SignalSpace

A consent-first signal ground where wireframe avatars turn movement and intentional signals into visible waves and actions.

Comment

Inspiration

Digital spaces usually represent people through static profile images or realistic 3D characters. I wanted to explore a different idea: what if presence looked like energy, movement left visible waves, and intentional signals could become small shared actions? SignalSpace is a prototype of that experience.

What it does

SignalSpace is an interactive social-space prototype for exploring what signal-driven communication could feel like.

A visitor chooses a name, enters a ground, and moves a wireframe avatar through a dark, responsive environment. Movement produces sonar-like rings and data particles. The visitor can enable a clearly labelled EEG-style simulation, watch its confidence change, and trigger a Pulse, Wave, or Glow. Echo, the single guide bot, explains what is happening, while the event stream shows whether an action came from a manual control or the local simulation.

The project is designed to make an unfamiliar idea understandable in the first minute. It does not claim to read thoughts or collect real sensor data. The current version is a browser simulation that demonstrates the interaction language, privacy boundaries, invitation flow, and overall experience.

How it was built

SignalSpace uses Next.js, React, TypeScript, the Canvas API, Web Audio, Web Share, and OpenAI Sites hosting. Canvas renders the perspective ground, wireframe bodies, particles, and expanding rings. React manages movement, identity, simulated confidence, chat, and the event stream. The hosted build also supports optional Sign in with ChatGPT.

I built the core experience with Codex using GPT-5.6. Codex helped translate the visual concept into an animation system, implement the interaction and sound layers, simplify the room to one guide bot, add newcomer guidance, clarify the privacy model, fix production issues, and deploy the working site. GPT-5.6 was especially useful for reasoning across product design, interaction design, implementation, and honest technical communication in one continuous build thread.

Challenges

The hardest part was making an abstract concept understandable without overstating what the prototype can do. The visuals needed to feel alive, but the interface also needed to distinguish manual actions, simulated confidence, and real sensor data. Another challenge was building a convincing animated scene without a 3D engine; the final version creates depth and movement with Canvas geometry, gradients, particles, and layered sound.

Accomplishments

The whole experience runs in the browser without an API key, dataset, or specialist hardware. A new visitor can understand the idea, enter the ground, move, trigger a signal, talk to Echo, and share the room in less than a minute.

Built With

Share this project:

Updates