Farm the Gap: A strategy game and learning platform to feed the future.

Developed with the software and tools below

📍 Overview

Farm The Gap, a new digital puzzle game from The Plotline, gives you the keys to a virtual global farm. Modeled on real-world data, the game’s objective is to close a 50% “food gap” by simply changing what we grow and eat.

This repository contains all the source code and data for the game.

👾 Development

System requirements

• Node v20+
• npm/pnpm

Steps

1. Clone the repo and run npm install to install dependencies
2. Start a development server: npm run dev
3. Build a production version: npm run build
4. Preview the production build: npm run preview.

Global state

The brains of the game are defined in src/lib/stores/state.ts.

In this file are five writable Svelte stores:

1. $farm: The foods on the farm and the respective production outputs
2. $gameState: Food inventory, current year, remaining undos
3. $gameSettings: High-level global settings
4. $userState: User preferences and interaction states
5. $gameHistory: History of moves (foods added/removed and where)

The first three stores feed into a derived Svelte store called $successMetrics. This store monitors the overall game status (win/loss) and the warning or fail states of each individual game metric (protein supply, emissions, etc). Any changes in the dependant stores triggers a recalculation of the derived store.

Data types

Global types and class declarations can be found in src/ambient.d.ts. Some important data structures:

Farm

class Farm {
  constructor(settings: GameSettings | null)

  // properties
  initialState: FarmState
  grid: FarmGrid
  items: string[]
  rows: number
  cols: number

  // methods
  getInitialState(): FarmState
  getFarmMetric(fn: (item: Food) => number): FarmMetric
  plantCrop(x: number, y: number, foodItem: Food): void
  getTotalSum(fn: (item: Food) => number): number
  getSumByFoodType(fn: (item: Food) => number): FarmGridFoodList[]

  // getters
  readonly yield: FarmMetric
  readonly landUse: FarmMetric
  readonly waterUse: FarmMetric
  readonly emissions: FarmMetric
  readonly eutrophy: FarmMetric
  readonly protein: FarmMetric
  readonly calories: FarmMetric
  readonly foodChanges: Count[]
}

Failure metrics

interface FailureMetric {
  value: number
  key: FailureMetricKey
  label: string
  suffix: string
  limit: number
  objective: string
  warn: boolean
  fail: boolean
  history: number[]
  farmMetricKey: FarmMetricKey
  foodMetricKey: keyof Food
  chartSettings: LineChartSettings
}

High-level game status

interface SuccessMetrics {
  hectaresPerPerson: number
  peopleAdequateCalories: number
  calorieProductionChange: number
  caloriesPerPersonPerDayValue: number
  proteinPerPersonPerDay: FailureMetric
  emissionsChange: FailureMetric
  waterUseChange: FailureMetric
  eutrophyChange: FailureMetric
  hasSucceeded: boolean
  hasFailed: boolean
}

Source data

Foods

I had to whittle down the hundreds of foods eaten globally into a sensible list of groups for easy gameplay. That involved making decisions about how to group and categorize foods, and finding the optimal trade-off between simplicity and accuracy. I arrived at the 13 foods and food groups in the stats table below.

Find these defined in src/lib/data/foods.json.

Food	globalLand (%)	kgYieldPerHa	caloriesPerKg	proteinPerKg	emissionsPerKg	waterUsePerKg	eutrophyPerKg	landPerKg
🐄 Beef	53	50	2,430	175	95	2494	428.7	295.3
🐑 Lamb	9	27	2,550	171	39.7	1803	97.1	369.8
🥛 Dairy	6	2,458	610	32.7	3.7	310	11.5	4.1
🌾 Rice	6	4,740	3,590	70.4	3.9	1,586	26.5	2.9
🍞 Wheat	6	3,483	3,700	151	1.6	648	7.2	3.9
🐖 Pork	4	737	2,280	178	9.8	1292	60.9	13.6
🥜 Nuts*	2	1,600	5,930	206.7	1.8	2993	16.6	11
🫘 Legumes*	3	1,896	3093	192	1.4	52	11.7	16.4
🐓 Poultry	3	1,200	1,330	179	8.2	483	34.3	8.3
🌽 Corn	2	5,847	3,640	62	1.7	216	4	2.9
🥦 Vegetables*	2	19,862	663	16.7	0.5	48	1.9	0.4
🍊 Fruit*	2	13,655	418	6.8	1.1	186	3.5	1
🥚 Eggs	2	1,465	1,430	124	4.4	830	20	6.8

Nuts impact data is the average of Nuts and Groundnuts (commodity data); nutrition data is the average of raw peanuts, almonds, and cashews.

Legumes data is a weighted average of Beans (75%), Chickpeas (10%), Lentils (5%), Garden Peas (10%)

Vegetables impact data is the weighted average of Root Vegetables (75%), Other Vegetables (20%), and Brassicas (5%). Nutrition data is the weighted average of gold potatoes, mature carrots and raw broccoli.

Fruit impact data is weighted average of Apples (11%), Bananas (14%), Berries (1%), Citrus Fruit (18%), Tomatoes (21%), and Other Fruit (35%) (commodity data). Nutritional data is weighted average of red delicious apples, bananas, strawberries, navel oranges, roma tomatoes and bartlett pears.