Understanding the Data

Note on Methodology: We use Fractional Attribution to ensure accuracy. If a single commit touches multiple categories (e.g., both "GUI" and "Tests"), we split its credit evenly (e.g., 0.5 each). This ensures the total height of the bars exactly matches the developer's unique commit count, avoiding the inflation common in other impact metrics.

Impact Profile (The Radar): Not all code is equal. We use a Risk-Weighted model (explained in Methodology) to visualize where a developer focuses their energy. "Security" code (Consensus/Crypto) is weighted significantly higher (50x) than "Education" code (Docs, 1x) because of the catastrophic risk involved. A shape skewed toward the top-right indicates a Protocol Engineer; a shape toward the bottom-left indicates a Product/UI Engineer.