Indiana Purging Voters Using Software That’s 99% Inaccurate • Hacker News Board

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#61

Post by fencepost » Mon, Oct 30, 2017, 7:45 PM UTC

ThrustVectoring wrote: post #50

pulisse wrote: post #7

De rigueur pointer to a discussion of the birthday paradox for those who don't understand why the combination of first name, last name, and birth date is an unavoidably flawed way of uniquely identifying individuals: https://www.scientificamerican.com/article/bring-science-hom... .

Personal anecdote: my parents used a pediatrician who identified minor patients by first + last + birth month. The office had to start using middle initial as well to disambiguate me from another patient.

Separate personal anecdote. My wife changed gynecologists after receiving a message on our answering machine with very sensitive and concerning medical data that was not for her.

She hadn't even seen the doctor in question for more than 9 months, had never had any indication of the problems in question (but had had tests done that could have identified such) and was generally thoroughly freaked out by this. Turns out the practice had another patient with the same name and was just a little sloppy about which person got called.

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#62

Post by liquidise » Mon, Oct 30, 2017, 7:45 PM UTC

brian-armstrong wrote: post #39

alistproducer2 wrote: post #26

Anyone with half a brain knows that these purges are designed to advantage a side not protect the system. I read an article the other day that stated that some America's voting system had essentially been compromised on many levels. It made me think about why so many of our polls have, in the last few years, been so off the mark. Now I'm beginning to think it's not because the polling models suddenly don't work. It i…

Yeah, I remember the outrage when the original Diebold hacking announcement was made in the Bush Jr era. The concerns in that were never addressed and doubtless things have simply since got worse. If Google, Facebook et al wanted to make amends for biasing the election, they could bring some of their engineering prowess to bear on defeating systematic voting machine fraud.

> Google, Facebook et al wanted to make amends for biasing the election

I'm still surprised by the anger for this. For the purposes of the viewer, Google and Facebook are effectively media companies. By that measure, all media companies bias elections heavily during election cycles. This is done both with content and well as paid advertising.

But to "make amends" for something sounds as if their influence is either unprecedented or morally wrong. Perhaps the argument is that the bias from international influences was wrong. Something else entirely?

I suppose i fail to see where the real complaint here is. What behavior or lapse triggered these accusations?

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#63

Post by CalChris » Mon, Oct 30, 2017, 7:45 PM UTC

drzaiusapelord wrote: post #55

alistproducer2 wrote: post #26

Anyone with half a brain knows that these purges are designed to advantage a side not protect the system. I read an article the other day that stated that some America's voting system had essentially been compromised on many levels. It made me think about why so many of our polls have, in the last few years, been so off the mark. Now I'm beginning to think it's not because the polling models suddenly don't work. It i…

The 2016 polling models were pretty good, or at least, as good as they are usually[1]. The problem is, it was a very close race by electoral college standards. Trump ultimately won by 80,000 votes in three states in an election with 136m votes cast[2]. So the 'deciding' votes were just 0.06% of the vote total. I think we can have it both ways. Voter suppression is being done and we have statistical heats in competiti…

It would take a Constitutional Amendment to get rid of the Electoral College. That’s 2/3rd of both houses of Congress and 2/3rds of the State legislatures.

Better to call a third Constitutional Convention.

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#64

Post by corey_moncure » Mon, Oct 30, 2017, 7:53 PM UTC

brian-armstrong wrote: post #39

alistproducer2 wrote: post #26

Anyone with half a brain knows that these purges are designed to advantage a side not protect the system. I read an article the other day that stated that some America's voting system had essentially been compromised on many levels. It made me think about why so many of our polls have, in the last few years, been so off the mark. Now I'm beginning to think it's not because the polling models suddenly don't work. It i…

Yeah, I remember the outrage when the original Diebold hacking announcement was made in the Bush Jr era. The concerns in that were never addressed and doubtless things have simply since got worse. If Google, Facebook et al wanted to make amends for biasing the election, they could bring some of their engineering prowess to bear on defeating systematic voting machine fraud.

What does the fact that they don't do so, tell you?

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#65

Post by fencepost » Mon, Oct 30, 2017, 7:54 PM UTC

kevin_thibedeau wrote: post #3

New York purged me back in 2010 and I was forced to vote provisional despite voting there consistently for many years. The fail extends all around.

And odds are high that your provisional ballot was simply tossed unless you contacted your county Board of Elections, probably with documentation to prove that you were allowed to vote and probably within just a few days after the election.

What happens with provisional ballots is very highly variable, but quite frequently ends up with them simply being tossed.

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#66

Post by craftyguy » Mon, Oct 30, 2017, 7:56 PM UTC

alistproducer2 wrote: post #26

Anyone with half a brain knows that these purges are designed to advantage a side not protect the system. I read an article the other day that stated that some America's voting system had essentially been compromised on many levels. It made me think about why so many of our polls have, in the last few years, been so off the mark. Now I'm beginning to think it's not because the polling models suddenly don't work. It i…

Pretty much. I highly suspect the shit security of the voting system is by design, not some 'accident.'

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#67

Post by thephyber » Mon, Oct 30, 2017, 7:59 PM UTC

CalChris wrote: post #63

drzaiusapelord wrote: post #55

Earlier quoted context omitted.

The 2016 polling models were pretty good, or at least, as good as they are usually[1]. The problem is, it was a very close race by electoral college standards. Trump ultimately won by 80,000 votes in three states in an election with 136m votes cast[2]. So the 'deciding' votes were just 0.06% of the vote total. I think we can have it both ways. Voter suppression is being done and we have statistical heats in competiti…

It would take a Constitutional Amendment to get rid of the Electoral College. That’s 2/3rd of both houses of Congress and 2/3rds of the State legislatures. Better to call a third Constitutional Convention.

There is another way to get a similar effect.[1]

[1] https://en.wikipedia.org/wiki/National_Popular_Vote_Intersta...

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#68

Post by wang_li » Mon, Oct 30, 2017, 8:10 PM UTC

JumpCrisscross wrote: post #48

wang_li wrote: post #33

Earlier quoted context omitted.

It doesn't get charged often because there are parties that fight tooth and nail to prevent the implementation of processes that would allow the detection of ineligible voters voting. It is shown in every election cycle that there are cases of voters voting who should not be because of their citizenship or felon status or voting in multiple districts. When you say that only 2 have been charged, why is that relevant?…

> It is shown in every election cycle that there are cases of voters voting who should not be because of their citizenship or felon status or voting in multiple districts. Source? Because every [1] source [2] I find [3] says the opposite. [1] http://library.uwp.ac.id/digilib/files/disk1/13/--publicadmi... [2] http://open.mitchellhamline.edu/cgi/viewcontent.cgi?article=... [3] http://www.nytimes.com/2007/04/12/washing…

Your own links say that it happens, but then hand wave it away as not significant.

Your [1]:

>However, in only 26 cases was there a conviction or a guilty plea by the citizen.

From your [2]:

>n the state of Washington, for example, six cases of double voting and nineteen instances of individuals voting in the name of the dead yielded twenty-five fraudulent votes

From your [3]:

>Although Republican activists have repeatedly said fraud is so widespread that it has corrupted the political process and, possibly, cost the party election victories, about 120 people have been charged and 86 convicted as of last year.

I'm not saying it's widespread, but I suspect that there have been local elections where the outcome was changed due to fraudulent votes. The question is do we care that each fraudulent vote voids the vote of a legitimate voter? And if something such as voter ID requirements were in place, would that have any value? Would requiring states to cross check their voter registration have any value?

As an aside, people seem to enjoy comparing the US to places like Canada and Germany on many different things, health care, progressive tax rates, commitment to green policies, immigration, etc. yet surprisingly no one ever says brings up voter ID, which both of those countries require.

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#69

Post by thephyber » Mon, Oct 30, 2017, 8:10 PM UTC

mox1 wrote: post #2

Too add a data point, here is a list of the 10 most common pairs in the United States. https://blogs.ancestry.com/cm/calling-james-smith-10-most-co... I hesitate to guess ethnicity based on names, but I see mostly Caucasian and Latin names (Smith is represented heavily).

I think the goal is less to target racial groups directly, but rather to target those who are less likely to have the desire, the knowledge, or the resources to rectify the issue.

This tends to target the poorest Americans -- either those who can't afford to take time away from work to get the documents required to re-register or those who simply would give up after finding out they are not registered.

Re: Indiana Purging Voters Using Software That’s 99% Inaccurate

#70

Post by fencepost » Mon, Oct 30, 2017, 8:12 PM UTC

wang_li wrote: post #33

ch4s3 wrote: post #10

Earlier quoted context omitted.

That's complete nonsense. The Supreme Court and several district courts have found states like North Carolina to be actively using purges and registration changes for the purpose of disenfranchising voters. Furthermore, if you want to protect the integrity of the vote, registration is the last logical place to start, as in person fraud is vanishingly rare. Only 2 people were charged with doing so in this past electio…

It doesn't get charged often because there are parties that fight tooth and nail to prevent the implementation of processes that would allow the detection of ineligible voters voting. It is shown in every election cycle that there are cases of voters voting who should not be because of their citizenship or felon status or voting in multiple districts. When you say that only 2 have been charged, why is that relevant?…

You claim that there are a significant number of fraudulent votes happening. The corollary of what you're attesting to is that Kris Kobach and many other folks responsible for enforcement of the laws are utterly and completely incompetent.

I don't have time to track down where I did the math, but even being very generous with the claims of fraud (assuming that they're wildly overstating) and assuming that Kansas being full of those Fine Upstanding Conservative Religious Citizens had only half the fraud rate of the country as a whole, that means that Kobach's office was unable to find basically ANY of the tens of thousands of Kansans voting fraudulently.

This is a situation that holds true when you look at the other places where there are widespread claims of voter fraud and the people making those claims. Sure there are a few cases (e.g. Ann Coulter's fraudulent registration, a few voters who voted in multiple places including at least one non-citizen, etc.) but not very many.

As for Kobach, he actually got 4 convictions in 2016 (for elections in 2010 - 2014): http://www.kansas.com/news/local/crime/article75545277.html

Looking at a different state, seems that Washington State is a hotbed of fraud - the state has identified 72 cases of suspected vote fraud in the 2016 election. https://www.usnews.com/news/best-states/washington/articles/...