Repository navigation
Research
github-actions[bot] edited this page Jul 25, 2026
·
6 revisions
Background research that informs Volley!'s direction. Each doc stands alone; together they give the context behind the choices the design and art docs make.
| Doc | Purpose |
|---|---|
| The Case for Open Development | The published essay. ~15,900 words across fourteen sections plus bibliography. The case for open development as the most reliable practice for a new indie to be seen, with implications past games. |
The AI-tooling design docs live in the sister repo, volley-ai.
| Doc | Purpose |
|---|---|
| Visual Positioning | Where Volley! sits visually among idle, desktop, and sport games. Reasoning behind the hand-drawn, live-scene direction. |
| Early Clone Games | Why Breakout's clone lineage outlasted Pong's, and what cloning shaped in the early industry. |
| Open Development Plan | Earlier internal plan that seeded the essay. |
| Paddle Bounce | How paddle-and-ball games actually compute return angle. Background for 01-prototype/tech/03-paddle-bounce.md. |
| Points vs NoEstimates | The case for dropping story points: where points came from and the political reason they exist, why estimates miss (the planning fallacy) and why the velocity fix breaks (Goodhart), and the NoEstimates alternative of forecasting from flow. |
| Why AI Loves Em-Dashes | Research note on the em-dash prose tell, why the model reaches for it and how the no-em-dash rule counters it. |
| Folder | Tracked | Contents |
|---|---|---|
drafts/ |
no | Section files for the essay (00 title through 15 sources). Concatenated via the cat command in regen scripts to produce the published file. |
notes/ |
yes | Per-angle research notes the essay was built from. |
Contributing
The game
The Cast
Process
Reference