IF the election results were reversed...

[Gravekeeper]

Not a bad idea, but we'd have to make sure the person programming the computer isn't partisan, and that the computer cannot be hacked.

Well, math isn't partisan and this would be a straight forward math problem all things considered. If the calculation side of the code was open source it would effectively be impossible to hijack. Because anyone could compile the program and input the data to verify the official results themselves.

Any discrepancy would be apparent neigh instantly when the official results were released. Its effectively impossible to compromise an open source system and avoid detection. Compared to compromising a singular closed doors system no one but the people you already don't trust have the keys too.

[D_Yeti_Esquire]

Well a computer can't be partisan, but as Gravekeeper points out - if the code is open source then anything anything it comes up with can be verified. Remember, we're only talking about districting here - so verification that the districting is correct would happen well before an election. If I have a burr in my saddle and I can prove the math is wrong, someone screwed up.

Where I think we're stopping down is we're so focused on "partisan" as a thing, we're neglecting something. Something is mathematically true or its not.

You can be the hardest of hard core conservatives and program a system and the only way its an issue is if your outcome deviates from the expected.

I picked four arbitrary factors and assigned weight(racial represetation can't be minimized, equal population, safe districts minimized, safe districts equalized as much as possible.) There's actually only one correct districting that is going to create that once you assign weighting. Your districts might look crazy - but they're either going to be mathematically and statistically verifiable or not.

Your results might not follow - but at a certain point you can only attack causes of odd voting outcomes. But you can't force voters to the polls.

[TheHuckster]

Well, math isn't partisan and this would be a straight forward math problem all things considered. If the calculation side of the code was open source it would effectively be impossible to hijack. Because anyone could compile the program and input the data to verify the official results themselves.

One thing to watch out for is the data itself could still vary depending on what source you use. We do a census every 10 years, and data in between are estimates. Much of the data is pretty static, but especially in areas with high rates of growth or shrinkage, such as Austin, Detroit, and Las Vegas, numbers will change drastically in those ten years. So, someone who plugs in estimated Austin data for the current year when the official results were pulled from the last census will falsely report "different results."

And, of course, it doesn't matter if the dumbass didn't do it right, people will take those figures and run with it and claim it's off base.

[Talon]

...and that the computer cannot be hacked.

There is no such thing as a computer system that can't be hacked.

When I say computer system, I'm not just referring to software and hardware. People are part of that system, and it's disturbingly easy to trick people into clicking a suspicious link in an email.

I'd think with some good, solid security measures (data encryption, strong passwords) the possibility of a hack could be reduced.

All good starting points, but what seems secure today may have vulnerabilities you're not aware of until later. Any changes in the environment, be it software or hardware could introduce new security holes. The system would have to be tested for weaknesses and patched as part of an ongoing process.

[s_stabeler]

it's worth noting that a system does not have to be perfect- it just has to be better than the alternatives. Yes, a computer would have it's own flaws- but would it be better than the existing system- where, ironically, a computer is ALREADY used, IIRC, it's just that it is configured to draw up gerrymandered districts. (considering that AFAIK both sides use the exact same program, just configured to bias a particular party, you could use the same exact program, just with the bias set to 0.)

[D_Yeti_Esquire]

Here's the thing though - the code is open source.

The data set by law is going to come from the census or public info.

That means if you do it as Gravekeeper/stabler suggest, you don't have to worry about a computer being hacked in the sense that it really doesn't matter. The government runs it but any Tom, Dick, Harry, or IBM can run the same program and detect if the results are off. Compromise the computer - its fine. The bad results get detected and you could do this through hashing the source before its compiled, hashing the program executable, hashing the source data. If those hashes change, the admin knows - simply revert to a previous version.

At that point the risk/reward of tampering with the system especially when districting occurs years before an election actually takes place is completely off. It would be very easy to detect and very easy to fix. And because we're keeping the process open, you're going to have busybodies in both parties going through the code with a fine tooth comb plus data scientists who by their very nature tend to try to continually optimize.

[HYHYBT]

But even if electors were apportioned by which candidate "won" a Congressional District (thus splitting the number of electoral votes among the candidates, sort of like delegates in a Primary), I don't know how much that would matter.

"Even if?" That's a "solution" deliberately calculated (though presumably not by you) to extend the gerrymander to presidential elections! OF COURSE it wouldn't make things better; just the opposite!

Again, would that be the sentiment if the results were flipped? If Trump had taken Hillary's approach, and she won the EC and he won the Popular Vote, I'm sure a lot of Democrats would be thanking their lucky stars for the EC, and saying "It worked like it was supposed to!"

Even if "you wouldn't be complaining if you'd won" were a valid argument, an EC that simply threw in an extra element of chance would still be fair: the side that it causes to lose today, it may cause to win next time, with things evening out overall. But this, instead, is a system that consistently puts its thumb on the same side of the scale. No Democrat has EVER won the EC and not the popular vote.

...but it's also not supposed to be where highly populated areas decide the whole thing, too.

Why the hell not? Especially since that would only be true when there aren't enough other people to outvote them. One person, one vote, EQUALLY WEIGHTED AT THE FINAL COUNT is inherently fairer than "my vote counts more than yours because I live in Wyoming."

I honestly think it's pretty simple. The computing horsepower exists (which is why almost nothing is competitive anymore legislatively.) Amend the constitution that voting districts require 5 things: 1) Do not suppress racial voting(existing), 2) are apportioned on population evenly(existing), 3) the number of competitive districts are maximized, 4) The number of safe districts are even within 5-10%, 5) The test for the meeting of this requirement is House and Senate representation within 5-10(or within 1 vote since some states have very few Reps)% of total state #'s.

Three is too vague. 4 and 5 are a terrible idea because they require what isn't even going to be possible in some cases, and even where they would, would involve the same sort of convoluted, artificial district lines we have now.

Make it an equation that doesn't care.

This is the opposite of your proposal to require districts be drawn to keep the number of safe seats close together... or, in case I misunderstood, to keep the number of safe seats in total as small as possible. Either way requires a formula that cares very much who would vote how.

[mjr]

Why the hell not? Especially since that would only be true when there aren't enough other people to outvote them. One person, one vote, EQUALLY WEIGHTED AT THE FINAL COUNT is inherently fairer than "my vote counts more than yours because I live in Wyoming."

Your last quote is somewhat true, anyway. Look at the EC breakdown as to people per elector.

Even so, I'll try to answer your first question. Are you saying that "middle America" (i.e. most of the "red" states) doesn't/don't count? Because what you're proposing would result in cities determining the President -- every time.

The top 10 cities, according to Wikipedia, have approximately 25,861,298 people. That's just under 10% of the population.

And if we did it that way, these ten cities would probably decide the Presidential election every time. Most of those cities are generally in "liberal" areas.

Are you really OK with the same 10 cities deciding the Presidency every time? Is it an "as long as I agree with them" kind of thing?

Even if it were broken down by county. "Red" counties far outnumbered "blue" counties in this election. Similar things happened by Congressional district.

[s_stabeler]

if you look at the demographics, then as I pointed out earlier, it's suburban voters that actually swing it. Anyway, the problem is that while it is a laudable aim to avoid disenfranchising rural areas, in a sense the current system has the opposite effect: disenfranchising the cities ( and yes, it does: in the last election, 34% of voters were in cities. 17% were in rural areas. the suburban vote was pretty much 50-50. The cities voted overwhelmingly for Clinton ( about a 30-point margin) while rural areas voted about the same for Trump. As such, by rights, there should have been a tied election- since the implication is that the US was pretty much split 50-50 between the candidates. Instead, he had an 80-vote lead in the Electoral College. In short, in some ways, rural votes seem to count for as much as 3 or 4 city votes.)

to make it clear: I do agree that measures should be taken to ensure that cities don't dominate to the exclusion of rural areas. However, as it stands, the opposite is happening- there needs to be a balance, and at the moment, there isn't one.

[mjr]

to make it clear: I do agree that measures should be taken to ensure that cities don't dominate to the exclusion of rural areas. However, as it stands, the opposite is happening- there needs to be a balance, and at the moment, there isn't one.

IIRC (and I could be wrong about this), the way the EC originally worked was the electors were voted for by Congressional district. So let's say you have 10 districts. You might have 6 electors of one party, and 4 of the other selected, based on how the electorate voted. Because you were voting for "electors" who promised to vote for the candidate who won. I forget how it worked with the Senate, but IIRC it used to work something like that.

So then instead of the "winner take all" that most states use, each state would have it's electors match whomever won the district.

IIRC it used to be that way. I could be VERY wrong there.

But as has been stated previously, then you run into gerrymandering, and have to make sure that redistricting is done in a way that is as fair as possible.

[Gravekeeper]

IIRC (and I could be wrong about this), the way the EC originally worked was the electors were voted for by Congressional district. So let's say you have 10 districts. You might have 6 electors of one party, and 4 of the other selected, based on how the electorate voted. Because you were voting for "electors" who promised to vote for the candidate who won. I forget how it worked with the Senate, but IIRC it used to work something like that.

Originally, you were voting for an elector who would then do all the heavy lifting in terms of determining who was the best candidate then vote for them. Electors were independent ( and technically speaking still are ). The idea being that the common folk did not have the time nor access to information to be able to decide on the best candidate. So they decided on the best person to go figure out who the best candidate for them was.

Oh, and to make it "fair" to the states that had lots of slaves that thought said slaves should count for something at least. >.>

[wolfie]

One person, one vote, EQUALLY WEIGHTED AT THE FINAL COUNT is inherently fairer than "my vote counts more than yours because I live in Wyoming."

Someone already brought up the issue of cities having a very high proportion of the population. Did a quick search, and Wikipedia says Wyoming has a population of 586,107. Manhattan has a population of 1,644,518, Brooklyn 2,636,735, Queens 2,339,150, the Bronx 1,385,108, and Staten Island 474,558.

Of the 5 boroughs of New York City, only one has a population less than the entire state of Wyoming. Adding them up gives the whole city a population of 8,480,069 people, or roughly 14 1/2 times the population of Wyoming. If all votes were weighted equally, entire states would be effectively disenfranchised due to their population being outweighed by that of the large cities.

Because the population density is not uniform across the country, both extremes (more people per riding in cities than in rural areas, and equal number of people in all ridings) have elements of unfairness, and any action taken to correct one issue would make the other worse.

[s_stabeler]

The problem is that under the current system, the unfairness is too extreme. In this last election, there was a rural/urban split, with suburbia down the middle. That should have produced a result split more or less 50-50 in the Electoral College. Instead, it produced a convincing Trump majority in the Electoral College. THAT is the problem- the rural vote, provided the suburban vote is split, can overwhelm the city one.

[D_Yeti_Esquire]

Because the population density is not uniform across the country, both extremes (more people per riding in cities than in rural areas, and equal number of people in all ridings) have elements of unfairness, and any action taken to correct one issue would make the other worse.
Reply With Quote

Yes but I'm not sure that's the point. Federalism in its original design didn't seek to equalize all branches of government between each other. What it sought to do is offset one problem with another. "Liberty" was ensured by counteracting the advantages they figured would exist.

The Senate was your original "protect the less populous states" strategy. But the original design didn't forsee super-dense megacities. And the original design wasn't adhered to when new states were drawn. So instead of seeing an expected advantage in the House offset by states like Wyoming being important in the Senate - actually the opposite has sort of happened. The Senate is the one organization that sort of cancels out gerrymandering by being representative of the entire state. So it ends up being more representative than the House for urban voters. The closer a state is to having equal # urban and rural voters, the more representative it is.

That's really the problem - if we see Rural voters racking up #'s in one branch of Government - not a problem. When they're doing it in all the branches simultaneously without a population reason for that occurring, that's actually a big friggin problem. Liberty is not being protected in that case. An urban voter has no ability to stop a Rural voter from "meddling". And if you actually look at some of the stuff coming out of Texas, you now have large portions of city budgets being withheld by the State unless they accept State mandates. Again - in and of itself not a problem, but when cities are disproportionately electorally disenfranchised from those decisions - problem. And that's become a bigger issue as states political parties have become more in-line with national politics. When an Arizona senator arguably has as much to do with how Dallas city policy because the National Republican party is exercising that control, the entire purpose of Federalism is short-circuited.

This could have always been a problem in the past, but if we rewind to the 70's there was more policy diversity in both parties. It's really become an acute problems as that diversity has gone away.

[mjr]

That's really the problem - if we see Rural voters racking up #'s in one branch of Government - not a problem. When they're doing it in all the branches simultaneously without a population reason for that occurring, that's actually a big friggin problem.

Is the converse true?

Again - in and of itself not a problem, but when cities are disproportionately electorally disenfranchised from those decisions - problem.

Does that apply to rural areas, as well (i.e. is the "converse" true)? I'm asking because you seem like a reasonable person to have a discussion with, and I'd honestly like your opinion on it.