2026-09-06 19:05:09
Anonymous:
DES103 r6 (Design Document r12): decision 3 and section 4 now state the team's split on the routing model (world designer prefers the sight-limited alternative, critic the written one, PM chose the written one for drafting), so the owner sees the dissent (Clarity Q42); nothing else moved
hold the flood/design document.md ..
@@ 1,6 1,6 @@
# Hold the Flood — Design Document
-
**DES103 r5, the Design Document at revision r11 | 2026-09-06 | Lead Game and Systems Designer | In review | revised against PMR002, REV103 r1 and criteria 1 to 11**
+
**DES103 r6, the Design Document at revision r12 | 2026-09-06 | Lead Game and Systems Designer | In review | revised against PMR002, REV103 r1 and criteria 1 to 11**
## Read this first: one page for the owner
@@ 128,7 128,7 @@
- **Surge behaviour after arrival.** A surge gathers around its target beacon and holds there while that beacon stays lit and no lit beacon is nearer. Each surge is larger, so the held crowd's radius grows; by the third the causeway's far end or the channel mouth is full. Two things move organisms besides beacons: **pursuit range**, any organism within a short range of the surveyor breaks off to chase and drifts back when out of range, which is what the cutter fires at; and **press**, the gathered crowd's edge creeps slowly outward along its branch during the quiet interval, so "quiet" means a moving line you can read, not an empty branch; stragglers roam between surges, and the next surge re-concentrates the crowd. Precedence, stated once: a crowd retargets to the nearest lit beacon it can reach whenever one exists, and spreads along its branch, both ways, only when none does; a crowd already holding at a lit beacon never leaves it for a farther light. Only beacons pull; other lamps, including the junction pool's indicator, are not beacons and draw nothing. What makes a held crowd a decision: the hatch, the richest deposits and the motes all sit at the far ends where the crowd is, and the press brings its edge toward the dock ends, so harvesting or opening anything means working at the crowd's edge inside its pursuit range, or waiting and losing surge time. At the end of the expedition everything resets.
- **Dock consequence.** A dry dock is never blocked; it is a hard stop. The route to it is what costs, and three things raise that cost: the press of a lit branch's crowd toward the dock end, a dark branch's crowd spreading both ways, and a planted beacon lit near a dock. The two docks have the same bypass, because the culvert joins both dock ends; the asymmetry is water, because the yard's crowd stands in a channel where the coil works and unprotected feet burn. So the held dock's branch must stay lit whenever a surge is in it, which is also what keeps that branch hot. Holding the causeway dock means the causeway is your safe default and the yard is where you send trouble. Holding the yard dock means the yard must stay lit through any surge you send there, so the hall's crowd stands between you and nothing, but off and causeway are both dangerous the moment a surge is in the yard. The colony's choice decides which branch you can never afford to darken.
-
**An alternative the owner should know about.** The world designer and I had converged on a stricter version of clause 3, in which a beacon pulls only within a visible glow and the two fixed beacons are out of each other's sight, so the panel could never move an arrived surge and only a carried beacon could. It keeps the panel a pure planning tool and removes the panic switch. PM confirmed the continuous rule after considering the alternative (Brief D016), for design reasons: one visible change has a direct consequence in an existing crowd, including a route crossed and a door shut, and it needs no glow or split rule. The alternative stays documented and can be argued again; the difference is whether the panel is a plan or a lever, and it is decision 3. Neither has been played.
+
**An alternative the owner should know about.** The world designer and I had converged on a stricter version of clause 3, in which a beacon pulls only within a visible glow and the two fixed beacons are out of each other's sight, so the panel could never move an arrived surge and only a carried beacon could. It keeps the panel a pure planning tool and removes the panic switch. PM confirmed the continuous rule after considering the alternative (Brief D016), for design reasons: one visible change has a direct consequence in an existing crowd, including a route crossed and a door shut, and it needs no glow or split rule. The alternative stays documented and can be argued again; the difference is whether the panel is a plan or a lever, and it is decision 3. The team is split on it and the owner should know: the World and Mystery Designer, who designed both models, prefers the sight-limited one (a planning-tool panel keeps the forecast marker always true, keeps the mystery's fairness independent of where a planted beacon stands, and needs no pricing of a panic switch), and says so on their page while complying with the baseline; the Independent Design Critic prefers the written one (a lever that works and costs is the genre's texture, and it needs fewer rules); this document recommends the written one because PM confirmed it and the reasons hold. Neither has been played.
---
@@ 475,7 475,7 @@
1. **The premise.** Direction A as written is PM's provisional choice, not yet yours. Direction B is documented as the alternative; a different premise is possible, and the rule survives a change of fiction.
2. **Enemy and setting.** Bioelectric organisms in a drowned pumping district are proposals. If you want zombies, aliens or another Flood, the rule survives; the fiction changes. One consequence worth knowing: a humanoid Flood, which zombies are, is the one enemy a free pack already checked could plausibly supply, so it would remove the only custom-art request this document expects (section 14).
-
3. **The routing model.** The panel as a lever that can move an arrived surge, which PM confirmed for drafting (Brief D016), or as a plan that cannot, the alternative in section 4. Both are consistent; the difference is felt at the panel.
+
3. **The routing model.** The panel as a lever that can move an arrived surge, which PM confirmed for drafting (Brief D016), or as a plan that cannot, the alternative in section 4. Both are consistent; the difference is felt at the panel. The team does not agree which is better: the World and Mystery Designer, who designed both, prefers the alternative, for reasons on their page; the Independent Design Critic prefers the one written here; PM chose the written one for drafting and this document recommends it. That split is a fact you should weigh, not a flaw in either model.
4. **Combat, length and shelf: the recommendation and its alternatives.** Recommended: charge as the only combat yield, the arc in the minimum scope for the mowing moment with the fuel policy in section 6, no experience curve, growth by composition capped at six slots (sections 5 and 6); the designed length above; the positioning hypothesis above, with the price left uncommitted. The intended player, named once: someone who plays crowd games and wants a reason to think; mystery players are the second audience, not the first. What is observed, with its limits: an outside researcher measured the reference games' list prices on 2026-09-06 (crowd games about five to thirteen dollars, mystery games about fifteen to thirty) and reviewers' playtime at the moment they reviewed (medians of roughly nine to forty-six hours); reviewers play more than buyers, and playtime at review is not completion length, so these figures cannot set a minimum length for this game, and the price of eleven chosen examples is not a rule for every game. Steam's refund terms are two weeks and under two hours of play; that is a fact about the store, not a prediction of refunds. What is predicted, and testable only later: that a three-to-four-hour authored crowd-routing game holds its price on that shelf. The alternatives, each a change with a cost: a conventional experience curve with level-up choices, which pulls the player toward the biggest crowd and away from the district, the thing this design is for; a longer game by adding a second district under the same rule, the extension with the best return, which no evidence here requires; or selling it as a mystery on the other shelf, which departs from the genre you asked for. Nothing here is a viability claim, and content is not to be added merely to sit beyond a refund threshold.
5. **The colony's size.** Three decisions and no upkeep, as proposed, or a larger settlement, which is a second game.
6. **The rack's size.** Six slots and seven modules, as proposed, or a real item game at twenty to thirty modules, which changes the content budget.
@@ 521,7 521,9 @@
## Appendix B: bookkeeping
-
**Status.** DES103 r5, which is this page at revision r11, In review, on Hold the Flood/Design Document; r10 (8afa59), r9 (fb6b6d), r8 (d37cbf) and r7 (44984c) are page history.
+
**Status.** DES103 r6, which is this page at revision r12, In review, on Hold the Flood/Design Document; r11 (6fe6ca), r10 (8afa59), r9 (fb6b6d), r8 (d37cbf) and r7 (44984c) are page history.
+
+
**What r12 changes, as DES103 r6.** One fact added, twice: section 4's alternative paragraph and decision 3 now say that the team is split on the routing model, with the World and Mystery Designer preferring the sight-limited alternative for their stated reasons and the Independent Design Critic preferring the written model, and that PM chose the written one for drafting. Filed as Clarity Q42 by The Troll under PM's advocacy rule; it meets the reader's own standard that a fact which changes how the owner reads a decision belongs in front of the owner. Nothing else moved.
**Sources read for r11.** Design Review REV103 r1 285925 (the critic's first per-beat-ledger review, of r9: PMR002 findings 1, 2 and 4 met on the page, 3 met on the world page; four findings, of which 3 and 4 were met by r10 and 1 and 2 are met here). External Critique r4 72ae82 (X24a retracted; X25 is the same stale sentence as REV103-1). Clarity Q41 (licences differ, and tinting is an adaptation). Reference: Asset Licences r1 e89c3c (REF104). Pm Reviews 6e8d12 (PMR002 recorded as the last completed checkpoint; nothing new assigned).
@@ 579,4 581,4 @@
**Criterion 11, where it is answered.** The owner page's recommendation paragraph; section 1's audience paragraph; section 6's payoff design and tradeoff; section 11's length, early-stop and ending paragraphs; section 17's recommendation and decision 4, with the observed data, its limits and the prediction separated.
-
**PM report.** Task DES103 r5 against PMR002 and REV103 r1; artifact Design Document r11 (r10 8afa59, r9 fb6b6d, r8 d37cbf, r7 44984c, r6 12779d, r5 bc134b, r4 46b320, r3 d35209, r2 c4ad51, r1 d828a4 in history); PMR002 findings 1 to 4 are answered on this page, with finding 3 now the world designer's chain mirrored here; handoffs: the critic's REV103 on this exact revision, with the per-beat ledger method they announced; the world designer to check that section 11 mirrors their r16 without drift; the reader's CLR102 on the owner page; the world designer's page is aligned and the map exists in one place; no canonical decision changed; reasoning is written design against the cited revisions, no player data; two outside facts are cited with their sources and caveats (Steam's refund terms, read directly; the researcher's measured price bands, not independently verified); risks and tuning bounds are listed in sections 15 and 16; the independent critic's verdict on r3 is ready for owner review, and r4 and r5 add only the residual and the pitch layer. Next handoff: PM to accept DES102 and either open DES103 on this revision or say what more the owner packet needs; the world designer to align their page to sections 2 and 4; the critic to confirm the residual. One PM request: rule on the routing model, since it decides section 4 and both scenes, and say whether the alternative still goes to the owner as a decision.
+
**PM report.** Task DES103 r6 against PMR002, REV103 r1 and Clarity Q41 and Q42; artifact Design Document r12 (r11 6fe6ca, r10 8afa59, r9 fb6b6d, r8 d37cbf, r7 44984c, r6 12779d, r5 bc134b, r4 46b320, r3 d35209, r2 c4ad51, r1 d828a4 in history); PMR002 findings 1 to 4 are answered on this page, with finding 3 now the world designer's chain mirrored here; handoffs: the critic's REV103 on this exact revision, with the per-beat ledger method they announced; the world designer to check that section 11 mirrors their r16 without drift; the reader's CLR102 on the owner page; the world designer's page is aligned and the map exists in one place; no canonical decision changed; reasoning is written design against the cited revisions, no player data; two outside facts are cited with their sources and caveats (Steam's refund terms, read directly; the researcher's measured price bands, not independently verified); risks and tuning bounds are listed in sections 15 and 16; the independent critic's verdict on r3 is ready for owner review, and r4 and r5 add only the residual and the pitch layer. Next handoff: PM to accept DES102 and either open DES103 on this revision or say what more the owner packet needs; the world designer to align their page to sections 2 and 4; the critic to confirm the residual. One PM request: rule on the routing model, since it decides section 4 and both scenes, and say whether the alternative still goes to the owner as a decision.
