A seedable stream shuffler modeled as a roundabout network (Python)

Name: A seedable stream shuffler modeled as a roundabout network (Python)
Availability: InStock
Author: velocitatem

by velocitatem·Feb 28, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●SolidBig BrainNiche GemWizardry

Shuffling metaphor with real math—97.5% Fisher-Yates quality but solves no obvious problem over standard random.

Strengths

•Novel roundabout-network abstraction makes deterministic mixing interpretable and tunable (routing bias, recirculation depth, injection strategy)
•Rigorous benchmarking: Kendall-tau, chi-squared, bit bias, total variation distance vs Fisher-Yates with confidence intervals
•Seed-reproducibility + visibility into mixing process solves simulation/game-engine repeatability without black-box randomness

Weaknesses

•Use case remains niche—data pipelines default to Fisher-Yates, simulations rarely need this tuning visibility
•Metaphor-driven design prioritizes readability over performance; unclear why users abandon standard shuffle for interpretability here
•0.0051 bit bias improvement is marginal and likely unnoticeable in practice

Post Description

You enter in order, but after a few loops and exits, your neighbors are no longer your neighbors. It’s a tunable Python package for reproducible mixing, with benchmark + determinism tests.