Thoughts, Limitations, and Acknowledgments
Acknowledgments: Angela Perkins, Janna Avon, everyone else involved in DHSS, Professor Justin Corvino, Moon Duchin, Mira Bernstein, Nicholas Stephanopoulos, Eric McGhee, Pablo Soberon, MGGG, FiveThirtyEight, Vox, Princeton University, University of Chicago, The Atlantic, Quanta, WIRED, Center for American Progress, Brennan Center for Justice, Carnegie Mellon University, Duke University, The New Yorker, NPR, The Nation, Bloomberg, University of Toronto, NYTimes, Mazore, Victor Powell, TemplateMo.
Limitations:
1. Reliance on data analysis due to a lack of availability of data and necessary computing power
2. Natural gap in expertise/knowledge/understanding of source material - topics were covered from Geometry, Topology, Probability/Statistics, and Computer Science
3. Not enough time to discuss all possible methods so had to narrow it down to the two most relevant ones that had been utilized in court cases
Final Thoughts/Notes: Math against gerrymandering is definitely on the right track. Algorithms/models will obviously have issues in their initial phases - that's inevitable and nothing to worry about as long as they are being optimized and enhanced. The domino-effect that MCMC and EG had in their respective landmark court cases can be felt to this day as court cases are popping up in Indiana, Illinois, Michigan, Oklahoma, Pennsylvania, and other states where experts are being brought in to testify with MCMC or EG or one of the two methods is being considered in the analysis of the districting plans. There's a movement happening across the judiciary bodies of this country and more and more faith is being placed in the power of mathematics - and we need all the math we can get with the latest round of redistricting coming up.
The methods that I discussed in this project are by no means exhaustive. A lot of great work is being done by the Metric Geometry and Gerrymandering Group (MGGG at Tufts), FiveThirtyEight's Gerrymandering Project, the Princeton Gerrymandering Project, and many others. There are other novel approaches also being researched and tested out in the fight against partisan gerrymandering. They include, but are not limited to, curve-shortening flows, Districtr, Maup, GerryChainJulia, TopDown noising algorithm, and simulations/optimization.
All my sources and works cited can be found here.