monument water

falls the way it fell at the first cave"* — with the reveal note: *"the environmental waterfall cue

established as pure sensation, no reveal spoken and no reveal-figure named, HL_0043"* (payoff at

Ch 76-77; WGM Ch 5 key content: *"Waterfall reveal cue setup at Borobudur… final reveal Ch 76-77"*).

**A Bali→Sumatra travel card carrying water, falling-water, or cascade imagery would pre-frame the

arc's single most protected cue, one beat before canon plants it as sensation.** This is the sharpest

concrete reveal risk the fork carries, and it exists only for the cover form.

5.4 The perception ceiling

HL_0028 / HL_0017: substrate-perception unlocks at Adeptus Minor (Ch 15); Vril Sight is tier 6.

Every Ch 3-5 beat carries reveal: HELD(… at the Neophyte register …). **No transition may render a

substrate/vril-signature/polarity view** — no glowing-ley-line map wipe, no "energy" transition art,

no route drawn on a world map as a vril trace. A route line on a map is also a *scope* leak (it shows

the player geography they have not walked).

5.5 Age and elapsed time

TRANSITIONS_CH02_21.md L17: *"The protagonist's internal age is designer-only and never surfaced;

the passages carry it forward but do not announce it."* No "three weeks later," no "a season at sea,"

no dated caption. Both slice crossings carry no age advance anyway (§2.1e).

5.6 Concealment state (binds the playable leg too)

CH_03.md seam state-diff: *"ws.identity_hiding_status: ACTIVE_CONCEALMENT, **pressure HIGH at the

Ch-4 entry** (easing to MED across Bali's mask-normalized society)."* CH04_B12 writes the ease to

MED only on *departure from Bali*.

So: on the Flores→Bali crossing the protagonist is at HIGH concealment pressure. Any transition

art — cover *or* playable leg — showing the child's face and ears open on a crowded deck contradicts

committed ws state. The cloth stays on. (This is a constraint the playable leg must respect, not an

argument against it; it is also a *gift* — a hooded child at a rail among strangers is the seam's own

tension, free.)

5.7 The Child Familiar

TRANSITIONS_CH02_21.md L24 + every Ch 2-5 reveal note: the Child Familiar is *"lost at Ch 2,

believed dead by the player yet canonically escaped, its reveal held to Ch 58 on the Truly-Good

path."* No transition may show, echo, or hint the escape — and a "familiars riding the passage"

cover shot is exactly the composition where an over-helpful artist adds a fourth silhouette.

5.8 Consolidated: the transition's reveal contract

A transition at these seams MAY show: the boat, the captain, the water, the coast, the weather, Shell

in the water, Kira on the air, the hooded child, the cargo, the crew, the port.

It MAY NOT: name or pattern the family; narrate a thesis; carry Grand-Sage-adjacent art direction or

score; use falling-water imagery immediately before CH05_B01; render substrate perception; draw a

route on an unwalked map; state elapsed time or age; expose the protagonist's features at HIGH

pressure; hint the Child Familiar; carry mode taxonomy or internal vocabulary (DT-4, and the standing

no-mode-taxonomy-on-player-surfaces ruling).

---

6. FINDINGS FOR THE DIRECTOR (this axis only)

1. Canon does not elide the slice's crossings; it elides their middles. The boarding

(CH04_B12, a typed traversal beat with the captain as actor), the walk to the coast

(CH03_B12), the landfall "by river and road" (CH05_B01), and a whole route

(CH05_B07 + the kra_traverse site) are all authored. Only the open water is blank.

2. Canon already cast the player as crew, not cargo. CH04_B12.player_objective: *"he will take

a passenger who can hold a rope."* The playable-leg premise is not a proposal — it is shipped

canon text in the natural-voice column.

3. Each slice crossing has one named, non-fungible piece of content a card cannot render: Shell's

first water leg as a chosen bond (Ch 3→4), Kira riding the passage as the roster accumulates

(Ch 4→5). These are the first two payments on the range-scale familiar-accretion through-line.

4. The belt's longest leg is already a SITE. kra_traverse — functions [traversal, dialogue],

port audiences, tribute politics, an L2 inscription found en route. Canon's own answer for the

belt's biggest hop is a place, not a card.

5. The ethos verdict: a cover over the whole seam is less honest than a playable leg — it fails

CVD Layer 1.5 ("inhabited, not exposition"), spends the WGM §10 Ch-55 transformation 52 chapters

early, invents a narrator the Natural Voice Doctrine bans, and occupies four HL_0043/HL_0044

surfaces at once. A cover over the *middle only* is honest, because canon wrote nothing there.

6. The affordability objection is thinner than assumed. At Ch 3-5 the player owns no boat

(T0_Vehicle_Registry: first watercraft VH_002 at CH_08) and has no navigation trade

(T1_Trades_Master §3.10: Trade 10 home = Ch 8). The canon-correct leg is a walkable deck

which is simultaneously PRESENTATION_DOCTRINE §4.2's "THE CROSSING" cover (loopable by

construction, DT-1/DT-2/DT-5 satisfied) and the playable transition. **The two options are the

same object at these seams.**

7. The card's honest role, at zero voice cost: rung-2/3 fallback for the open-water middle and

the uncovered belt hops, lifting its line from the exit beat's existing player_objective

rather than composing new narration.

8. FLAG — game/canon arrival mismatch. MapForChapter("CH_05") = SamosirToba, called "the CH_05

ARRIVAL map" in TODO_CANON. Canon's Ch-5 arrival is Borobudur / Central Java (CH05_B01,

seq 1); Samosir is beat 5. The CH_04→CH_05 seam currently lands the player at the chapter's fifth

site.

9. FLAG — belt route shape. The staged belt reads samosir → palembang → borobudur → angkor.

Canon's route is `borobudur → palembang → samosir → [kra_traverse] → angkor → bayon → baray →

my son → back to samosir (climax) → palembang (exit)` — seven legs with two returns, with

local.doubled_back: true as a gating state on CH05_B12.

10. FLAG — DT-3's designated home does not exist. Zero chapter_bridging_cinematic rows across

all 79 nodes. Until that changes, the doctrine forbids the cover from carrying seam meaning and

provides no alternative carrier — which is itself an argument for putting the content in a beat.

11. The sharpest concrete reveal risk in this fork is cover-only: a water/falling-water motif on

the Bali→Sumatra card would pre-frame the Borobudur waterfall cue (CH05_B01, HL_0043-noted,

payoff Ch 76-77) one beat before canon plants it as pure sensation.

---

7. METHOD NOTE — POSITIVE CONTROLS RUN

Per the standing a-search-that-cannot-match-reports-zero rule, every absence-claim above was proven

by first showing the search matches known-present content:

absence claimedsearchpositive control (matched)slice result
no at-sea site at Ch 3-5`ocean\open_water\shipboard\deck over build/spine_rows/*/sites.csv`CH_33 caravel_and_open_ocean_edge; CH_38 deck_at_the_midpoint0
no voyage thread cell at Ch 3-510 maritime keywords over T0_Thread_Grid8 of 17 rows at CH_33/36/370 of 39
no bridging cinematicchapter_bridging_cinematic over all scenes.csv(n/a — 0 arc-wide; the grep itself was verified against boss_encounter_resolution / per_chapter_cutscene, which match)0 of 79 nodes
no voyage seedmaritime keywords over the 400-row housing ledgermatched The Ship That Carries the Teeth, Charon's Crossing, galleons of dream (all region/fairy-shelved)0 seam-keyed

Distances in §1 are my derivation (great-circle from WGM published coordinates), not canon text,

and are labelled as such wherever used.

Generated by harness/site/structure_site.py — the URL path is the repo path. review root