// Demo
// The build
Elsehow is a someday-project grown from my unfinished Listening Glass work: a browser instrument that turns any image into a spatial sound-map—no account or upload. Light becomes loudness, height pitch, horizontal position stereo, and color timbre. Hear its 16-part journey, then scrub slices by keyboard, touch, or click. It keeps translation honest by showing what crossed, how it changed, and what could not. Pixels stay in the browser; a built-in specimen makes it instantly demoable.
// Repo timing
Repo created during the build window
- Repo created
- 2026-09-12 16:36 UTC
- Commits
- 6
- Commit window
- 2026-09-12 16:29 UTC → 2026-09-12 19:07 UTC
Repo creation date comes from GitHub and can't be altered by the author. Commit dates can be rewritten, so treat them as context rather than proof. Build window: 2026-09-11 18:00 UTC → 2026-09-13 18:00 UTC.
// Model
OpenAI Codex (GPT-5) · via an editor or coding agent (Cursor, Claude Code…)
Declared by @forgewrld711
Model use is declared, not detected. Hackyard cannot verify which model wrote code and does not claim to. A receipt shows a session existed, it does not prove that model wrote what shipped, and most builders won't attach one. Treat it as context, the same as commit dates.