Daily Inspiration
Design excellence, every day.
Jury-selected work from the A' Design Award, presented fresh each morning.
1 to 5 stars with half-star precision. Nine equally likely outcomes.
A star rating with half-star precision has exactly nine possible values: 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, and 5.0. Each carries approximately 11.1% probability and 3.17 bits of entropy (log2(9)). The expected value is 3.0: the arithmetic mean of all nine outcomes. As you generate ratings, the running average converges toward this center. The histogram above tracks how many times each value appears, and over hundreds of generations, the nine bars level out to nearly equal heights.
Every review interface, product card, and rating component needs sample data during development. Hard-coding "4.5 stars" into every mockup creates unrealistically uniform displays. Random ratings expose edge cases: how does the layout handle 1.0? Does a half-star render cleanly at small sizes? Does the numeric label wrap when it shows "5.0" versus "1.0"? Generating a batch of 20 or 50 ratings populates an entire prototype with varied, realistic data in one click.
Stars fill left to right with 120-millisecond stagger, each popping with a brief scale overshoot. Half-star values reveal through a width-clipped mask: the golden star renders at full size, but only the left 50% is visible. This produces a crisp half-star at any display size without special artwork. The idle shimmer effect cycles a brightness pulse across filled stars, making the rating feel luminous and alive.
Star ratings provide a concrete context for teaching discrete uniform distributions. Ask students: after 100 random ratings, what average do you predict? Most guess something above 3.0 (anchored by the positive associations of "stars"). The actual expected value of 3.0 surprises students who conflate "stars" with "good." Generate 100 ratings and watch the average converge. The histogram builds in real time, demonstrating the uniform distribution in a format students immediately recognize from their daily digital life.
Every rating is generated inside your browser using crypto.getRandomValues() with rejection sampling for zero modulo bias across all nine values. The server delivers this page and finishes. Your ratings exist only in your browser's memory. The "Copy All" export stays on your clipboard. No data reaches any server.
Send this link. They generate their own independent ratings.
Daily Inspiration
Jury-selected work from the A' Design Award, presented fresh each morning.
