Gerrymandering Highlighter

[fileunderjeff]

The process of drawing districts is often flawed, with boundaries drawn largely down partisan lines. What if there were data-driven ways to draw districts? How could it provide a "check" on gerrymandering, so that if you stack the computer's districts and the people's districts, we can easily see the battle lines?

Background: http://www.informationweek.com/government/open-government/wanted-honest-algorithms-for-voter-redistricting/a/d-id/1297859 http://www.drdobbs.com/genetic-algorithms-optimal-solutions/184405617 http://www.newsweek.com/gerrymandering-districts-can-be-undone-math-359204

Open Primaries brought this idea our way, and they would be a partner and subject matter expert on redistricting. We will also seek out an academic partner.

[fileunderjeff]

We have a couple of subject matter experts on board with scoping out this project. There is also a possible partnership with the ACLU of Texas. Will invite them all to comment here.

[rgarrett]

This looks interesting. If anyone else wants to discuss it, let me know.

[rlgreen91]

Hmm, so I actually remember seeing something like this on Reddit. Let me look.

Here we go: https://www.reddit.com/r/dailyprogrammer/comments/3lf3i2/20150918_challenge_232_hard_redistricting_voting/?ref=search_posts

[zenjen777]

This definitely is a timely issue given the recent ruling. http://www.mystatesman.com/news/state--regional-govt--politics/court-voids-texas-congressional-districts/FJoxFl3QTJxL1PAqYnMKoO/

[betterways]

Theres also a need for a Nextdoor organized by districts since they have so little to do with neighborhoods

[jordanweinstein]

How can I get involved / help with this?

[shantilise]

Sounds like a great project! How do I get involved?

[rlgreen91]

So, I'm not @fileunderjeff, but I imagine that the first thing would be to develop an algorithm that draws district lines. I know I linked to the challenge above, but I imagine the first thing to do would be to create a program that, given a graph, can divide it into a number of requested districts. So, if I give you a 5x5 area and ask for 3 districts, it could do that, with the only requirement being that none of the districts overlap. Does anyone have experience with developing algorithms based on graph theory?

[doostee]

I am also interested in this idea. Here is someone who has developed an algorithm and generated maps for all 50 states: http://bdistricting.com/2010/

He also has good information about work done by others using different metrics and algorithms: http://bdistricting.com/about.html#other

[TommyJW]

I'm interested in this as well. For Hackathon scope, could we look at running the algorithm on our local district lines then highlighting the areas most impacted? Then, we would need to add additional parameters to pull demographic data, because as John Oliver explained well, not all oddly shaped districts are Gerrymandered.

[fileunderjeff]

Here is an interesting, open source project that might help: https://georeactor.github.io/gerrymandering/

[betterways]

"What if there were data-driven ways to draw districts?" There are. That's how they gerrymander.

[fileunderjeff]

@betterways my understanding is that the districts are negotiated. data plays a part, but I don't think anyone would call the process data driven.

[betterways]

https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.bustle.com/p/how-does-packing-cracking-work-last-week-tonight-tackled-the-intricacies-of-gerrymandering-video-50089&ved=0ahUKEwiO5Lbr9_7TAhUh24MKHWbGBTEQFghvMA8&usg=AFQjCNF6tIfFT5NN_I4tsqhn73tnNWkf6A&sig2=PwChcZXK2ZXBMTKRvwF74Q