Another setback for the Right Wing and the GOP

[Tigers!]

You yanks have more trouble than I realised. If you cannot appoint a neutral body for this then you are truly stuffed.

You don't look at voter affiliation, just look at population.

That is why compulsory voting beats voluntary. Voter affiliation is not an issue.

How do you appoint a non-partisan committee when the appointers are partisan?

I should find out how we do it in Australia. We have been doing it seemingly quite successfully since the Commonwealth Electoral Act 1918.

--------------------------------------------------------------------
Here is a quick and dirty primer https://en.wikipedia.org/wiki/Redistribution_(Australia)

 

[Loren Pechtel]

You yanks have more trouble than I realised. If you cannot appoint a neutral body for this then you are truly stuffed.

You don't look at voter affiliation, just look at population.

That is why compulsory voting beats voluntary. Voter affiliation is not an issue.

How do you appoint a non-partisan committee when the appointers are partisan?

At the very least, you appoint an independent entity and stop allowing the politicians themselves to take part in redrawing district lines.

Are you really waiting for perfection before you would agree to remove the politicians themselves from the process?

You seem to have missed my position on this. I don't believe we can get a truly impartial board to draw the lines so I think the lines should be drawn by formula. Take choice out of the picture and it doesn't matter if you have partisan people doing it.

 

[Loren Pechtel]

You yanks have more trouble than I realised. If you cannot appoint a neutral body for this then you are truly stuffed.

You don't look at voter affiliation, just look at population.

That is why compulsory voting beats voluntary. Voter affiliation is not an issue.

How do you appoint a non-partisan committee when the appointers are partisan?

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

Observations:

1) We have some things that are supposed to be equal numbers of Democrats and Republicans. Oops, when Democrats leave the GOP hasn't been appointing replacements.

2) The system you describe can also be used to deadlock when the status quo favors the party who is gaming the system.

 

[bilby]

At the very least, you appoint an independent entity and stop allowing the politicians themselves to take part in redrawing district lines.

Are you really waiting for perfection before you would agree to remove the politicians themselves from the process?

You seem to have missed my position on this. I don't believe we can get a truly impartial board to draw the lines so I think the lines should be drawn by formula. Take choice out of the picture and it doesn't matter if you have partisan people doing it.

That doesn't change the problem though. How do you get an impartial committee to codify your formula, and how do you ensure that it's not later amended in a partisan way?

If you can do that (and you can), why not allow that committee to set the district boundaries, such that non-partisan considerations not part of the original formula can be taken into account?

 

[bilby]

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

Observations:

1) We have some things that are supposed to be equal numbers of Democrats and Republicans. Oops, when Democrats leave the GOP hasn't been appointing replacements.

2) The system you describe can also be used to deadlock when the status quo favors the party who is gaming the system.

I have very little patience with your 'Americans stuff it up, therefore it's impossible to do it right' principle. You base a lot of arguments on this foundation, but a quick glance at the other 95% of the world says it is nonsense.

 

[Jason Harvestdancer]

You yanks have more trouble than I realised. If you cannot appoint a neutral body for this then you are truly stuffed.

You don't look at voter affiliation, just look at population.

That is why compulsory voting beats voluntary. Voter affiliation is not an issue.

How do you appoint a non-partisan committee when the appointers are partisan?

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

Each party. Yep. There's no way that would hurt third parties.

This is easy though. Write a computer program with publically auditable source code. Program it with population information and natural boundaries. Have it draw the districts.

 

[bilby]

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

Each party. Yep. There's no way that would hurt third parties.

You have first past the post simple majority voting, so third (and subsequent) parties are irrelevant.

This is easy though. Write a computer program with publically auditable source code. Program it with population information and natural boundaries. Have it draw the districts.

Sure. Who writes the software?

A computer program is not fundamentally different from a biro in this context - it draws boundaries where the person in control of it wants them, unless he's incompetent.

 

[Jason Harvestdancer]

If the code is publically auditable, it doesn't matter who writes the code. That's why I included "publically auditable". Anyone, including members of third parties, can check the code.

OTOH, since third parties are irrelevant, perhaps it would be best to staff the board that draws the districts with third party voters.

 

[Elixir]

If the code is publically auditable, it doesn't matter who writes the code. That's why I included "publically auditable". Anyone, including members of third parties, can check the code.

OTOH, since third parties are irrelevant, perhaps it would be best to staff the board that draws the districts with third party voters.

A laudable idea, but ... since only a tiny fraction of the public are code-literate, the rest of them are subject to"alternative facts" about it. And if those alternative facts are endorsed by say, a Donald Trump, very few will believe it's actually non-partisan. Remember - the orange turd has tens of millions of people convinced that Mueller, Comey, McCabe etc. are all DEMOCRATS! And that is a mundane, demonstrably alternative "fact". It isn't some esoteric assertion like "this code is non-partisan".

 

[Bomb#20]

At the very least, you appoint an independent entity and stop allowing the politicians themselves to take part in redrawing district lines.

Are you really waiting for perfection before you would agree to remove the politicians themselves from the process?

You seem to have missed my position on this. I don't believe we can get a truly impartial board to draw the lines so I think the lines should be drawn by formula. Take choice out of the picture and it doesn't matter if you have partisan people doing it.

Why are all you guys trying to band-aid a broken neck? What the heck are lines good for? What the heck are districts good for? When my neighbor and I disagree about practically everything, why the heck does anyone believe it's even theoretically possible for the government to command that the same individual will "represent" both of us in the legislature and have the results be fair in any sane sense of the word? Geographic districts are an 18th-century solution to an 18th-century problem caused by 18th-century communication technology. Abolish the dumb things. Let me and my neighbor both be represented, the only way we can be: by two different people.

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

Each party. Yep. There's no way that would hurt third parties.

^^^^^^^ This ^^^^^^^

You have first past the post simple majority voting, so third (and subsequent) parties are irrelevant.

If we get rid of districts we get rid of first past the post simple majority voting.

 

[bilby]

Why are all you guys trying to band-aid a broken neck? What the heck are lines good for? What the heck are districts good for? When my neighbor and I disagree about practically everything, why the heck does anyone believe it's even theoretically possible for the government to command that the same individual will "represent" both of us in the legislature and have the results be fair in any sane sense of the word? Geographic districts are an 18th-century solution to an 18th-century problem caused by 18th-century communication technology. Abolish the dumb things. Let me and my neighbor both be represented, the only way we can be: by two different people.

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

Each party. Yep. There's no way that would hurt third parties.

^^^^^^^ This ^^^^^^^

You have first past the post simple majority voting, so third (and subsequent) parties are irrelevant.

If we get rid of districts we get rid of first past the post simple majority voting.

If.

 

[Bomb#20]

If.

There are lots of ways you can do that. For example, you can establish a board of appointers whose rules require it to consist of equal or approximately numbers of members of each party, and also require a supermajority or even unanimity for any decisions. Such a board cannot make appointments at all unless the parties agree that a candidate is not partizan.

If.

Your way depends on the members of your committee deciding IF they want to compromise. You'll get reasonable boundaries IF both sides decide that's in their interest. Otherwise they will simply fail to reach the required supermajority. Each party has veto power, and whichever party the status quo favors will have an incentive to exercise it.

If by "If.", what you mean is that incumbent legislators won't enact my proposal because they don't have an incentive to abolish geographical districts, yes, that's true; but incumbent legislators also don't have an incentive to appoint a neutral body that will try to be fair to everyone. Legislators benefit from gerrymandered safe seats. From the achievability point of view, the great merit of abolishing districts is that California voters could directly abolish our state representatives' single-member districts, by referendum, whether the legislators like it or not. In contrast, the referendum process provides us the people with no power to abolish corrupt voting behavior by the appointees to any "neutral" district-drawing committee we might establish.

 

[Loren Pechtel]

You have first past the post simple majority voting, so third (and subsequent) parties are irrelevant.

This is easy though. Write a computer program with publically auditable source code. Program it with population information and natural boundaries. Have it draw the districts.

Sure. Who writes the software?

A computer program is not fundamentally different from a biro in this context - it draws boundaries where the person in control of it wants them, unless he's incompetent.

Make the formula and raw data public information. A dirty programmer will certainly be exposed.

- - - Updated - - -

If the code is publically auditable, it doesn't matter who writes the code. That's why I included "publically auditable". Anyone, including members of third parties, can check the code.

OTOH, since third parties are irrelevant, perhaps it would be best to staff the board that draws the districts with third party voters.

I've seen code with some pretty subtle deceptions in it--dirty code might not be caught. However, if the inputs are public (and they should be) anyone can write code and compare the results. That's how I would expect deception to be detected.

 

[T.G.G. Moogly]

Your way depends on the members of your committee deciding IF they want to compromise. You'll get reasonable boundaries IF both sides decide that's in their interest. Otherwise they will simply fail to reach the required supermajority. Each party has veto power, and whichever party the status quo favors will have an incentive to exercise it.

I can downright depressed thinking about how people are more loyal to party than to the law. It's all about power, not constitutional mandate. I think the era of compromise is over, if there ever was one.

My hope is that when SCOTUS decides these cases Roberts doesn't go with party.

I've seen code with some pretty subtle deceptions in it--dirty code might not be caught. However, if the inputs are public (and they should be) anyone can write code and compare the results. That's how I would expect deception to be detected.

Yes. If someone is cooking the books/code it will be known. Hopefully we can get at least that far.

 

[bilby]

Your way depends on the members of your committee deciding IF they want to compromise. You'll get reasonable boundaries IF both sides decide that's in their interest. Otherwise they will simply fail to reach the required supermajority. Each party has veto power, and whichever party the status quo favors will have an incentive to exercise it.

I can downright depressed thinking about how people are more loyal to party than to the law. It's all about power, not constitutional mandate. I think the era of compromise is over, if there ever was one.

My hope is that when SCOTUS decides these cases Roberts doesn't go with party.

I've seen code with some pretty subtle deceptions in it--dirty code might not be caught. However, if the inputs are public (and they should be) anyone can write code and compare the results. That's how I would expect deception to be detected.

Yes. If someone is cooking the books/code it will be known. Hopefully we can get at least that far.

A computer adds nothing except perhaps speed.

If the rules and process are transparent, a committee can do the job - and probably do it better than a computer program, because unlike a computer, a committee of humans can reject or modify crazy solutions that fit the letter of the rules, but breach their spirit. Real voting districts are not all neat grids of streets with clear geographical features - and all maps include errors.

Suggesting that a computer can do a better job is a category error - a computer is analogous to a pen, not to a redistricting official.

When people say 'why not use a computer to ensure fairness', I hear 'why not use a pen to draw the lines, because a pen cannot be biased'.

Computers are tools; Not magic decision making boxes.

All the problems with a committee that sets boundaries still exist if you change to a system where a committee writes the algorithms that are used to program a computer to set boundaries. It's just kicking the can down the road.

 

[Jason Harvestdancer]

My suggestion includes that the public can audit the code. The only valid objection that I've seen is Loren's that you have a subtle programmer writing dirty code. It is not analogous to using a pen instead of a person.

Yes, most people are not code literate, but there are code literate people in both major parties and all third parties, and maybe even in the minor parties.

 

[Loren Pechtel]

A computer adds nothing except perhaps speed.

If the rules and process are transparent, a committee can do the job - and probably do it better than a computer program, because unlike a computer, a committee of humans can reject or modify crazy solutions that fit the letter of the rules, but breach their spirit. Real voting districts are not all neat grids of streets with clear geographical features - and all maps include errors.

Suggesting that a computer can do a better job is a category error - a computer is analogous to a pen, not to a redistricting official.

When people say 'why not use a computer to ensure fairness', I hear 'why not use a pen to draw the lines, because a pen cannot be biased'.

Computers are tools; Not magic decision making boxes.

All the problems with a committee that sets boundaries still exist if you change to a system where a committee writes the algorithms that are used to program a computer to set boundaries. It's just kicking the can down the road.

This is an optimization problem. A key aspect of my approach is that you choose the best solution--it's unique, thus anyone else working with the same inputs will come up with exactly the same answer. Such problems are almost always very compute-intensive, it would be unlikely that the correct solution could be found manually.

 

[bilby]

A computer adds nothing except perhaps speed.

If the rules and process are transparent, a committee can do the job - and probably do it better than a computer program, because unlike a computer, a committee of humans can reject or modify crazy solutions that fit the letter of the rules, but breach their spirit. Real voting districts are not all neat grids of streets with clear geographical features - and all maps include errors.

Suggesting that a computer can do a better job is a category error - a computer is analogous to a pen, not to a redistricting official.

When people say 'why not use a computer to ensure fairness', I hear 'why not use a pen to draw the lines, because a pen cannot be biased'.

Computers are tools; Not magic decision making boxes.

All the problems with a committee that sets boundaries still exist if you change to a system where a committee writes the algorithms that are used to program a computer to set boundaries. It's just kicking the can down the road.

This is an optimization problem. A key aspect of my approach is that you choose the best solution--it's unique, thus anyone else working with the same inputs will come up with exactly the same answer. Such problems are almost always very compute-intensive, it would be unlikely that the correct solution could be found manually.

It's very unlikely that the provably optimum solution for any reasonably large number of districts and voters is even computable in a reasonable amount of time.

 

[Loren Pechtel]

A computer adds nothing except perhaps speed.

If the rules and process are transparent, a committee can do the job - and probably do it better than a computer program, because unlike a computer, a committee of humans can reject or modify crazy solutions that fit the letter of the rules, but breach their spirit. Real voting districts are not all neat grids of streets with clear geographical features - and all maps include errors.

Suggesting that a computer can do a better job is a category error - a computer is analogous to a pen, not to a redistricting official.

When people say 'why not use a computer to ensure fairness', I hear 'why not use a pen to draw the lines, because a pen cannot be biased'.

Computers are tools; Not magic decision making boxes.

All the problems with a committee that sets boundaries still exist if you change to a system where a committee writes the algorithms that are used to program a computer to set boundaries. It's just kicking the can down the road.

This is an optimization problem. A key aspect of my approach is that you choose the best solution--it's unique, thus anyone else working with the same inputs will come up with exactly the same answer. Such problems are almost always very compute-intensive, it would be unlikely that the correct solution could be found manually.

It's very unlikely that the provably optimum solution for any reasonably large number of districts and voters is even computable in a reasonable amount of time.

Why do you think it's NP-hard?

 

[bilby]

It's very unlikely that the provably optimum solution for any reasonably large number of districts and voters is even computable in a reasonable amount of time.

Why do you think it's NP-hard?

Why do you think it's not?