584K views · Jun 28, 2021 · 26:48 · 2 HN recommenders · 3 mentions · discussed 2023–2025
AlphaPhoenix uses Monte Carlo simulation and annealing-style methods to draw North Carolina congressional district maps, both gerrymandered and fair. He focuses on the fascinating computational problem rather than the politics.
Why Hacker News points people here: HN recommends it as an entertaining demonstration of how powerful algorithmic gerrymandering can be.
“For a great (and entertaining!) look into how you can use simulations to generate these types of maps”HN comment, 2023
“For a great (and entertaining!) look into how you can use simulations to generate these types of maps”
Entertaining look at generating district maps with simulations.
For a great (and entertaining!) look into how you can use simulations to generate these types of maps you should check out this video by the superb channel AlphaPhoenix: this video
“Required watching to put things into perspective of how powerful gerrymandering can be”
Shows how powerful gerrymandering can be.
Required watching to put things into perspective of how powerful gerrymandering can be this video There was a proposed (and failed) ballot initiative in Ohio that tried to address this in the only way you actually can. The tl;dr is that there is no such thing as a fair map and so the state would be required to gerrymander the districts to match the state-wide election results. Basically proportional representation with extra steps, but brings the "secret metric" in map making to the forefront and requires it be a specific value so you can't tune it to your favor.
“reminds me of AlphaPhoenix’s automated gerrymandering project”
Related example of automated gerrymandering with randomness.
This aspect of adding randomness to avoid local randomness reminds me of AlphaPhoenix’s automated gerrymandering project: this video
