Accusations of Electoral Fraud in NYC
Last week there was a diary at the top of the recommended list on Daily Kos citing voting problems in NYC with some accusations (in the commnets) that Hillary's people are disenfranchising black voters. A genuine red flag was raised when it was noticed that a heavily black election district in Harlem was recorded as having zero votes for Barack Obama. This is clearly unlikely and therefore genuinely raised people's concern. And it brings up several important issues regarding voting machines. But one thing this should NOT cause is accusations of fraud on the part of Clinton's campaign. From first hand experience, I can suggest that the problem was not an intentional undercount of Obama voters, but a maintenance problem with the voting machines used in NYC.
In NYC we use very old lever machines that record the votes using mechanical counters inside the machine. It is all mechanical with each vote for a candidate tuning a counter wheel inside by one digit. As long as these machines are properly set to zero before voting begins and the votes in the end are tallied in an open manner, and as long as the machines are not re-set before the vote is properly tallied, fraud can be avoided. Of course you can see how if an election board is corrupt there is room for fraud, but with bipartisan election boards and with a chance for any campaign or local political club to see the backs of the machines at the end of voting and record their own tally, the system seems pretty fair and open.
But there is one problem. The machines are very old and because it has been assumed for some time that they will eventually be replaced, they are not properly maintained. The mechanical wheels that record each vote are kind of sticky, so they don't always fully move to the next digit. I have seen this personally comparing tallies at the end of a primary election in 2006. If a wheel gets stuck going from 9 to 10 (or 19 to 20, or 29 t0 30, etc.) of from 99 to 100, someone looking at the back of the machine might either see it as 10 or zero (or 10 or 20, or...you get the idea). So it can happen that a tally gets undercounted by 10 or 100 votes simply because of the fact the machines are poorly maintained.
I first noticed this when I was looking in the back of a machine to get a tally for a friend running for office. I saw my count for him was 10 votes higher than the unofficial tally written down by the poll worker. I called him over and politely disagreed with his count. He was very nice about it, looked at the machine, and agreed with me, and changed his final tally.
The same night at the same polling place I noticed two other discrepancies between tallies another friend made and the tally made by the poll workers. This was after the machines were being sealed and the unofficial poll worker tallies finalized. In one case the poll worker recorded zero votes for Hillary running for Senate in one election district, and my friend had written down 100 votes. Once again the wheel was stuck between digits and one person interpreted it as zero votes, the other as 100 votes. Given that the people doing this are all tired after a long day, errors are bound to happen.
In theory all these machines are sealed at the end of election night, hauled to a warehouse and then later an official count is done. Presumably this official count is done far more carefully, though I have never been a witness to this so I can't vouch for it. Presumably any campaign can question tallies if their counts don't match the unofficial or official counts from the election boards. I welcome information on the final, official counting system.
So let's get back to the undercount in Harlem. The initial red flag of a district in Harlem having no votes for Obama in the unofficial tally by the poll worker is almost certainly due to this kind of mechanical problem leading someone to record what should have been 100 votes as zero votes. In theory it SHOULD be corrected when the official count is done. It is almost certain that Hillary Clinton had NOTHING TO DO WITH IT. People jumping to that conclusion are being way to quick to judge.
It also brings up questions about voting machines in general. Every system will have flaws. Every system will be open to some kind of fraud or another. The key is to have a system where fraud and errors CAN be detected and corrected.
Our lever machine in NYC, if the system is sufficiently transparent and monitored, and the machines were properly maintained, is a good one. The counts can be recorded by many people, both official poll workers and individual campaigns, newspapers, etc. before the machines are sealed. They are sealed in front of all these people then stored in a warehouse. Official counts are then made. Assuming these are properly done and monitored, and each campaign has a chance to check for discrepancies with the counts they made on election night, all will be well. The main problem is lack of maintenance.
PB/OS machines clearly have problems. Many of the problems in Ohio in 2004 were on PB/OS machines. But in the case of PB/OS machines there is still a paper ballot that can be recounted. There is a check on the official tallies if an election board handles the situation properly. The problem in Ohio was a corrupt election board dominated by one party, the Republicans. This meant that even court orders to check the counts were ignored by an election board that was partisan. New York has bipartisan election boards, so in theory we can avoid the fraud that probably took place in Ohio in 2004. Partisan election boards are also a problem in Florida. In 2004 Jimmy Carter's organization that monitors elections around the world and is highly respected around the world refused to monitor Florida's elections because they did not meet the minimum requirement for a fair election. The reason they did not qualify as a fair election was because they have a partisan election board. For PB/OS machines to be acceptable you need bipartisan (or maybe even multipartisan?) election boards and transparency. Maintenance will also, of course, be an issue.
Then we come to the DRE machines. These are the most flawed and unacceptable. Even with a good election board, you cannot have a fair election because there a.) is no paper trail, and b.) there is no transparency. With DRE machines it is all recorded digitally...even the tiny paper "receipts" that some machines give the voter are NOT an officially recognized paper trail. This means THERE CAN BE NO RECOUNT. There is no independent count for the vote that can be checked and compared like the lever machines in NYC or PB/OS machines have. Additionally there is no transparency. The software for doing the counts is owned by a company and that company keeps it secret. So no one knows what the machine does. If fraud were committed either by the company on behalf of a candidate they like or by a hacker, it could never be caught by the public. No paper trail...no recount...no independent check and balance...no transparency. These problems have led to the National Institute of Standards and Technology (NIST) to recommend decertification of DRE machines. NIST is a non-regulatory federal agency within the U.S. Commerce Department's Technology Administration that "promotes U.S. innovation and industrial competitiveness by advancing measurement science, standards, and technology in ways that enhance economic security and improve our quality of life." The decision by NIST to recommend decertification should be definitive. When the techies tell you the technology is unacceptable, we should listen.
Every system will have flaws. But proper maintenance of machines, fair and bi- or multi-partisan election boards, a paper trail, good opportunities for recounts if desired, and full transparency are the way to avoid problems with elections. This is really very simple, yet it isn't being done. DRE machines are the worst possible solution because they lack a paper trail, good opportunities for recounts if desired, and full transparency. PB/OS and mechanical machines have problems, but at least they allow for all of the above criteria for a fair election.
So let's stop blaming Hillary Clinton for an Obama undercount in Harlem and instead take this chance to recognize what we really do need for fair elections.
In NYC we use very old lever machines that record the votes using mechanical counters inside the machine. It is all mechanical with each vote for a candidate tuning a counter wheel inside by one digit. As long as these machines are properly set to zero before voting begins and the votes in the end are tallied in an open manner, and as long as the machines are not re-set before the vote is properly tallied, fraud can be avoided. Of course you can see how if an election board is corrupt there is room for fraud, but with bipartisan election boards and with a chance for any campaign or local political club to see the backs of the machines at the end of voting and record their own tally, the system seems pretty fair and open.
But there is one problem. The machines are very old and because it has been assumed for some time that they will eventually be replaced, they are not properly maintained. The mechanical wheels that record each vote are kind of sticky, so they don't always fully move to the next digit. I have seen this personally comparing tallies at the end of a primary election in 2006. If a wheel gets stuck going from 9 to 10 (or 19 to 20, or 29 t0 30, etc.) of from 99 to 100, someone looking at the back of the machine might either see it as 10 or zero (or 10 or 20, or...you get the idea). So it can happen that a tally gets undercounted by 10 or 100 votes simply because of the fact the machines are poorly maintained.
I first noticed this when I was looking in the back of a machine to get a tally for a friend running for office. I saw my count for him was 10 votes higher than the unofficial tally written down by the poll worker. I called him over and politely disagreed with his count. He was very nice about it, looked at the machine, and agreed with me, and changed his final tally.
The same night at the same polling place I noticed two other discrepancies between tallies another friend made and the tally made by the poll workers. This was after the machines were being sealed and the unofficial poll worker tallies finalized. In one case the poll worker recorded zero votes for Hillary running for Senate in one election district, and my friend had written down 100 votes. Once again the wheel was stuck between digits and one person interpreted it as zero votes, the other as 100 votes. Given that the people doing this are all tired after a long day, errors are bound to happen.
In theory all these machines are sealed at the end of election night, hauled to a warehouse and then later an official count is done. Presumably this official count is done far more carefully, though I have never been a witness to this so I can't vouch for it. Presumably any campaign can question tallies if their counts don't match the unofficial or official counts from the election boards. I welcome information on the final, official counting system.
So let's get back to the undercount in Harlem. The initial red flag of a district in Harlem having no votes for Obama in the unofficial tally by the poll worker is almost certainly due to this kind of mechanical problem leading someone to record what should have been 100 votes as zero votes. In theory it SHOULD be corrected when the official count is done. It is almost certain that Hillary Clinton had NOTHING TO DO WITH IT. People jumping to that conclusion are being way to quick to judge.
It also brings up questions about voting machines in general. Every system will have flaws. Every system will be open to some kind of fraud or another. The key is to have a system where fraud and errors CAN be detected and corrected.
Our lever machine in NYC, if the system is sufficiently transparent and monitored, and the machines were properly maintained, is a good one. The counts can be recorded by many people, both official poll workers and individual campaigns, newspapers, etc. before the machines are sealed. They are sealed in front of all these people then stored in a warehouse. Official counts are then made. Assuming these are properly done and monitored, and each campaign has a chance to check for discrepancies with the counts they made on election night, all will be well. The main problem is lack of maintenance.
PB/OS machines clearly have problems. Many of the problems in Ohio in 2004 were on PB/OS machines. But in the case of PB/OS machines there is still a paper ballot that can be recounted. There is a check on the official tallies if an election board handles the situation properly. The problem in Ohio was a corrupt election board dominated by one party, the Republicans. This meant that even court orders to check the counts were ignored by an election board that was partisan. New York has bipartisan election boards, so in theory we can avoid the fraud that probably took place in Ohio in 2004. Partisan election boards are also a problem in Florida. In 2004 Jimmy Carter's organization that monitors elections around the world and is highly respected around the world refused to monitor Florida's elections because they did not meet the minimum requirement for a fair election. The reason they did not qualify as a fair election was because they have a partisan election board. For PB/OS machines to be acceptable you need bipartisan (or maybe even multipartisan?) election boards and transparency. Maintenance will also, of course, be an issue.
Then we come to the DRE machines. These are the most flawed and unacceptable. Even with a good election board, you cannot have a fair election because there a.) is no paper trail, and b.) there is no transparency. With DRE machines it is all recorded digitally...even the tiny paper "receipts" that some machines give the voter are NOT an officially recognized paper trail. This means THERE CAN BE NO RECOUNT. There is no independent count for the vote that can be checked and compared like the lever machines in NYC or PB/OS machines have. Additionally there is no transparency. The software for doing the counts is owned by a company and that company keeps it secret. So no one knows what the machine does. If fraud were committed either by the company on behalf of a candidate they like or by a hacker, it could never be caught by the public. No paper trail...no recount...no independent check and balance...no transparency. These problems have led to the National Institute of Standards and Technology (NIST) to recommend decertification of DRE machines. NIST is a non-regulatory federal agency within the U.S. Commerce Department's Technology Administration that "promotes U.S. innovation and industrial competitiveness by advancing measurement science, standards, and technology in ways that enhance economic security and improve our quality of life." The decision by NIST to recommend decertification should be definitive. When the techies tell you the technology is unacceptable, we should listen.
Every system will have flaws. But proper maintenance of machines, fair and bi- or multi-partisan election boards, a paper trail, good opportunities for recounts if desired, and full transparency are the way to avoid problems with elections. This is really very simple, yet it isn't being done. DRE machines are the worst possible solution because they lack a paper trail, good opportunities for recounts if desired, and full transparency. PB/OS and mechanical machines have problems, but at least they allow for all of the above criteria for a fair election.
So let's stop blaming Hillary Clinton for an Obama undercount in Harlem and instead take this chance to recognize what we really do need for fair elections.
posted by mole333 | 3:40 PM
|
1 comments
1 Comments:
Just like intentionally faulty paper discredited the punch card, in favor of electronics, which were to be the future, we see discrediting of mechanical processes, like the levers, when in truth, they are 100% recanvassed and less vulnerable to theft and error.
What was missing in the story was the open secret that the worst machines are always used in minority neighborhoods. Just like Ohio and elsewhere.
To amplify about DREs, and direct electronic ballots, we have trails as a NY state law, but they are bad because voters don't check them and counties don't count them. Mostly because we'll never know if the secret vote inside the machine and trail are the same intent. There were paper trail failures in recent elections of 20-72%.
Post a Comment
<< Home