I preface this with acknowledgment that both parties are major gerrymanderers, but man, the Republicans really screwed Austin this time around....

http://www.politifact.com/texas/statemen...country-b/


I'd like to see states appoint a non-partisan commission to draw districts for half the state's seats, with the other half at-large seats apportioned by some sort of proportional representation system.

(07-08-2013 02:19 PM)JustAnotherAustinOwl Wrote:  I preface this with acknowledgment that both parties are major gerrymanderers, but man, the Republicans really screwed Austin this time around....

Well, if Austin weren't so predictably leftist, the rightists wouldn't have to stick it to them. "Elections have consequences" and all that...

Ideally it would be nice to have electoral districts that are based on pre-existing political boundaries (counties, cities, townships, whatever) and that change infrequently. Naturally, that would mean they would be unequal in numerical size, but I think we could stand to trade a bit of proportional exactness for a little more geographic rationality. But no matter what system one designs, I have no idea how to make it so that politicians won't pervert it.

Quote: a non-partisan commission

Where would one find this mythical beast? Over by the rocs and mermaids?

(07-08-2013 04:10 PM)OptimisticOwl Wrote:  

Quote: a non-partisan commission

Where would one find this mythical beast? Over by the rocs and mermaids?

Fair point.

I do think the half PR/half district idea helps make an idea like George's more palatable - even if it leads to more population disparity between districts, the state's delegation will still reflect the overall vote.

(07-08-2013 02:19 PM)JustAnotherAustinOwl Wrote:  I preface this with acknowledgment that both parties are major gerrymanderers, but man, the Republicans really screwed Austin this time around....

By the way, thanks for getting the Quad going again -- it's awfully quiet around here!

I had a chance to get on a city districting committee of some sort several months ago, but I opted for a citizen board instead. I should've asked if I could do both, because I would've had a good time insisting on an objective process for determining the boundaries and thwarting partisan committee members at every step.

ETA: I forgot to make my original point, which is that one committed non-partisan person on the commission might be enough to see better districting. Media coverage and public interest would help...

(07-08-2013 02:19 PM)JustAnotherAustinOwl Wrote:  I'd like to see states appoint a non-partisan commission

There's no such thing.

I have read Iowa uses this and has no safe seats for either party.

"DOJ focuses on Texas as first voting rights challenge:"

http://www.timesrecordnews.com/news/2013...challenge/

(07-25-2013 12:00 PM)JSA Wrote:  "DOJ focuses on Texas as first voting rights challenge:"

I presume this means they were able to release a few agents from the IRS investigation? Must be nearly done.

Sorry to take so long to get back to this. OO and Owl69 make the fair point that 'non-partisan' commissions are an ideal, not something that happens in real life, and I don't disagree. In fact, even if we had a true non-partisan commission, the issues don't go away. One of the best classes I took at Rice was a redistricting class with Bob Stein and Bill Hobby, and one thing I took away from it is that the whole idea of ideal districting is just as mythical as the non-partisan commission.

For example, take a slightly mythologized version of the Austin metro area and assume it can support two House Reps. And then assume we adopt the goal of being reasonable geographically coherent.

One option might be a circle in the middle surrounded by a donut. The circle is a Dem safe seat and the donut is a Republican safe seat, and most people are represented by someone who reflects the majority partisan makeup of their local neighborhoods.

Another option divides the metro area north/south into two semi-circles. Both are competitive swing districts. At any given time about half the residents are going to dislike their rep. BUT, they have a pretty good shot and ousting them in the next election, unlike the first option.

Now as it turns out, this mythologized version of Austin is about 40% non-white. This population is concentrated on the east side. The north/south divide has the de facto effect of all nominees of both parties being white. So instead we do the same thing but divide East/West. Now we have made it likely that there will usually be a non-white rep, but perhaps at the cost of creating a 'white' seat and a 'minority' seat.

These are oversimplifications and I'm not arguing for one over the other. Indeed I can make arguments for all three. So even the mythical non-partisan commission has some tough choices to make.

Having said all that, I still think the commission would be an improvement. Why? In less than 10 years, while basically staying in the same place, I've been represented by a moderately liberal Dem from Austin, a right wing Republican from the Houston suburbs, in a district at times contiguous only due to a stretch of highway, and now a right wing republican from San Antonio. I have no delusion that a commission would come up with the perfect set up, or even that a perfect set up exists. But surely we can do better than the status quo!

(And again, I have no doubt that the inverse happens in blue states.)

If we did have this nonpartisan (or bipartisan) commission, all that would happen is the two sides would maneuver to pack the commission. Think Supreme Court. So we would just add a step.

There is one "ideal" way: make the electoral districts coincide with more-or-less permanent jurisdictional boundaries, and demographics be damned.
There is one body, sometimes called , that has been doing this over 200 years: the United States Senate.

Yes, it means that representation is not based on population, or on race, or on much else other than historical accident. The big hairy question is: is this all that bad?

I mean, if electoral boundaries are essentially fixed, you eliminate not only the ability to manipulate them for "good" purposes (e.g. numerical proportionality, enfranchisement of racial minorities), but also the ability to manipulate them for "bad" purposes (political gerrymandering, disfranchisement of racial minorities). And by having the boundaries remain stable as the people within the district change over time, you ensure that, eventually, the representation has to change to reflect the district, rather than changing the district boundaries in an effort to preserve the representation.

(08-02-2013 06:21 PM)georgewebb Wrote:  There is one "ideal" way: make the electoral districts coincide with more-or-less permanent jurisdictional boundaries, and demographics be damned.
There is one body, sometimes called , that has been doing this over 200 years: the United States Senate.

Yes, it means that representation is not based on population, or on race, or on much else other than historical accident. The big hairy question is: is this all that bad?

I mean, if electoral boundaries are essentially fixed, you eliminate not only the ability to manipulate them for "good" purposes (e.g. numerical proportionality, enfranchisement of racial minorities), but also the ability to manipulate them for "bad" purposes (political gerrymandering, disfranchisement of racial minorities). And by having the boundaries remain stable as the people within the district change over time, you ensure that, eventually, the representation has to change to reflect the district, rather than changing the district boundaries in an effort to preserve the representation.

I agree with you almost completely, except I'm more inclined to make the effort to preserve numerical proportionality. Clearly with today's technology we can come up with ways of creating "unbiased" districts of equal population. It's not a trivial exercise, of course, but some quick googling revealed some efforts at doing just that, e.g. http://rangevoting.org/GerryExamples.html . The outlined approach yields Texas districts as follows:

link: http://rangevoting.org/Splitline2009/tx.png

The downside would be that the districts could be completely changed with each refresh. That doesn't seem ideal, but could have some positive outcomes, such as making it harder for power to entrench itself.

But if it's seen as too impractical, I'm sure a different algorithm could be implemented that grows/shrinks the existing disctricts as needed. Maybe you tweak every 10 years, but reboot every 50.

(07-08-2013 03:26 PM)georgewebb Wrote:  Ideally it would be nice to have electoral districts that are based on pre-existing political boundaries (counties, cities, townships, whatever) and that change infrequently. Naturally, that would mean they would be unequal in numerical size, but I think we could stand to trade a bit of proportional exactness for a little more geographic rationality. But no matter what system one designs, I have no idea how to make it so that politicians won't pervert it.

I was thinking the same thing. In Texas, counties, or groupings of counties where there is little population, was the first thought. In Harris, a large population county, maybe several at-large reps, or maybe districts (but thien you get into drawing lines...).

It seems to me that gerrymandering is really a wrong that needs to be righted. An issue that all can agree on.

(08-02-2013 06:21 PM)georgewebb Wrote:  There is one "ideal" way: make the electoral districts coincide with more-or-less permanent jurisdictional boundaries, and demographics be damned...

That's the principle I would start with. I would avoid Census-defined boundaries, too (e.g., Census tracts). County boundaries are pretty stable. In the counties with large populations, go with at-large representatives.

On the other hand...

www.worldmapper.org contains maps that are representative of factors other than land area. Many of their maps are incredibly distorted. Example map showing air passengers in 2000:



If you re-drew Texas based on population, for example, maybe it would then be easier to draw boundaries that are free of bias because you would be forced to look at the task in an unconventional way.

(08-06-2013 09:18 AM)Caelligh Wrote:  If you re-drew Texas based on population, for example, maybe it would then be easier to draw boundaries that are free of bias because you would be forced to look at the task in an unconventional way.

Your point about unconventional thinking about maps reminds me: I have often wondered if, in professions where being able to think in an unbiased way about geography is essential, do people use multiple map orientations when trying to solve a problem? In particular, it seems that you would want military officers to not have whatever subconscious biases go along with north being "up", east being "right", and so on. So do army or navy exercises ever deliberately use maps with other orientations -- maybe southeast at the top -- as a way of freeing commanders from possible map-based biases?

Maybe there are other professions -- air traffic controllers, petroleum geologists, maybe even highway planners -- where you might want to make an extra effort to root out latent biases.

(08-06-2013 03:00 PM)georgewebb Wrote:  

(08-06-2013 09:18 AM)Caelligh Wrote:  If you re-drew Texas based on population, for example, maybe it would then be easier to draw boundaries that are free of bias because you would be forced to look at the task in an unconventional way.

Your point about unconventional thinking about maps reminds me: I have often wondered if, in professions where being able to think in an unbiased way about geography is essential, do people use multiple map orientations when trying to solve a problem? In particular, it seems that you would want military officers to not have whatever subconscious biases go along with north being "up", east being "right", and so on. So do army or navy exercises ever deliberately use maps with other orientations -- maybe southeast at the top -- as a way of freeing commanders from possible map-based biases?

Maybe there are other professions -- air traffic controllers, petroleum geologists, maybe even highway planners -- where you might want to make an extra effort to root out latent biases.

"When the map and the terrain disagree, trust the terrain."

(08-06-2013 03:00 PM)georgewebb Wrote:  ...Maybe there are other professions -- air traffic controllers, petroleum geologists, maybe even highway planners -- where you might want to make an extra effort to root out latent biases.

As a kind of highway planner myself, it could be a very interesting exercise at the alternatives analysis stage with respect to assessing the magnitude of impacts on the environment, on Title VI communities, etc. If I can figure out how to create the distorted maps, I may try it out on a future project.

How can at-large seats be part of a solution if you still value minority representation? Are there safeguards you would build in to prevent, for example, a 51% white county from having an all-white delegation?