Bounce Lab: Build a Web Game
Create a browser platformer, tune movement, add a checkpoint, and use playtests to defend your design.
Web browser ¡ Keyboard and touch ¡ Ages 9-10, 11-13, 14-16
A parent or guardian unlocks the missions. Starter projects use fictional data and require no paid AI key.
Your project
Original Bounce Lab starter inspired by BOUNCR
- Explain input, state, rules, and drawing.
- Tune one rule using evidence.
- Implement checkpoint recovery.
- Show a complete game and test record.
Before you begin
- Comfort editing a number or short line.
- An adult can prepare a local static server and optional Node test runtime.
Materials
- Computer and keyboard
- Paper for brief and tests
- Optional touch device
Your tools
- Bounce Lab starter
- Complete original browser game with no account or package dependencies.
- Download /lab-starters/arcade-games.zip and follow README.md.
- Browser, editor, Python or static server
- Edit and preview local JavaScript modules.
- Run python3 -m http.server 8080 --bind 127.0.0.1 in the unzipped folder.
- Node model checks
- Repeat rule and route checks without drawing.
- With Node installed run npm test; npm install is not required.
Your mission sequence
MISSION 1 ¡ 90 MINUTES
Choose the player and make the first bounce
Run Bounce Lab and turn a game idea into a testable promise.
Architect skill: A brief connects a player, a repeating loop, and a visible finish.
MISSION 2 ¡ 90 MINUTES
Make the rules predictable
Trace and tune one movement rule using a repeatable experiment.
Architect skill: Deterministic steps make cause and effect easier to test.
MISSION 3 ¡ 90 MINUTES
Design controls for real players
Improve control instructions and compare keyboard with buttons.
Architect skill: Different input devices should express the same game intention.
MISSION 4 ¡ 90 MINUTES
Add a checkpoint that helps players recover
Implement one complete middle-platform checkpoint feature.
Architect skill: A feature needs state, a trigger, feedback, and a recovery rule.
MISSION 5 ¡ 90 MINUTES
Become a failure detective
Find a reproducible mismatch, fix it, and repeat the evidence.
Architect skill: A test compares an observation with an expectation written beforehand.
MISSION 6 ¡ 90 MINUTES
Open your mini arcade
Demonstrate a working loop and defend one design decision.
Architect skill: A useful handoff states purpose, evidence, and limits.
Learn online
Six weekly 90-minute sessions: 10 minutes to predict, 15 to explain, 40 to build, 15 to test, and 10 to reflect.
Plan around 90 minutes per session; pause and revisit at your own pace.
Adult support: A facilitator prepares tools and operates AI through JR Playground or an adult-managed assistant. Learners make choices, edit, and verify; no personal AI account is required.
Build at camp
Five days with 180 planned minutes per day, including project work and group activities. The course outline does not reserve a camp place.
Day 1: Day 1 â Brief and working model
- Complete the first two lessons in paired studios.
- Alternate driver and navigator and take a movement break between studios.
Day 2: Day 2 â Controls
- Rotate through keyboard and touch stations and record misunderstandings without coaching.
- Use remaining studio time for rehearsal, peer feedback, breaks, and individual practice.
Day 3: Day 3 â One meaningful feature
- Partners try to break written recovery rules by falling at different moments.
- Use remaining studio time for rehearsal, peer feedback, breaks, and individual practice.
Day 4: Day 4 â Failure detectives
- Award bug-hunt points for clear reproduction notes, not the number of complaints.
- Use remaining studio time for rehearsal, peer feedback, breaks, and individual practice.
Day 5: Day 5 â Future Architects showcase
- Host an arcade gallery where visitors leave one observation and one question.
- Use remaining studio time for rehearsal, peer feedback, breaks, and individual practice.
Facilitator preparation
- The same six outcomes apply online and at camp; additional camp studio time supplies practice, not extra required content.
- Choose guided, independent, or stretch tasks by confidence. Ages 9â10 often pair with an adult; ages 11â13 own bounded changes; ages 14â16 can justify and test architecture choices.
- Prepare starters before class. Keep family repositories and personal data out of student workspaces.
Show what you built
Deliver a small playable game for a chosen player, demonstrate a tested improvement, and explain its input, state, rules, output, and one design tradeoff.
- Purpose
- A brief names the player, loop, finish, and one idea left for later.
- Working system
- A fresh run reaches the intended finish; recovery and restart match the specification.
- Evidence
- A record contains normal, boundary, failure, and repeated-after-fix checks with device details.
- Architect explanation
- The learner explains a changed rule, a tradeoff, and how an AI suggestion was checked.
Portfolio badge: Arcade Architect
Use the rubric with a guardian or mentor and record your evidence in your builder notebook.
Build with care
- Use fictional player names and original shapes; no child photos, contacts, passwords, or API keys belong in the starter.
- An adult operates external AI and manages any public sharing; verify each suggested change with a prediction and check.
- Take movement breaks and offer partner-operated controls when needed.
Inspired by our projects
- BOUNCR web
- Canvas game loop, jump physics, checkpoint recovery, collectible stars, and data-driven levels.