SLICE_SCOPE_CASE_2026-07-27.md

pipelines/SLICE_SCOPE_CASE_2026-07-27.md

The slice-scope question — THE EMOTIONAL / COMMERCIAL CASE

Analyst axis: what an Early Access buyer must FEEL for the game's promise to land, which authored beats

arrive PLAYED vs TOLD per shape, and what each shape does to the Josh Gate. Read-only pass; nothing here

is applied. Every claim carries its canon anchor. Proposal-tier, presented for the director's synthesis.

Grounding read this pass: docs/proposals/AMULET_PROVENANCE_BRIEF.md (the full integration plan for

Josh's third-sitting ruling), docs/spine/CH_PROLOGUE.md, docs/spine/CH_01.md, docs/spine/CH_02.md,

docs/spine/CH_03.md, docs/spine/CH_05.md, docs/spine/CH_13.md, docs/ROADMAP_POST_5090_TO_SHIP.md,

docs/proposals/REALM_DESIGN_PROGRAM.md §9, docs/proposals/FINALE_FAIRY_ENRICHMENT.md,

docs/PRE_5090_BUILD_PLAN_VOL2.md, the ship-rulings + Josh-Gate memory records, and the built map set in

C:/dev/Humanity/Humanity/Content/Maps (FloresIsland, Bali, BorobudurMerapi, PalembangMusi, SamosirToba,

AngkorKhmer — six maps, all Ch 2-5 regions; nothing for Ch 6-13 exists).

---

0. The direct answer to the question inside the question

"…all the way through giving the amulet to the healer's child (is this ch 13?)"

Half yes, and the half that is "no" is the load-bearing half.

• The JOIN is Ch 13. Ntala (CHAR_0002) steps from unsigned watcher into active traveling companion at

Ch 13, and the evidence-journal begins in that child's hand — a canonical event bound by HL_0070

[SRC: docs/spine/CH_13.md # Continuity anchor] [SRC: T0_Hard_Lines @ HL_0070].

• The GIFT is not assigned to any chapter. Josh's own ruling left it explicitly open: "the gift's

staging beat at Ch 13 or later is an OPEN integration point" [SRC: AMULET_PROVENANCE_BRIEF.md §1].

The brief's five candidate stagings run Ch 13 → the first real scare → the Ch-38 journal transfer →

before the Realm War → player-choice, and its standing recommendation is **the first real scare, with

Ch-13 as fallback** [SRC: AMULET_PROVENANCE_BRIEF.md §7].

**Consequence: the scope question and the gift-staging question are COUPLED, and the coupling runs one

way.** "Prologue → the gift" is a Ch-13 scope *only if Josh rules the gift to the join*. Under the

standing recommendation (first real scare) the boundary floats past 13 with no fixed endpoint — which

means the scope cannot be costed until the gift is staged. Ruling the gift to Ch 13 is the cheapest thing

Josh can do to make any of these shapes decidable, and §7 below argues it is also, under Early Access,

the artistically correct staging for a reason the brief could not have known when it was written.

A second correction to the question's framing, offered because it changes the answer: the healer's child

is already inside the current slice. They begin leaving unsigned marks where the protagonist sleeps at

Ch 3 — "the arc's first human-character mystery and the primary human companion who active-joins at Ch 13"

[SRC: docs/spine/CH_03.md @ CH03 beat 159, # Key ally NPCs]. The relationship the gift pays off is

*planted* in the shipped slice and *joined* eight chapters past it. That asymmetry — planted here, paid

there — is the actual shape of the problem, and it is not the same problem as the Prologue.

---

1. The frame: what does an EA buyer need to FEEL?

State the bar honestly before grading shapes against it, because the obvious answer is wrong.

The thesis cannot land in any slice, by construction. Nothing of the Grand Sage or the game's thesis

may be shown, hinted, or foreshadowed before its Ch-76/77 window (HL_0116 / HL_0046 / HL_0043), and the

Prologue and Ch 1 are the *most* constrained nodes in the arc on exactly this point — show-don't-tell is

"ABSOLUTE" on both entries [SRC: docs/spine/CH_PROLOGUE.md header] [SRC: docs/spine/CH_01.md header].

So adding the Prologue does not bring the thesis one inch closer to an EA buyer. Any argument for

scope expansion that rests on "the thesis needs to land" is arguing against canon. The Prologue buys three

other things and only three: the CEILING (what was lost), the AMULET'S PROVENANCE (why the gift costs),

and the BOND (why the Ch-2 opening hurts).

What an EA buyer actually needs to feel, in the order they need to feel it:

1. Within ~10 minutes — "this world is real and it is not like other games." The differentiator is the

real world rendered as a place its own peoples built and knew: Liang Bua's stratigraphy, the Manggarai

lingko fields, the polo named by the Nage and Lio in their own register [SRC: docs/spine/CH_02.md

# Civilizational marks]. This is the promise, and it is 100% deliverable in Ch 2 alone.

2. Within the first hour — "I am small, stripped, and climbing." The vril-depleted floor, the first

waterfall recharge learned through the hands, not a tutorial box [SRC: docs/spine/CH_02.md @ CH02_B04].

Fully deliverable in Ch 2.

3. By the end of the purchase — "something was taken from me, and I am owed a return." This is the

engine of the whole 79-node arc (Thread 6, dying-rising, anchored at the Ch-1 fall)

[SRC: docs/spine/CH_01.md # Threads at origin]. **This is the one an EA buyer of Ch 2-5 receives as

an assertion rather than an experience** — see §3.

4. By the end of the purchase — "I am not alone anymore," or at minimum "I have been noticed." An

ending. The current slice has neither; see §4.

5. Standing throughout — open questions I want the next drop to answer. Ch 2-5 is genuinely rich here

(the familiar's fate, the unsigned watcher, the sealed letter at Ch 8 — outside the slice, the amulet).

Items 1, 2 and 5 are already met by the built slice. Items 3 and 4 are the two real gaps, and **they are

different problems with different cures** — which is the finding the four shapes obscure.

---

2. The two gaps, named precisely

Gap A — THE OPENING GAP (structural; expansion forward does not fix it)

The slice's very first beat is a payoff beat whose setup lives in Ch 1.

CH02_B01, the first thing a buyer plays: *"Wake on cold stone in the dark of Liang Bua after the rift-fall;

the Child Familiar bites once, bolts, and is watched dying from a crack in the rock — arrival as loss, the

first morning of a life measured from a wound"* [SRC: docs/spine/CH_02.md @ CH02_B01].

What Ch 1 authored to make that land: the Child Familiar is one of the four player choosings — chosen

from the twelve royal-communion species, and Josh's authored prose says *"it chooses back, and after that

it is simply always there, a warmth at the child's side that has a heartbeat and a mind and a way of being

glad"*, under the explicit design rule *"None of it is a menu. All of it is a life"* [SRC:

docs/spine/CH_01.md # Body prose] [SRC: T3_Core_Characters § 2.2].

So the shipped opening asks a buyer to grieve a creature they never chose, never bonded with, and never met.

And it goes further than bereavement: **the three-path fork of Ch 2 is keyed on how the player reads the

bite as a BETRAYAL** — *"on the generative path the betrayal stays a wound carried whole; on the middle path

it is swallowed unfelt; on the corrosive path it hardens into the first small permission to distrust"*

[SRC: docs/spine/CH_02.md # Body prose]. Betrayal requires a prior bond. **The integrity system's first

fork ships with its emotional premise absent.** That is not a polish note; it is the corrosion axis opening

under-motivated in the first ten minutes of the game.

Expanding the slice to Ch 13 does not touch this. Gap A is behind the slice, not ahead of it. This is

shape (b)'s blind spot and it is worth stating plainly: *Ch 2-13 fixes the ending and leaves the opening

exactly as broken as it is today.*

Gap B — THE ENDING GAP (presentational, and much cheaper than it looks)

Ch 5 does not end anything. Its close: *"doubles back down the whole long route to the sea — carrying

north-country knowledge the way they carry everything, filed and unnamed… Somewhere beneath the great

monument, water falls the way it fell at the first cave, and the child always stops to listen"*

[SRC: docs/spine/CH_05.md # Body prose]. Beautiful, and it resolves nothing. An EA buyer finishes and

feels the game *stopped* rather than *paused*.

But Ch 5 already contains, in canon, a first-rate cliffhanger that is currently buried mid-chapter: the

corrupt datu is the first antagonist who reads back — *"Every corrupted vessel before him was blind to

what the protagonist truly is; this one has been briefed — told, by powers he only half-understands, to

watch for exactly the kind of wrongness their vital-force gives off… that recognition is a new kind of

danger the concealment they have practiced cannot answer"* [SRC: docs/spine/CH_05.md # Body prose].

"You have been noticed by something that briefed him" is a shipping-grade EA curtain line, it is

already canon, it is reveal-safe (no name, no family, no pattern — the datu himself only half-understands),

and staging it as the slice's closing note is a presentation decision, not a content-scope decision.

Therefore: Gap B does not justify expanding to Ch 13. It justifies re-staging the Ch-5 close. That

single observation removes the strongest emotional argument for shape (b) — and it is the argument most

likely to have been driving the instinct that the slice "has to be longer."

---

3. PLAYED vs TOLD — the authored arc, beat by beat, per shape

The authored emotional spine of the third sitting, and where each beat actually lands. P = played,

T = told/implied, A = absent (not even referenced), H = lands later as a hindsight recast of

something the player already played.

\* Gap-A dependent — see §6, the minimum-viable-setup question, which can move this cell without the realm.

Read the table by column, not by row.

• (a) delivers the world-promise and the survival engine at full strength and delivers zero of

Josh's newly authored emotional spine, in any register. The amulet is a worn unexplained object; the

guilt engine does not exist.

• (b) buys exactly two rows over (a): the join+journal, and a gift that arrives **stripped of the only

thing that makes it mean anything.** In (b) the gift reads "I gave away a nice thing I've always worn."

Josh's authored line — *"the protection that cost the mother her life, given away to protect someone they

love"* — is 0% delivered. **This is the honest answer to whether expanding to 13 makes the amulet chain

land: it does not. It only makes the object change hands.**

• (c) is the only shape that delivers the authored arc as authored, and §5 costs what that purchase

actually requires.

• (d) delivers everything (b) does on the same forward schedule, and converts every A in the Prologue

block into an H — a hindsight recast — which §7 argues is not a consolation prize.

---

4. Where the Ch-2 opening stands without the Prologue: the in-media-res case, worked honestly

The instinct that "great games open unexplained" is correct and it is being applied to the wrong beat.

The distinction that decides this: an unexplained opening works when the missing information creates a

QUESTION. It fails when the missing information is supposed to have created a FEELING. You cannot

retroactively install a feeling in a player who was not there; you can absolutely install a question.

• Dark Souls, Hollow Knight, Outer Wilds, Elden Ring open with the player knowing nothing — and each asks

the player to feel curiosity and unease, both of which are generated in real time by what is on

screen. None of them opens by asking you to mourn a specific named creature you have not met.

• Ch 2's opening asks for a bereavement (arrival as loss) *and* a betrayal-read (the three-path fork).

Both are relational. Both require a prior relationship. Neither is a mystery-engine beat as authored.

But the beat can be re-pointed onto the channel that does work, and canon already supplies the material.

1. The bite is an anomaly, not just a tragedy. The Child Familiar *canonically escaped and is alive*;

the death is a player-belief, held from every surface artifact until the Ch-58 Truly-Good reveal

[SRC: docs/spine/CH_02.md # Continuity anchor, @ CH02_B01 reveal binding]. So the bite is genuinely

strange — a creature that bites its bonded child once, hard, and runs. Staged as anomalous rather

than merely fatal, the beat generates the question ("why did it do that?") the in-media-res channel

can actually carry, and it does so without leaking a single frame of the Ch-58 reveal.

2. The amulet is a free Chekhov's gun and it costs nearly nothing. It crosses the seam because it is

worn, and it is unreadable and unidentified on the far side under HL_0095; the CH_02 continuity anchor

needs one clause reconciling "a worn thing, not inventory" [SRC: AMULET_PROVENANCE_BRIEF.md §2.3,

§10]. Make it visibly worn and remarked-on-but-unexplained from the first minute of the EA build.

An unexplained object the player has worn since beat one converts the missing origin from an ABSENCE

into an ACTIVE MYSTERY — and it pays off twice, at the gift and again at the Prologue drop.

3. The child's condition carries what the bond cannot. Cold, dark, no fire, nothing eaten, and the

faint unexplained lightening of the air on the worst night [SRC: docs/spine/CH_02.md # Body prose].

This is a survival-empathy channel and it works on a stranger. It is a real and sufficient opening.

It is simply a *different, quieter* opening than the one Josh authored.

What CANNOT carry the backstory in shapes (a)/(b) — stated plainly, because the question assumed it can:

• The townspeople's grudge is realm-side and cannot appear in an earthly chapter at all. It is a Ch-1

fairy-realm beat about the realm's own people [SRC: AMULET_PROVENANCE_BRIEF.md §6]. There are no

townspeople in Flores, Bali, Java or South Africa who know the protagonist's mother died. The grudge

ships with the Prologue/Ch-1 or it does not ship.

• Dialogue surfaces are structurally blocked by the concealment mechanic. Ch 2 runs

identity_hiding_status FIRST_CONCEALMENT, escalating toward HIGH as village contact approaches

[SRC: docs/spine/CH_02.md # Entry header]; by Ch 13 it is

CONCEALMENT_WITH_HOSTILE_RECOGNITION [SRC: docs/spine/CH_13.md # Entry header]. **The protagonist's

core mechanic is not telling anyone what they are.** A told-backstory delivered through NPC dialogue

would require the protagonist to violate the arc's central discipline in its opening chapters. This is

the sharpest structural finding in this section: **shapes (a) and (b) cannot TELL the backstory even if

we wanted them to.** The only legal carriers are interior register, the worn object, and — from Ch 13 —

the journal.

• Flashback is the wrong tool and is dangerous here. A memory cut of the mother's death front-loads a

proximate emotional cause onto the fall, and the single sharpest reveal hazard in the whole integration

is that the fall must read as *accident, never omen or consequence*

[SRC: AMULET_PROVENANCE_BRIEF.md §9]. A flashback is also a cheap-looking first pass at the game's

most expensive art surface — see §5. Recommend against, in every shape.

Verdict on §4: the Ch-2 opening stands up *as a mystery-and-survival opening* without the Prologue,

and it does so at real quality, provided the bite is staged as anomalous and the amulet is worn and

unexplained. It does not stand up as the *bereavement-and-betrayal* opening it was authored to be.

That authored version is unrecoverable without Ch 1, at any slice length.

---

5. What shape (c) actually buys, and what it charges — the collision worked, not waved

Shape (c) is the only shape that delivers Josh's authored arc as authored. On the emotional axis it is the

clear winner, and it should be argued at full strength before it is argued down:

The case FOR (c). The four choosings land as a life, not a menu. The familiar's death lands as the

betrayal the fork needs. The ceiling is felt, so "stripped" means something in the body rather than in a

tooltip. The mother's death lands, so the amulet has weight for every hour the player wears it. The gift

at 13 closes the guilt engine on schedule. A buyer would receive a *complete emotional movement*: given

everything → lose it → fall → survive alone → find one person → give away the last thing you have of home.

That is a better product than anything the other three shapes describe. **No honest analysis should

pretend otherwise.**

The charge, in the currency this axis cares about — first impressions, which are spent once.

• It inverts Josh's same-day ruling at the point where the ruling is load-bearing. The realm program

is explicit: the mythological realms and the fairy realm are *"the grandest displays of the artistic

capability"*; the fairy realm builds last, after all mythological realms, with the Prologue and Ch 1

built last of all [SRC: docs/proposals/REALM_DESIGN_PROGRAM.md §intro, § the Prologue row]. The reason

is craft sequencing, not scheduling convenience: the establishing art board rules the realm grammar,

and the program pilots that grammar on Atlantis and the Duat *before* the fairy realm inherits it.

Building the fairy realm first means building it without the grammar it is supposed to crown — i.e.

building the best thing in the game worst.

• It spends the crown jewel's first impression at the lowest bar it will ever be at. This is the

commercial version of the same point. The fairy realm is the marketing asset with the highest ceiling in

the entire property. Shipping it as the *first* generated realm, in the *first* EA build, against a

grammar nobody has piloted, is spending it. First impressions do not refund.

• It is a second full playable kit for one chapter. The Prologue is playable as the Elven King at the

sovereign ceiling — Sovereign Vril Projection, the keeping of the veil, Temporal Sight, Familiar

Communion [SRC: docs/spine/CH_PROLOGUE.md # Entry header; T3_Core_Characters §6.4]. That is not the

protagonist's Neophyte kit with numbers turned up; it is a different verb set, built once, used once,

gating the launch of everything else.

• It requires a character creator, twelve familiar species, and a best-friend NPC (the four choosings)

[SRC: docs/spine/CH_01.md # Entry header] — plus Josh's newly authored beats: the realm-service

sandbox, the catastrophe, the playable run-back, the death scene, and an amended chapter tail

[SRC: AMULET_PROVENANCE_BRIEF.md §4.2].

• It has no terrain chain. Every earthly region rides the deterministic GIS → Gaea → WorldCover →

PCG pipeline that is already generalized and proven [SRC: ROADMAP_POST_5090_TO_SHIP.md P2.1;

PRE_5090_BUILD_PLAN_VOL2.md W-SPACE]. The fairy realm rides none of it. It is 100% invention, and

its ruleset registry, harmony doctrine and art-board pipeline are at PROPOSAL with three forks still

pending Josh [SRC: docs/proposals/REALM_DESIGN_PROGRAM.md DOC_MAP row].

• It moves the amendment surface into hard-line nodes. Both the Prologue and Ch 1 are in the

eighteen-node canonical hard-line set with show-don't-tell ABSOLUTE, and the integration touches ~18

files including registry rows, hard-line-bound scene rows, and T3 characterization

[SRC: AMULET_PROVENANCE_BRIEF.md §10]. Every one of those needs the full critic + reveal-discipline

treatment before it can be built, let alone polished to the AAAAA bar.

• The program's own §9 already ruled on this collision. Where the objection lands: *"the four later

lenses and the volume of board work genuinely should not compete with the Ch 2-5 slice"*

[SRC: docs/proposals/REALM_DESIGN_PROGRAM.md §9]. Shape (c) is that competition, maximised.

Verdict on §5: (c) wins the emotional argument and loses the *first-impressions* argument, which is

the same currency measured over a longer horizon. Recommend against, with the specific reason stated in

Josh's own terms: **the fastest way to make the fairy realm less than the grandest thing in the game is

to build it first.**

---

6. The minimum-viable-setup question (the shape the four options do not contain)

Because Gap A is structural and Gap B is not, the useful question is not "how much further forward?" but

"what is the least that must sit in front of Ch 2 for its opening beat to work?" Four rungs, cheapest

first. This is a presentation-tier question, not canon — the four selections and what they set are canon

either way [SRC: T3_Core_Characters §2.2].

1. Nothing (pure in-media-res). Re-point the beat onto mystery + survival empathy per §4; wear the

amulet unexplained; stage the bite as anomalous. Cost ≈ 0. Recovers the question channel, not the

bereavement channel. The bereavement re-reads at full strength when the Prologue drops.

2. The four choosings as a front-end creator. The player picks the familiar from the twelve, the face,

the name, and the best friend before Ch 2 loads. **Recovers most of the bereavement payload — the

creature that bites and dies is one you chose** — at the cost of the design rule Ch 1 exists to enforce

(*"None of it is a menu. All of it is a life."*). A creator is needed eventually regardless, so much of

this cost is brought-forward, not new. **The honest downside: EA players' first-play memory of the

choosings is permanently the menu version**, and the lived version arrives as a replacement rather than

a first experience. Director-level call; worth a brief of its own.

3. A cut-down playable Ch-1 childhood vignette (one small realm space, no kingship). Recovers nearly

the whole bond payload. Argue this DOWN hard, and for a non-obvious reason: it establishes the

fairy realm's visual grammar at a reduced bar, and the real Prologue then has to *match or exceed a

look already shipped*. It is the worst of both — it spends the first impression AND does not deliver

the ceiling. Rung 3 is a trap.

4. The full Prologue + Ch 1 = shape (c), §5.

On this axis the choice is between rungs 1 and 2, and 1 is the safer default with 2 as a live upgrade.

Rung 1 preserves the lived-choosings experience for the eventual drop; rung 2 buys the opening beat back

now and spends that experience. Both are compatible with every shape below.

---

7. The gift as a slice ending, and the ordering accident that turns out to be canon

Ch 13 as the act-break. If the base slice ended at Ch 13, the emotional shape of the purchase would be

a genuine and complete movement, and it is worth naming because it is the strongest thing shape (b) has:

Ch 2 opens with *"the only warmth left in the world"* dying [SRC: docs/spine/CH_02.md # Body prose].

Ch 13 closes with *"For the first time the protagonist has a witness who is also a keeper"*

[SRC: docs/spine/CH_13.md # Body prose].

Alone → witnessed. That is a real act, its two ends rhyme, and it needs no fairy realm. If the slice

must be longer, Ch 13 is the correct place to stop — not because of the gift, but because of the join

and the journal. Any expansion that stops at 6, or 8, or 10 buys length without buying an ending.

But the gift itself is a weak act-break in shape (b), and this is the part worth being blunt about.

The gift's authored payload is *entirely* provenance. Without the Prologue, giving the amulet is a kind

gesture between two children. With it, it is the protagonist knowingly handing away the thing that cost

their mother her life. Shape (b) ships the gesture and withholds the meaning — and worse, it *spends

the beat*: once the object has changed hands, the later Prologue drop can only recolor a memory, never

stage the giving.

Which turns out to be exactly right, and this is the key finding of the whole analysis.

The arc's own doctrine is that the origin is built last and read first, and that it *"recast[s] the

whole arc without ever hinting"* — Ch 1 *"reads as innocent childhood on first play, and only in HINDSIGHT

reframes the whole arc"* [SRC: docs/proposals/FINALE_FAIRY_ENRICHMENT.md; docs/spine/CH_01.md header].

Under Early Access, chapters are consumed in release order, so an EA player receives the Prologue after

the journey — and a Prologue received after the journey does precisely the thing canon says the Prologue

does: it recasts everything already played.

**The EA ordering does not break the amulet chain. It performs the arc's own signature move at companion

scale.** Every EA player who plays the Prologue drop re-reads their own gift: *the thing I handed to Ntala

without knowing was the thing that killed his mother.* That is not damage control — it is the same

mechanism as the Ch-77 reveal, one order of magnitude smaller, delivered free by the release schedule.

Two readings, both canonical:

• Post-EA (1.0) buyers get the designed order — Prologue → Ch 1 → Ch 2 → … — and the gift lands with

full charge on first play.

• EA buyers get the hindsight order and the recast. Different, and *legitimately* good.

The sequencing constraint this creates, stated honestly: under a strict build-last ruling the Prologue

ships after Ch 77, so **the gift will land unprovenanced on first play for EA buyers no matter where it

is staged** — Ch 13, Ch 38, or before the Realm War. There is no staging that escapes it. The choices are

therefore (i) accept the hindsight recast, which is what the arc does anyway, or (ii) move the Prologue

earlier in the *drop* order than in the *build* order — e.g. after the first mythological realm has proven

the grammar rather than after all of them. **(ii) is a genuine amendment to Josh's ruling and belongs in

its own brief; it is not this analysis's to assume. Under (i), staging the gift at the Ch-13 join** is

now the better choice than "the first real scare," reversing the brief's standing recommendation for a

reason the brief could not have had: at the join the gift is *inside the base slice*, it gives the act-break

an object as well as a moment, it closes the coverage gap the brief worried about (a non-combatant on live

fields with no ward), and the recast has the longest possible runway to land on.

---

8. The Early Access reality: precedent, the first hour, and the marketing problem per shape

8.1 Precedent — the honest finding

**The precedent class for "ship a narrative game without its opening and add the opening later" is close

to empty, and this must not be glossed.** The dominant EA/episodic pattern is the exact inverse: Baldur's

Gate 3 shipped Act 1 alone for years; Kentucky Route Zero released Act I first over a seven-year run;

Telltale and Life is Strange shipped strictly in order. Hades — the closest useful analogue — shipped the

loop first and landed large amounts of narrative, characters and its true ending across EA, training its

audience to expect *story to arrive*, and its players re-read earlier runs in light of later reveals. That

is the mechanism shape (d) relies on, but Hades was never missing its beginning.

Implication: shape (d) is not derisked by precedent, and the risk has a specific shape — a review

sentence, written at EA launch and permanent in the aggregate: *"it starts in the middle and never explains

why."* EA reviews do not expire.

The mitigation is framing, and Josh's existing store identity is unusually well suited to carry it.

The always-on-sale, AAAAA, honest-about-the-tooling posture [SRC: ship-rulings memory; D-PRICING] is a

*declare-the-strategy* brand. Declaring the build order on the store page converts the weakness into a

roadmap hook:

*Early Access begins at Chapter 2, where the story's protagonist begins: alone, stripped, on a real

island. The Prologue and Chapter 1 — the fairy realm — are the last thing we will build, because they

have to be the most beautiful thing in the game.*

That sentence is true, it is on-brand, it turns the deferral into an argument for the ceiling, and it gives

the EA roadmap a headline finale to point at for its entire run. Undeclared it is a defect; declared it is

a wishlist driver. Recommend declaring it, in whichever shape ships.

8.2 The first hour and the trailer — the counterintuitive finding

The differentiator no competitor has is the real world rendered as its own peoples' achievement. The

Ch 2-5 trailer is Liang Bua's dark, Mount Inerie's cone country, Kelimutu's three colored lakes, Balinese

temple country, Borobudur's terraces, Toba's caldera, Angkor's five towers and its mirror-lake — none of

which any competitor is shipping.

A fairy-realm opening is beautiful, and it is also **the most contested aesthetic real estate in the

market. On the marketing axis specifically, opening the game on Flores is a stronger commercial

position than opening it on an elven kingdom** — and shapes (a)/(b)/(d) do that by construction. This

argues that the pressure to front-load the Prologue is an *authorial* pull, not a commercial one, and the

two should be scored separately.

8.3 Value at the ruled price

At the ruled direction (~$20 as a permanent discount, EA at or under that), the value bar for four

early-light chapters is comfortably met by Steam norms; the density curve puts Ch 2-5 deliberately at the

light end of a 100-200h main-path game [SRC: ship-rulings memory, THE DENSITY MODEL;

ROADMAP_POST_5090_TO_SHIP.md P1]. **The commercial risk of shape (a) is narrative shape, not content

volume.** Expanding to Ch 13 buys volume nobody is asking for at this price and spends the one thing EA

exists to protect — see 8.4.

8.4 The EA feedback argument (the decisive commercial point against shape (b))

Josh's own ruled rationale for Early Access: *"chapters ship as updates; the EA feedback channel feeds the

care backstop + density curve"* [SRC: ROADMAP_POST_5090_TO_SHIP.md D-STEAM-FORMAT; ship-rulings memory].

EA's entire value is getting real players onto a small surface before the large surface is built.

Shipping twelve chapters before any feedback arrives means eight chapters — including three or more

elevated-care regions (Ch 11 Congo, Ch 12 West Africa, Ch 13 South Africa is explicitly

cultural_depiction_tier elevated [SRC: docs/spine/CH_13.md # Entry header]) — get built, polished to

the P3 zero-bug bar, and shipped against unvalidated assumptions, with the care backstop arriving

after they are already public. Shape (b) does not just delay the loop; **it wastes it on the eight

chapters where it would have been worth the most.** On the commercial axis this is the single strongest

argument against expanding the first ship, and it is Josh's own ruling arguing it.

---

9. The Josh Gate per shape — what he would actually feel

He plays nothing until the AAAAA-polished slice at P3-exit [SRC: THE JOSH GATE, ROADMAP_POST_5090_TO_SHIP.md].

So the shape decides what his *first contact with his own game in years* is.

• (a) Ch 2-5. He starts in a cave with a creature he never chose dying, and finishes mid-journey. The

world and systems will land hard — they are the strongest thing in the build. But he is likely to name

Gaps A and B himself, from the chair, after four chapters have been carried through the full P2

generative passes and the P3 zero-bug bar. That is the most expensive possible place to discover a

structural note. Mitigated substantially by §4's re-point and §2's Ch-5 curtain, both cheap and both

landable long before the gate.

• (b) Ch 2-13. A far better-shaped sitting (alone → witnessed, twelve chapters, a real companion) — but

he waits much longer for it, and it still contains none of the material he wrote yesterday. The

gift he would play is the unprovenanced version, which risks the worst outcome of all: *his first

experience of his own authored chain is the version with the meaning removed.*

• (c) Prologue → 13. Maximum emotional satisfaction and the one shape that gives him his authored arc.

And the specific harm is exactly the thing the Josh Gate exists to protect: **his first playthrough of

the fairy realm would be of the fairy realm at its worst version** — built before the realm grammar, before

the art boards, before Atlantis and the Duat proved the pipeline. The Josh Gate was created so his hours

go to the ceiling. Shape (c) shows him a floor and calls it the ceiling. **This is the humane argument

against (c), and it is the one worth putting to him**: not "it costs too much," but *"you only get to

see the fairy realm for the first time once, and this shape spends it early."*

• (d) staged. His gate sitting is the base slice — feel, ceiling, care backstop, exactly as designed —

and his first playthrough of the Prologue happens at the true ceiling, at the end, with seventy-plus

chapters of climb behind it. That is the best first impression of the game's most important surface that

any shape can produce, for him and for every EA buyer simultaneously.

---

10. Shapes ranked on this axis, with the two additions the analysis supports

Ranked by emotional payload delivered per unit of first-impression spent, which is the right metric

when the scarcest asset is a thing you can only show someone once.

1 · (d+) STAGED EA, base slice Ch 2-5, hardened — RECOMMENDED on this axis. Shape (d) plus four cheap

amendments, none of which is a scope expansion:

• (i) Re-point CH02_B01 onto the mystery + survival-empathy channel: the bite staged as anomalous,

not merely fatal (§4). Reveal-safe; costs staging, not content.

• (ii) The amulet worn, visible, remarked-on and unexplained from minute one (§4) — the

Chekhov's gun that converts the deferred origin into an active mystery and pays off twice.

• (iii) Re-stage the Ch-5 curtain on the datu who reads back — *"you have been noticed"* (§2,

Gap B). Already canon, already reveal-safe, and it is the ending the slice is missing.

• (iv) Declare the build order on the store page (§8.1) — the deferral as the EA roadmap's

headline finale rather than an unexplained absence.

Plus a director-level rung-1-vs-rung-2 call on the choosings (§6), and the gift ruled to the **Ch-13

join** when 13 drops (§7).

*What it delivers:* every EA-critical feeling (§1 items 1, 2, 4-as-cliffhanger, 5) now; item 3 as an

active mystery now and as a full recast later; Josh's authored arc at full strength, in the order that

protects it, for every buyer eventually and for 1.0 buyers in the designed order.

2 · (b+) Ch 2-13, IF and only if the slice must be longer. Ch 13 is the right stopping point if any

expansion happens — the join and the journal are a real act-break and the alone→witnessed rhyme is genuine.

But it fixes the ending problem that §2 shows is fixable for near-nothing, does not touch the opening

problem, spends the EA feedback loop on eight unvalidated chapters including elevated-care regions, and

ships the amulet gift with its meaning removed. Do not expand to 13 to make the gift land — it does not.

3 · (a) status quo, unhardened. Cheapest and fastest, and it ships two open ends. Note carefully: the

distance between rank 3 and rank 1 is four cheap staging decisions, not a scope change. **Shape (a) is not

really a rival to (d) — it is (d) with the hardening left off.**

4 · (c) Prologue → 13. Wins the emotional argument outright and loses on first impressions, build-order

craft, schedule, and the Josh Gate's own purpose. Recommend against.

Added shapes the analysis supports (examples-not-limits):

• (e) SHIFTED DROP ORDER — the Prologue ships earlier than it builds last. Keep the build order exactly

as ruled (fairy realm after the mythological realms prove the grammar), but let the Prologue drop

once that grammar is proven rather than after Ch 77. This is the only shape in which an EA buyer gets

the gift provenanced on first play, and it is a genuine amendment to a same-day ruling — **it needs its

own brief and it is not this analysis's to assume.** Worth naming because it is the only escape from

§7's sequencing constraint.

• (f) THE ORIGIN AS A FREE STANDALONE PROLOGUE. When the Prologue is built, release it as a separate

free front-end (a common and well-understood Steam pattern) as well as folding it into the main build.

It is a wishlist engine at the exact moment the EA run is trying to convert to 1.0, and it lets a new

buyer start in the designed order without asking existing owners to restart. Costs a store page and a

build target; no canon implications. Purely commercial, and it composes with (d) and (e) both.

---

11. The recommendation, in one paragraph, and the strongest objection to it

RECOMMENDATION. Hold the base slice at Ch 2-5 and harden it (the anomalous bite, the worn unexplained

amulet, the Ch-5 "you have been noticed" curtain, the declared build order); ship staged Early Access with

chapters dropping forward; rule the amulet gift to the Ch-13 join so the base act-break has an object;

and land the Prologue + Ch 1 as Early Access's crowning finale, where the fairy realm is built after the

grammar that makes it the grandest thing in the game — which also gives Josh, and every EA buyer, the

single best possible first sight of it. The EA ordering does not damage the amulet chain; it performs the

arc's own hindsight recast at companion scale, and 1.0 buyers still receive the designed order intact.

THE STRONGEST OBJECTION. *Every one of those arguments is a production argument wearing an artistic

costume, and the artistic truth is simpler: Josh wrote the Prologue and Ch 1 yesterday because they are

the emotional heart of the game, and the recommendation ships the game without its heart and asks players

to trust a promise. Hindsight-recast is a real mechanism, but it is a mechanism for material the player

ALREADY PLAYED — the Ch-77 reveal recasts seventy-seven chapters the player lived. Nothing recasts a

Prologue the player never played; there is no memory there to reinterpret, only a gap that gets filled

late. A player who buys EA, plays Ch 2-5 with a stranger's pet dying in beat one and an unexplained trinket

on their chest, and then waits years for the chapter that explains both, has not been given a hindsight

gift — they have been given an IOU, and IOUs are what Early Access is most distrusted for. Meanwhile the

"first impressions are spent once" argument cuts the other way with equal force: EA launch IS the first

impression, and shipping it without the game's authored opening spends that impression on a version of the

game its own author does not consider complete.*

The answer to the objection, and where it lands. It is right that the recast is weaker for material

never played, and §3's table should be read with that discount applied — H is not P, and the table says so.

But it is wrong that the alternative recovers the heart: shape (c) does not deliver the heart at its

authored quality, it delivers the heart built before the craft that makes it beautiful, which is a

different and permanent kind of loss. And the objection's IOU risk is real and is addressed head-on rather

than waved off — that is precisely what amendment (iv) is for. **Where the objection genuinely lands: the

minimum-viable-setup question (§6) must not be skipped.** The gap between "a stranger's pet dies" and "the

creature I chose dies" is the entire difference between the objection being decisive and being answerable,

and rung 2 closes most of it for a fraction of shape (c)'s cost. **If one thing in this analysis is

escalated to Josh alongside the scope call, make it the choosings question, not the scope question.**

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