SoylentNews is people
SoylentNews
from the the-simple-answer-is-that-there-isn't-a-simple-solution dept.
The problems of gerrymandering are manifold, often debated and lamented. Now, a group of computational geometers from Tufts, MIT and others are working the problem from different fronts. From https://sites.tufts.edu/gerrymandr/
The Metric Geometry and Gerrymandering Group (MGGG) is a small Boston-based team of mathematicians launched by Moon Duchin of Tufts University. Our mission is to study applications of geometry and computing to U.S. redistricting. We believe that gerrymandering of all kinds is a fundamental threat to our democracy.
Our goals are:
- to pursue cutting-edge research in the practically relevant applications of geometry, topology, and computing to the redistricting problem;
- to foster collaboration with researchers in statistics, supercomputing, law, political science, and other fields;
- to facilitate direct civic engagement by training scholars from a variety of quantitative backgrounds to serve as expert witnesses and consultants in redistricting cases;
- to educate the public, both through direct outreach and by helping college and high school teachers incorporate units on voting, gerrymandering, and civil rights into the mathematics curriculum;
- to build a diverse community of mathematically inclined people around the country and give them the knowledge and the tools to hold map-drawers accountable when 2020 comes around.
And from https://sites.tufts.edu/gerrymandr/get-involved/
We are assembling a team of mathematicians, lawyers, statisticians, and active citizens of all stripes to work on practical metrics and solutions for gerrymandering in advance of the 2020 U.S. Census. If you're interested in joining our community, please fill out our Skills and Interest Inventory form.
For anyone that wants to get up to speed on this complex and important topic, https://sites.tufts.edu/gerrymandr/resources/ is a page of links to a variety of related papers and articles.
SN discussed the math of the gerrymander back in 2014,
https://soylentnews.org/article.pl?sid=14/12/27/1148245
(Score: 1, Interesting) by Anonymous Coward on Thursday July 05 2018, @08:43PM
I like my method:
To be legal a district map must have:
1) the right number of districts
2) the entire area (state) must be covered by exactly one district
3) each district must be contiguous (even if it has a thin strand connecting two areas)*
4) the least populous district must have at least 95% the population of the most populous state
Everyone in either congress can submit one map for consideration, the winner is the legal map with the lowest total district perimeter. Existing county borders count at half weight (* those thin strands get expensive an only one of the two borders can be ON a county line). Perhaps existing district boundaries count at 3/4 weight to keep boarders in place over time a little bit.
The map that will get picked is the most unfair map that one party can manage that is more fair than the most fair map the other party can manage, unless there is a third party in there to spoil it and make it even more fair.
Bruce Schneier called such a system a "self-fulfilling protocol", where the options of an opponent keep your actions more fair. Another example is the splitting a snack where one person splits it, the other picks the "half" they want.