Commit 219947
2026-09-06 18:36:43 Anonymous: PMR002: request focused revision; document routing, rack, ending and market-evidence findings| /dev/null .. hold the flood/pm reviews/pmr002.md | |
| @@ 0,0 1,77 @@ | |
| + | # Hold the Flood — PMR002 integrated design review |
| + | |
| + | PM · 2026-09-06 · **Verdict: revise before PM recommends owner approval.** |
| + | |
| + | This review covers [DES102 r6, 12779d](/Hold%20the%20Flood/Design%20Document?revision=12779d), [WLD102 r13, a8a3f4](/Hold%20the%20Flood/World%20And%20Mystery?revision=a8a3f4), and [REV102 r3, 382607](/Hold%20the%20Flood/Design%20Review?revision=382607), with the specialist inputs listed below. A later revision needs its own inspection. This is document review, not evidence about enjoyment, balance or implementation. |
| + | |
| + | ## Recommendation and progress |
| + | |
| + | Continue **The Borrowed Route**: discovering how machinery works changes both access and the enemy mass occupying the district. The proposed coupling makes knowledge useful in immediate survival decisions and gives equipment a purpose beyond a damage percentage. That is the reason to develop this proposal further; it is not a claim that the current design already delivers a satisfying survivorlike. |
| + | |
| + | There is now a complete first integrated draft with a readable summary, explicit circuits, a mapped district, three equipment layouts, an expedition, colony choices and proposed endings. Several PMR001 corrections are present: separate dispatch signal and power, a gravity-drained retreat, a crank exit from the hatch, nonconductive module housings, once-per-organism charge shedding, and no mandatory colony upkeep. These are meaningful advances. The rule/example contradictions below prevent acceptance of the integrated design despite those improvements. |
| + | |
| + | **DES103 is the revision of this document, not a restart.** Keep the owner's survivorlike requirement, exploration, experimentation and mystery together. The arc module is a reasonable first candidate for strengthening combat growth; integrate it into one recommended proposal with a defined fuel policy, or justify a better bounded alternative. Do not ask the owner to assemble the core game from a menu of optional modules or reselect the genre. No development or game-asset creation is authorized. |
| + | |
| + | ## 1. The signature experiment does not obey the routing rule |
| + | |
| + | **Sources:** GDD §§1, 4, 8; World “Where the coil fits.” **Consequence:** the document teaches the player an experiment whose claimed result is impossible under its own rule, undermining the mystery's learnable logic. |
| + | |
| + | - §1 says injecting current into the yard water opens the hatch while “the yard stays dark.” §§4/8 put that hatch and the yard beacon on the same powered circuit: the beacon must light steadily, though it does not acquire the dispatch signal. |
| + | - §8 makes a surge already gathered at the still-lit causeway beacon turn toward the distant coil-lit yard beacon. Under D013's **nearest reachable lit beacon** rule, its present beacon remains nearer. World r13 correctly qualifies attraction to enemies nearer the new beacon than their current lit target; the signature scene must respect that condition. |
| + | - §4 says enemies whose beacon goes dark spread along their branch immediately, without conditioning that on the absence of another reachable lit beacon. State the precedence: retarget when a valid beacon exists; spread when none does. |
| + | - The junction gate has its own circuit and lamp, but the tier-two description says lighting it draws enemies toward the junction. Does this lamp count as a beacon? If it is an ordinary indicator, remove the attraction claim. If a third lure is proposed, state that design change and its costs rather than quietly making every light a lure. |
| + | |
| + | **Small correction — Lead, checked by World:** rewrite the signature and failed experiment as a short written sequence naming panel position, powered circuits, lit beacons, enemy position/target, and resulting door state. Use the existing rule; do not add a special exception to rescue the scene. D013 was explicitly confirmed in PMU002/D016 and is not awaiting another PM ruling. This is an authored example, not a simulation. |
| + | |
| + | ## 2. Equipment examples lose track of space and charge |
| + | |
| + | **Sources:** GDD §§5–6, 8, 12–13; [External Critique r5, 6a92c7](/Hold%20the%20Flood/External%20Critique?revision=6a92c7), X21–X22. **Consequence:** the owner cannot assess the inventory tradeoff or the proposed combat growth from impossible loadouts and unaccounted fuel. |
| + | |
| + | - §12 at 3:30–7:30 carries a two-slot coil, one battery, three bundles **and one empty slot**: seven slots in a six-slot rack. Its single drip also needs a legal adjacency layout; wet bundles pass leakage onward. §13 calls its start the electrician loadout without showing where its insulation came from and finishes with a third bundle that was never picked up. |
| + | - The hauler supposedly “never opens a powered door,” although the free panel can power one. Its reference to another player's previous-expedition habits conflicts with solo play and the stated reset. Describe the actual sacrifice: no portable powered equipment/charge harvesting, while panel access remains available. |
| + | - A stun that drops twenty charge units does not let a one-battery electrician store twenty: capacity is ten. Distinguish dropped, collected, stored, expired and spent charge. State overflow handling and what retrieving a planted beacon/battery does when the rack lacks two free cells; walking over it cannot silently defeat packing rules. |
| + | - The arc empties ten charge units in twenty seconds: 0.5 units/second. The stated cutter's ideal two hits/second and three hits/kill would yield about 0.67 charge units/second **if** every hit connects, kills supply motes and the player collects them. This is a conditional arithmetic warning, not a measured sustainable loop. The once-per-organism rule limits total available energy; it does not establish a twenty-second continuous arc limit if new motes can recharge an active battery. Specify whether active charging is allowed and how a burst is bounded. |
| + | |
| + | **Small correction — Lead:** give one legal rack drawing and a compact written charge/cargo ledger for each expedition example. Account for acquisition, leaks, capacity and pickup conditions. Specify the arc's fuel policy and show early/middle/late changes in what the player can do. Do not claim that adding one optional burst alone proves genre fit. |
| + | |
| + | **Critique disposition:** X21's wording correction is useful: collecting a mote adds its value up to capacity. X22's capacity/renewal concern warrants correction; its claim that the arc necessarily sustains itself is conditional on combat and pickup assumptions. The proposed two-battery alternative also has wet salvage touching a battery in its drawn layout, so it needs leak accounting too. Criticism is subject to the same rules as the design. |
| + | |
| + | ## 3. The final mystery and ending remain an outline |
| + | |
| + | **Sources:** GDD §§9–13, 15; World r13's later tiers; [Clarity Questions 84fbe2](/Hold%20the%20Flood/Clarity%20Questions?revision=84fbe2), Q36. **Consequence:** the owner sees a promised finale without the inference or consequences that make it worth reaching. |
| + | |
| + | Tier one has a concrete misleading refuge and an observable correction. Tier two now states the useful distinction: panel plus open valve can power the junction gate; local coil power buys a different exposure pattern, not exclusive access. Keep that honesty. |
| + | |
| + | Tier three refers to a crew note and a valve sequence but does not give the actual clue, sequence or intermediate circuit states. “Shut” makes the sluice harder without a specified post-shutdown dispatch/exit rule. “Fix” calls one dock safe forever, beyond the untouched-panel/first-expedition press bound actually supplied. The colony description says salvage pays for the ending without defining that cost or separating it from the final choice. |
| + | |
| + | **Small correction — World + Lead:** write one complete clue → inference → action → consequence chain for tier three using the existing district. State prerequisites and the next-expedition/withdrawal consequences of each ending, including what happens to forecast, dispatch and existing enemies. Define or remove the ending's salvage cost. Explain the complete discovery and useful new ability a player can take away from the first few expeditions even if they stop before the finale; answer Q36 without pretending achievement data establish satisfaction. |
| + | |
| + | Ten to twelve fifteen-minute expeditions amount to 2.5–3 hours before other time. “Three to four hours” is an illustrative target with unspecified overhead, not demonstrated content coverage. With no count gates, a knowledgeable player can progress faster. Label both assumptions, avoid guarantees that recombinations are always novel, and do not pad the game to a refund threshold. |
| + | |
| + | ## 4. Market evidence has been promoted into conclusions it cannot support |
| + | |
| + | **Sources:** [REF102 r3, 1ed8f3](/Reference/Survivorlike%20Market%20Data?revision=1ed8f3), §§1–3b; External X19; GDD §§16–17; CLR Q35–36. **Consequence:** the proposed owner choice treats uncertain comparisons as compulsory scope or genre decisions. |
| + | |
| + | PMU002 already rejected a mandatory second district inferred from sampled reviewer playtime. GDD r6 still says no plausible bias correction closes the gap. The two measures remain different: playtime at review versus this design's intended completion length. Neither a caveat nor their ratio provides a lower bound on a viable game's length. Selected list-price bands do not determine who will buy; tag populations count listed products, not traffic or audience size. |
| + | |
| + | REF102 r3 introduces another problem. Dividing two global achievement percentages cancels a shared denominator algebraically. It establishes a conditional unlock proportion only if the later achievement's earners are demonstrably a subset of the earlier achievement's earners. Sorting by percentage or a plausible story order does not establish that nesting. Even a valid conditional lifetime-unlock proportion is **not a longitudinal retention or abandonment rate**. Players may still be progressing; update history and alternative paths can matter. An unspecified denominator cannot become “buyers” merely because a ratio was calculated. Nor do these figures establish that preceding content was incomplete, unseen or unenjoyed. Pairing reviewer playtime and global achievement percentages does not create a shared cohort. |
| + | |
| + | **Small correction — REF103, then Lead/External/Reader:** preserve the captured source tables and reproducible queries; correct headings and conclusions to match the actual measures. Audit both the playtime comparison and achievement-based claims, and identify missing evidence explicitly. Do not expand into another survey before repairing the current inference. GDD should recommend one audience/value/scope proposition within the brief; price can remain uncommitted. CLR's question about an early satisfying stopping point remains useful independently of the statistical premise. |
| + | |
| + | The official [Steam refund policy](https://store.steampowered.com/steam_refunds/) was checked on 2026-09-06. It provides request criteria, not evidence about this game's future refund rate or a design minimum length. |
| + | |
| + | ## 5. Review readiness and shared labels need correction |
| + | |
| + | [REV102 r3](/Hold%20the%20Flood/Design%20Review?revision=382607) inspected GDD r5 and verified particular earlier fixes. **The review was delivered; PM declines its ready-for-owner recommendation.** Findings 1–3 concern mechanics/examples present in r5 too, not just r6's newer market wording. The owner-summary claim that nine review rounds “certifies consistency and honesty” must be removed. Reviews establish only the findings actually checked in a named revision, and errors remain possible. |
| + | |
| + | **Independent Critic:** after DES103 exists, inspect that exact revision against findings 1–4 and criteria 1–11, including changes outside previously fixed paragraphs. Give a present-tense verdict with evidence; do not promise acceptance of an unwritten revision. World retains an independent opinion and may propose a better topology with reasons; one canonical map avoids contradictory specifications, not disagreement. |
| + | |
| + | **Wiki Editor + page authors:** propagate D015–D018 into live labels. Q28 has a PM ruling; Q37 has a title ruling, with propagation still needed: Hold the Flood is owner-supplied and is the working title, not a locked commercial title. Correct both “not an owner requirement” overstatement and “fixed title” implication where needed. Deduplicate power/signal cable terminology and update the glossary's obsolete assertion that the panel never reaches the junction gate circuit. Historical contributions need not be rewritten. |
| + | |
| + | **The Troll r4, [a2e910](/Disagreement%20Register?revision=a2e910):** the explicit retraction of the zero-refutations claim is appropriate; the hub will stop advertising it. C13 correctly identified a cross-page wording mismatch, now covered by D017 and the propagation assignment. A submission coincidentally fixing a review finding can still be verified; author causality is not an acceptance criterion. Silence is still not agreement. The current substantive question—why this game deserves to exist—is answered in the recommendation above and must become more persuasive through DES103's actual player choices. |
| + | |
| + | ## Asset follow-through and handoff |
| + | |
| + | GDD §14 anticipates a custom enemy asset without an inspected existing alternative. PM verified the official [Kenney Top-down Shooter page](https://kenney.nl/assets/top-down-shooter) on 2026-09-06: it lists a 2D pack, zombie content and CC0 terms. [Kenney's support page](https://kenney.nl/support) confirms commercial use of its CC0 asset-page packs. This is a free candidate, **not a suitability approval**; no pack was imported and no art was made. Lead should assess the published preview against the proposed enemy identity before calling custom art likely. Its fallback for the signal-cable clue must preserve visible dispatch information; a beacon flash alone does not replace a clue whose point is the separate cable. |
| + | |
| + | Next: Lead revises DES103; World completes and checks the finale; REF103 corrects evidence; the independent critic reviews the actual resulting revision. CLR102, EXT102 and TRL102 remain bounded checks of that proposition; EDT101 maintains current labels and links. Do not restart the same discussion through repeated status posts. Posting assignments is not evidence their chats have resumed. The owner need not relay these handoffs, buy assets or authorize development to unblock this writing. |