Here’s how we do it:

1. Create a canvas element in the HTML body.
   - Width: 800px
   - Height: 600px

2. Initialize drawing context on the canvas.
   - Set color to black and line style to none.

3. Define power dynamics rectangle with initial position and rotation speed.
   const rect = { w: 150, h: 300 };

4. Create voices (particles) array.
   const VOICE_COUNT = 80;
   for(let i=0; i<VOICE_COUNT; i++){
      voices.push({
         x: Math.random() * canvas.width,
         y: Math.random() * canvas.height
      });

5. Define drawBackground function to fill canvas with blue gradient.
   ctx.fillStyle = bgGrad;

6. Define drawPowerRect function to draw power dynamics rectangle on canvas.
   rect.angle += rect.rotSpeed;
   ctx.save();
   ctx.translate(canvas.width/2,canvas.height/2);
   ctx.rotate(rect.angle);

7. Draw voices by iterating over array and applying physics simulation for particle motion.

8. Call drawConnections() to draw lines connecting voices around the circle.

9. Run loop to continuously update canvas state based on voice positions and rotation calculations.

This is how we design a feedback ritual that empowers marginalized voices without exacerbating power dynamics: by acknowledging their presence and giving them space to speak while we hold our own axes straight