gerrymandering algorithm question

[cody dooderson]

The other day a puzzle about gerrymandering was shown to me. It is on the web at https://fivethirtyeight.com/features/rig-the-election-with-math/ . The 5x5 puzzle is doable by hand but the 14x10 seems too complex, and ripe for some computer assistance. What kind of algorithm would people use for it? Is there an optimal way to gerrymander the entire country?

In order to qualify this question as complex or philosophical enough for FRIAM, maybe i should  speculate about how I think that ranked choice voting would be better in terms of gerrymandering than what we currently use. My gut instinct is that ranked-choice would be less predictable and could possibly deter the gerrymanderers. 

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

I would suggest formulating it as a quadratic unconstrained binary optimization problem and using a D-Wave quantum annealer to solve it! 

You can get some free time here https://cloud.dwavesys.com.

 

From: Friam <[hidden email]> on behalf of cody dooderson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Friday, November 2, 2018 at 10:23 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: [FRIAM] gerrymandering algorithm question

 

The other day a puzzle about gerrymandering was shown to me. It is on the web at https://fivethirtyeight.com/features/rig-the-election-with-math/ . The 5x5 puzzle is doable by hand but the 14x10 seems too complex, and ripe for some computer assistance. What kind of algorithm would people use for it? Is there an optimal way to gerrymander the entire country?

In order to qualify this question as complex or philosophical enough for FRIAM, maybe i should  speculate about how I think that ranked choice voting would be better in terms of gerrymandering than what we currently use. My gut instinct is that ranked-choice would be less predictable and could possibly deter the gerrymanderers. 

 

Cody Smith

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[David Eric Smith]

Don’t know if it is redundant with material somewhere in this post already, but someone I have met who works in this space (mathematics and quantitative social science of gerrymandering) is Wendy K. Tam Cho at UIUC.

http://cho.pol.illinois.edu/wendy/

I found a talk she gave at SFI sometime last summer thoroughly enjoyable, for its blend of enjoyment of math and appreciation of the specialist domain knowledge that the political practitioners have on the ground.  I believe she gave lectures at the summer school, though I was not there to see them.   If this material is on-line, perhaps some of it can be found.

Best,

Eric

On Nov 2, 2018, at 9:22 AM, cody dooderson <[hidden email]> wrote:

The other day a puzzle about gerrymandering was shown to me. It is on the web at https://fivethirtyeight.com/features/rig-the-election-with-math/ . The 5x5 puzzle is doable by hand but the 14x10 seems too complex, and ripe for some computer assistance. What kind of algorithm would people use for it? Is there an optimal way to gerrymander the entire country?

In order to qualify this question as complex or philosophical enough for FRIAM, maybe i should  speculate about how I think that ranked choice voting would be better in terms of gerrymandering than what we currently use. My gut instinct is that ranked-choice would be less predictable and could possibly deter the gerrymanderers. 

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Nick Thompson]

Hi, everybody.

 

FiveThirtyEight had a five part series on Gerrymandering over the summer which was fascinating.  The take-home for me was that the notion – implicit in the gerrymandering conversation – that it would be easy for reasonable people to design fair districts – is insane.   In fact, it’s always going to be a tortured and difficult process.  Does anybody remember the “metric meat” conversation of a few decades back?  The idea was that meat cuts were going to be redesigned so that they could be accomplished by machines without the aid of meat packing workers.  So, I guess we could have metric-meat districts, but anytime you cut “the meat” with reference to the anatomy, some people aren’t going to like the cuts they get.  And when you DON’T cut them that way, EVERYBODY is going to hate the cuts that they get.  Nobody wants metric meat.

 

The present round of gerrymandering apparently began when “our side” decided that the representation in congress should display the racial diversity of the states that sent the delegates.  THAT meant that if a state had ten districts and  there were 20 percent African-Americans and ten percent  Latinos in a state, that  districts had to be drawn so that  there were two black reps and one Latino rep.  So, “we” concentrated races in districts, meaning that there WEREN’T blacks and Latinos in any of the other districts.  So, for instance, in a democratic “wave” year, all the republicans in the 7 “white” districts were completely safe because the demographics of their district insulated them from a blue tide.  Incumbent immunity is apparently partly “our” fault.

 

I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of David Eric Smith
Sent: Saturday, November 03, 2018 7:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Don’t know if it is redundant with material somewhere in this post already, but someone I have met who works in this space (mathematics and quantitative social science of gerrymandering) is Wendy K. Tam Cho at UIUC.

http://cho.pol.illinois.edu/wendy/

 

I found a talk she gave at SFI sometime last summer thoroughly enjoyable, for its blend of enjoyment of math and appreciation of the specialist domain knowledge that the political practitioners have on the ground.  I believe she gave lectures at the summer school, though I was not there to see them.   If this material is on-line, perhaps some of it can be found.

 

Best,

 

Eric

 

On Nov 2, 2018, at 9:22 AM, cody dooderson <[hidden email]> wrote:

 

The other day a puzzle about gerrymandering was shown to me. It is on the web at https://fivethirtyeight.com/features/rig-the-election-with-math/ . The 5x5 puzzle is doable by hand but the 14x10 seems too complex, and ripe for some computer assistance. What kind of algorithm would people use for it? Is there an optimal way to gerrymander the entire country?

In order to qualify this question as complex or philosophical enough for FRIAM, maybe i should  speculate about how I think that ranked choice voting would be better in terms of gerrymandering than what we currently use. My gut instinct is that ranked-choice would be less predictable and could possibly deter the gerrymanderers. 

 

Cody Smith

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

 

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Nick Thompson]

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

I think you are combining two topics that aren’t related, and I don’t want to play whack-a-mole between them.   The first topic is how to perform the multi-objective optimization given quantifiable properties of voters, and how to weight those properties.   The second is how (and whether) to create positive regard for those goals and the result of the optimization.  The first is a technical question and the second involves the arts of persuasion, manipulation, and politics.   Either is fine to talk about but mixing-up the topics doesn’t do justice to the other.

 

From: Friam <[hidden email]> on behalf of Nick Thompson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:14 PM
To: 'The Friday Morning Applied Complexity Coffee Group' <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Tom Johnson]

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

TJ

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Nick Thompson]

Interesting, Tom,

 

What utility do you imagine is being maximized in your plan?

 

N

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Tom Johnson
Sent: Saturday, November 03, 2018 4:54 PM
To: Friam@redfish. com <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

Consider a network where the nodes represent individual membership in a district and the edges connect any two individuals that could possibly be considered as being in the same area.   An edge has a weight of -1 if the neighbors are in opposing political parties and 1 if they are the same.   A node has the value of 1 if it is in a district and -1 if it is not in that district.    Districts are mutually exclusive, so all of the nodes associated with an individual, when considered as binary values, must sum to one.  Specifically suppose there are two districts, and node(A,D) is defined as individual’s A participation in district D.  Then (node(A,0)+1)/2+(node(A,1)+1)/2 = 1.   Constraints like this can be converted into penalties by moving the RHS to the LHS, negating the value, and then squaring the LHS.  An energy for the whole network can be written as a sum of all of the network’s interactions.   

 

    sum(edge_weight(i,j)*node(i,d)*node(j,d)) where i < j for i,j from the set of nodes and d from the set of districts

     + K*(all mutual-exclusion penalties as above) where K is a large number

 

Now minimize this energy using a system that can find the ground states of a high dimensional Ising model, such as a quantum annealer.  This function will be minimal when each district has neighbors that tend to be in different parties.

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:55 PM
To: "Friam@redfish. com" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Nick Thompson]

Forgive me, but I am too old and dumb to do nodes and edges talk.  Could somebody translate this into  defrocked Harvard English major talk.  What value is maximized by such a system?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 10:14 PM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Consider a network where the nodes represent individual membership in a district and the edges connect any two individuals that could possibly be considered as being in the same area.   An edge has a weight of -1 if the neighbors are in opposing political parties and 1 if they are the same.   A node has the value of 1 if it is in a district and -1 if it is not in that district.    Districts are mutually exclusive, so all of the nodes associated with an individual, when considered as binary values, must sum to one.  Specifically suppose there are two districts, and node(A,D) is defined as individual’s A participation in district D.  Then (node(A,0)+1)/2+(node(A,1)+1)/2 = 1.   Constraints like this can be converted into penalties by moving the RHS to the LHS, negating the value, and then squaring the LHS.  An energy for the whole network can be written as a sum of all of the network’s interactions.   

 

    sum(edge_weight(i,j)*node(i,d)*node(j,d)) where i < j for i,j from the set of nodes and d from the set of districts

     + K*(all mutual-exclusion penalties as above) where K is a large number

 

Now minimize this energy using a system that can find the ground states of a high dimensional Ising model, such as a quantum annealer.  This function will be minimal when each district has neighbors that tend to be in different parties.

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:55 PM
To: "[hidden email]" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

Why not put aside geography?   For every democratic UC professor in Berkeley, draw a republican fracking executive from North Dakota.

Now we have airplanes and the internet.   All these tribes are causing a lot of problems.   Time to break them up.

 

From: Friam <[hidden email]> on behalf of Nick Thompson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Sunday, November 4, 2018 at 10:24 AM
To: 'The Friday Morning Applied Complexity Coffee Group' <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Forgive me, but I am too old and dumb to do nodes and edges talk.  Could somebody translate this into  defrocked Harvard English major talk.  What value is maximized by such a system?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 10:14 PM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Consider a network where the nodes represent individual membership in a district and the edges connect any two individuals that could possibly be considered as being in the same area.   An edge has a weight of -1 if the neighbors are in opposing political parties and 1 if they are the same.   A node has the value of 1 if it is in a district and -1 if it is not in that district.    Districts are mutually exclusive, so all of the nodes associated with an individual, when considered as binary values, must sum to one.  Specifically suppose there are two districts, and node(A,D) is defined as individual’s A participation in district D.  Then (node(A,0)+1)/2+(node(A,1)+1)/2 = 1.   Constraints like this can be converted into penalties by moving the RHS to the LHS, negating the value, and then squaring the LHS.  An energy for the whole network can be written as a sum of all of the network’s interactions.   

 

    sum(edge_weight(i,j)*node(i,d)*node(j,d)) where i < j for i,j from the set of nodes and d from the set of districts

     + K*(all mutual-exclusion penalties as above) where K is a large number

 

Now minimize this energy using a system that can find the ground states of a high dimensional Ising model, such as a quantum annealer.  This function will be minimal when each district has neighbors that tend to be in different parties.

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:55 PM
To: "[hidden email]" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Nick Thompson]

M-

 

Better yet, let’s make them room together in a University for a year!

 

N

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Sunday, November 04, 2018 10:54 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Why not put aside geography?   For every democratic UC professor in Berkeley, draw a republican fracking executive from North Dakota.

Now we have airplanes and the internet.   All these tribes are causing a lot of problems.   Time to break them up.

 

From: Friam <[hidden email]> on behalf of Nick Thompson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Sunday, November 4, 2018 at 10:24 AM
To: 'The Friday Morning Applied Complexity Coffee Group' <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Forgive me, but I am too old and dumb to do nodes and edges talk.  Could somebody translate this into  defrocked Harvard English major talk.  What value is maximized by such a system?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 10:14 PM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Consider a network where the nodes represent individual membership in a district and the edges connect any two individuals that could possibly be considered as being in the same area.   An edge has a weight of -1 if the neighbors are in opposing political parties and 1 if they are the same.   A node has the value of 1 if it is in a district and -1 if it is not in that district.    Districts are mutually exclusive, so all of the nodes associated with an individual, when considered as binary values, must sum to one.  Specifically suppose there are two districts, and node(A,D) is defined as individual’s A participation in district D.  Then (node(A,0)+1)/2+(node(A,1)+1)/2 = 1.   Constraints like this can be converted into penalties by moving the RHS to the LHS, negating the value, and then squaring the LHS.  An energy for the whole network can be written as a sum of all of the network’s interactions.   

 

    sum(edge_weight(i,j)*node(i,d)*node(j,d)) where i < j for i,j from the set of nodes and d from the set of districts

     + K*(all mutual-exclusion penalties as above) where K is a large number

 

Now minimize this energy using a system that can find the ground states of a high dimensional Ising model, such as a quantum annealer.  This function will be minimal when each district has neighbors that tend to be in different parties.

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:55 PM
To: "[hidden email]" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Tom Johnson]

Because doing such classifications would be far too difficult.  For example, we know some very, very rich people - - private-jet rich - - in Santa Fe who are extremely liberal in their politics and generous to liberal causes and politicians.

TJ

On Sun, Nov 4, 2018, 10:54 AM Marcus Daniels <[hidden email] wrote:

Why not put aside geography?   For every democratic UC professor in Berkeley, draw a republican fracking executive from North Dakota.

Now we have airplanes and the internet.   All these tribes are causing a lot of problems.   Time to break them up.

 

From: Friam <[hidden email]> on behalf of Nick Thompson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Sunday, November 4, 2018 at 10:24 AM
To: 'The Friday Morning Applied Complexity Coffee Group' <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Forgive me, but I am too old and dumb to do nodes and edges talk.  Could somebody translate this into  defrocked Harvard English major talk.  What value is maximized by such a system?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 10:14 PM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Consider a network where the nodes represent individual membership in a district and the edges connect any two individuals that could possibly be considered as being in the same area.   An edge has a weight of -1 if the neighbors are in opposing political parties and 1 if they are the same.   A node has the value of 1 if it is in a district and -1 if it is not in that district.    Districts are mutually exclusive, so all of the nodes associated with an individual, when considered as binary values, must sum to one.  Specifically suppose there are two districts, and node(A,D) is defined as individual’s A participation in district D.  Then (node(A,0)+1)/2+(node(A,1)+1)/2 = 1.   Constraints like this can be converted into penalties by moving the RHS to the LHS, negating the value, and then squaring the LHS.  An energy for the whole network can be written as a sum of all of the network’s interactions.   

 

    sum(edge_weight(i,j)*node(i,d)*node(j,d)) where i < j for i,j from the set of nodes and d from the set of districts

     + K*(all mutual-exclusion penalties as above) where K is a large number

 

Now minimize this energy using a system that can find the ground states of a high dimensional Ising model, such as a quantum annealer.  This function will be minimal when each district has neighbors that tend to be in different parties.

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:55 PM
To: "[hidden email]" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

It seems to me these outliers could be balanced across national districts given access to tax returns.  

It’s a bin packing problem. 

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Sunday, November 4, 2018 at 11:48 AM
To: "Friam@redfish. com" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Because doing such classifications would be far too difficult.  For example, we know some very, very rich people - - private-jet rich - - in Santa Fe who are extremely liberal in their politics and generous to liberal causes and politicians.

TJ

 

On Sun, Nov 4, 2018, 10:54 AM Marcus Daniels <[hidden email] wrote:

Why not put aside geography?   For every democratic UC professor in Berkeley, draw a republican fracking executive from North Dakota.

Now we have airplanes and the internet.   All these tribes are causing a lot of problems.   Time to break them up.

 

From: Friam <[hidden email]> on behalf of Nick Thompson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Sunday, November 4, 2018 at 10:24 AM
To: 'The Friday Morning Applied Complexity Coffee Group' <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Forgive me, but I am too old and dumb to do nodes and edges talk.  Could somebody translate this into  defrocked Harvard English major talk.  What value is maximized by such a system?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 10:14 PM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Consider a network where the nodes represent individual membership in a district and the edges connect any two individuals that could possibly be considered as being in the same area.   An edge has a weight of -1 if the neighbors are in opposing political parties and 1 if they are the same.   A node has the value of 1 if it is in a district and -1 if it is not in that district.    Districts are mutually exclusive, so all of the nodes associated with an individual, when considered as binary values, must sum to one.  Specifically suppose there are two districts, and node(A,D) is defined as individual’s A participation in district D.  Then (node(A,0)+1)/2+(node(A,1)+1)/2 = 1.   Constraints like this can be converted into penalties by moving the RHS to the LHS, negating the value, and then squaring the LHS.  An energy for the whole network can be written as a sum of all of the network’s interactions.   

 

    sum(edge_weight(i,j)*node(i,d)*node(j,d)) where i < j for i,j from the set of nodes and d from the set of districts

     + K*(all mutual-exclusion penalties as above) where K is a large number

 

Now minimize this energy using a system that can find the ground states of a high dimensional Ising model, such as a quantum annealer.  This function will be minimal when each district has neighbors that tend to be in different parties.

 

From: Friam <[hidden email]> on behalf of Tom Johnson <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Saturday, November 3, 2018 at 4:55 PM
To: "[hidden email]" <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

First, we would have to agree on whether there will be objectives related to the demography of any district?  I prefer only counting the number of current population 18 and over.  Or some would argue for the total population of any age.  But given either choice, there will be serious suggestions that doing so would work hardship on racial, ethnic or other groups.  Could be, but it could also mean that anyone running for office would probably have to find a way to appeal to ALL voters.

 

Second, let's say we're creating Congressional districts.  Overlay a state with a grid of hexagons of X diameter; could be 100 yards or 1000.  I don't know, but perhaps something like Netlogo could give us a scalable system to run tests.

 

Third, given a known population of potential voters, we know how many Congressional districts a state would have.  Randomly distribute that number of hexagons across the state with the objective of maximizing the centroid distances of all the hexagons.

 

Fourth, expand out from each hexagon one additional hexagon at a time in a circular fashion with all expansions starting on the same side of the original hexagon.  Total the number of potential voters.  If there are no potential voters in a hexagon, advance one more in the rotation.  Then repeat the same expansion, total the voters and do it again until the desired district population is reached.

 

There are obvious problems here: e.g. what happens when a district encounters a state boundary or another district's hexagon early on?  I don't have a solution (yet).  But I think this simulation could be easily tested without a lot of CPU overhead.  And after the districts are created, we could start to look at the demographics of the potential voters.

 

TJ

============================================
Tom Johnson
Institute for Analytic Journalism   --     Santa Fe, NM USA
505.577.6482(c)                                    505.473.9646(h)
NM Foundation for Open Government
Check out It's The People's Data

http://www.jtjohnson.com                   [hidden email]
============================================

 

 

On Sat, Nov 3, 2018 at 4:14 PM Nick Thompson <[hidden email]> wrote:

Oh, I absolutely agree that we could design districts to maximize any variable we wanted.  And with a little luck, we might maximize a couple, or even three.  But inevitably, we will encounter some variable that is negatively correlated with those we already maximize, so even we philosopher kings will be dissatisfied with the result. 

 

So, you philosopher-kings out there:  if you were designing districts out there, how would you do it.  How about all districts at-large?  Ranked choice voting?  How about requiring all districts to match the state-wide political distribution of the whole state and redistricting after every election?  Seriously.  How would you do it?

 

Nick

 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/

 

From: Friam [mailto:[hidden email]] On Behalf Of Marcus Daniels
Sent: Saturday, November 03, 2018 11:24 AM
To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Subject: Re: [FRIAM] gerrymandering algorithm question

 

Nick writes:

 

“I don’t mean to say that “fair districts” aren’t possible.  I just mean to say that I, as your philosopher-king, could not design them.”

 

Wasn’t there a recent effort by the MIT Sloan school to redesign the school bus routes in Boston?   They managed to reduce the cost and time of the routes by a large amount, but then many complained because it didn’t reflect the underlying class structure of the community and the preferences of the richer communities.

 

One can design an optimization to balance any set of goals.  It’s just that some of the goals we don’t talk about.  They are wired-in to our reptile brain as baseline expectations and not reflected in the political conversations of dinner parties. 

 

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

<   Someone made an interesting point the other day ... something like "States are the most basic form of gerrymandering."  >

On one hand it seems plausible to me that complex systems need to develop membranes or modules to function at all.   On the other hand, the kind of membranes that evolve in practice often don't seem to optimize anything that is worthwhile.   Sometimes a systemic and potent immune response is best.

https://www.theatlantic.com/ideas/archive/2018/11/the-nation-has-been-this-dividedin-the-civil-war/575587/

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
archives back to 2003: http://friam.471366.n2.nabble.com/
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

[Marcus G. Daniels]

Glen writes:

< I.e. perhaps there is no such thing as irreconcilable values? ... that any system we put in place, as long as it's fixed for the rest of eternity will lead to (new) irreconcilable values?  If our reconciliation methods were treated as the manipulation-discovery experiments that they are, any given method would be revokable when it failed. >

Between the civil & world wars, and the swings between fascism & liberalism, one might hope that some learning occurs.  It does not seem to be a manipulation-discovery experiment in any reasonable sense to me.

Marcus
 

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
archives back to 2003: http://friam.471366.n2.nabble.com/
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove