Posted on 10/17/2008 8:28:38 PM PDT by RetSignman
I don't know how electronic machines are handled and the chain of command so it seems to me that there could be a POSSIBILITY of tampering way before actual voting begins.
I don't think there is any doubt now that this election is facing massive fraud but it has been focused on fraudulent ballots but what about the electronic voting machines?
Is it possible to pre-load these electronic machines with votes PRIOR to the first votes are cast by voters? I would think that IF this possible, how hard would it be for corrupt officials in key districts and key states to accomplish this without any oversight by the authorities?
All it would take are corrupt officials and we all know there are plenty of them in EVERY state.
OK, you are free to flame away and call me a conspiratorial nut job but in this election being the last best chance for the Socialists and ultra left wing in this country to finally take control, is anything they won't do?
You are powerless to insure a fair election.
Voters are already disenfranchised thanks to Ohio ACORN fraud, with the SCOTUS dismissal of the 14th amendment and its enforcement.
ACORN tried this crap in Ohio in 2004, but couldn’t deliver Ohio to Kerry—too many voters came out for W.
But hey, there’s always the WA governor to thank and JENNIFER BRUNNER, the current Ohio SoS. Never mind that she’s f’ked us all, each and every one.
Purple fingers and Xs on ballots. Drivers’ license or, even better, birth certificate as ID. There, a great solution.
In districts that lean Republican pre-loading has never been a problem.
It’s possible for the counters to be non-zero before an election. In fact tests are run on the machines, and the results are then supposed to be cleared, but who knows. The best guarantee against that happening is for each party to have election watchers at the polls. They can verify that each machine starts with zero.
Electronic voting has always made me very nervous. I don’t like it. Too easy to manipulate.
Thank you, I was looking for someone who had some knowledge of these machines, you were the first one with at least some modicum of knowledge about them.
There is a long history of both US and local prosecutors ignoring vote fraud. This is been taken by many as a green light. It’s like an athlete who sees that the ref isn’t calling a certain type of violation and then engages in that violation knowing the ref won’t call it.
The mechanical vote machines, the paper ballots and the computerized machines all have the same flaw. They depend on honest people. As long as there are dishonest people who think they can get away with it, technology is not going to protect us.
As an official Poll Judge, I will set up and/or verify setup of all voting machines at my precinct. Part of the process is to open a sealed compartment and insert a memory card that is set up for the specific precinct, that I receive in a sealed/locked bag from HQ. During setup, the machine prints a tape showing zero votes for each office, that is examined, signed by each of the 3 Poll Judges, but left in the machine, still attached to the printer.
When the polls close, we will insert a “supervisor” card that prevents any more votes from being recorded, and causes the vote totals to be printed on the tape.
The memory cards will be removed from each machine except the one chosen and set up to accumulate totals. Then each of the memory cards from the other machines will be plugged into this machine until all have been accumulated. Each memory card is numbered, and the machine will not accept any more that once. And the cards numbers must (of course) match the official list.
Could the system be manipulated somehow? Yes, but it would require getting into the programming and making some pretty sophisticated changes, without detection or “fingerprints”, either physical or electronic. The machines are tested, then locked and sealed with the primary memory cards removed - and these memory cards are the official record of the machine vote. They will be returned to HQ by a team of one R and one D official for counting there.
When I was growing up in New York, my parents were regularly poll watchers for the Conservative Party. In every gubernatorial election, the machines would come in with 3-5 votes for Rockefeller on each one.
As you indicate, and as with all voting systems, the security is in the way the system used more than it is in the system itself. Erasers could easily defeat a hand-written ballot except for the fact that it’s deposited in a sealed container, and there are multiple witnesses each time it’s handled. The same situation was true of punch cards, cards themselves could be faked, the system was designed to keep you from doing that.
Sounds like your Registrar has some well thought out rules in place.
Oh... and... another obvious problem with pre-loading would be that the total vote count for the precinct would not equal the total number of voters who signed in for that precinct. Pretty obvious fraud indicator.
By Sarah Palin, in person.
Or Robert Bork.
Or Jim Robinson.
Cheers!
This type of fraud is difficult. It would require a teenage hacker, at the least.
I have been a precinct election judge in West Texas for 14 years now. I have been involved in two recounts. The first was all Opscan paper ballots, the second a mix, with voters allowed to chose their preference.
The first was tedious, with watchers watching counters, and the candidates allowed to roam (silently) observing it all. Took up most of a Saturday, but no problem. The second was a rolling disaster that made national headlines as we took two weeks to come up with a result that still didn’t quite match the numer of recorded voters. We had to wait for a vendor rep. to fly in to tease some of the E-Slates into coughing up results, and then all we got was a reprint of what he had election night, in fact, had we gotten anything else, there would have been something wrong. As it was, something obviously didn’t quite add up, but a roomfull of fairly bright people never did untangle it and the trailing candidate finally conceded, I suspect out of sheer exhaustion.
I heard Con. Rush Holt (D) of New Jersey Friday on NPR’s Talk of the Nation, his bill last year to demand more rigorous testing of E-vote systems was amended into uselessness then flushed, Friday Holt specifically called for all-paper elections.
Opscan ballots, scanned at precinct level, then scanner chip AND ballots are securely returned to Elections, are proven to have the lowest error rate of any system out there. Why go A** around the elbow to get a machine to give you a VVPT, when the Opscan ballot is its own paper trail?
We had a case here in South Texas, Hidalgo county, Republicans often don’t even bother to run there, had a Dem. overwhelming favorite against a third party (Constitution Party I think). Midway through early returns, it was noticed the Dem was losing badly. Administrator checked, sure as God made little green footballs, the Sequoia machines had been set up wrong and were flipping votes.
Everybody up to Sec of State crowed about how catching this proved the validity of E-voting! EXCUSE ME? If that race had been, say 55-45, there might have been mild surprise, but nobody would have checked and Hidalgo county would have the wrong man in the legislature.
I don’t trust these damned things half as far as I can throw one, and I’ve lugged around enough of them, they’re heavy.
That is scary. The question I would have is, can the card be scanned for malicious embedded code PRIOR to inserting it in the machine?
[these memory cards are the official record of the machine vote]
After watching the video post #18 which showed how to hack a voting machine, I can see that the final step of inserting the official card can be manipulated. It also showed how this card can be spread to other voting machines.
It showed me that even making sure that the machine has been set to zero isn’t enough of a security measure.
Shouldn’t that ‘verified card’ undergo one final fail safe method of scanning that card for embedded malicious code? That card’s integrity is, as I see it, the critical arbitrator in a honest election. This malicious code wouldn’t be evident to the judges and the assumption would be that all measures have been exhausted to ensure an honest vote.
Sorry for belaboring this concern.
I am at least as worried by the absentee ballots that ACORN will have so much access to, as I am by the voting machines.
Why would “Mickey Mouse” or “Tony Romo” risk being caught voting at the polls when they can just vote absentee?
All electronic voting machines should be connected by both land line and cell phone to a central server that records votes and the time when votes were cast to see if any voting machine reflects votes before the polls open, after they close, or an impossibly fast cluster of votes.
To the best of my knowledge, the memory cards contain only data, not executable code. That barrier is not absolute, but it can be controlled by accessing the memory card as an I/O device instead of being mapped to a memory address. It would be possible to write code onto the card, but the resident computer program would have to load that code into memory for execution, and then set its instruction address to where the code was loaded.
Also, to avoid detection, the malicious code would need to remain inactive during testing AND while totals are still low, and then only move votes from one candidate (or non-votes) to another instead of just adding or discarding votes.
Techniques to prevent these tricks exist, but the battle is ongoing. For instance, all program modules and files should have digital signatures, which can be checked against a read-only list of such signatures as it loads, as well as by periodic memory scans. Additional data could be collected internally, such as the total number of times each button is activated on and off, which could be used to cross-check the vote totals, for instance.
It would seem that the most obvious solution would be to issue a paper receipt to the voter showing him his votes, but that invites intimidation and vote-selling.
I remain convinced that the best automated system was the punch cards. There was a physical record of the voter's choices, and by maintaining a secure chain of custody fraud was impossible. California had used the cards for decades without problems, and they even stood up to the trial run Gore's consultant tried in the mid 90’s. The big difference between California and Florida was that the California legislature had an established ‘hanging chad’ standard.
What is needed is an unambiguous, unalterable, machine readable physical ballot. Recalling Florida 2K, hand-punched cards do not suffice. As I see it, the voting machine should produce a ballot that is both machine readable AND readable by - but not accessible to - the voter, BEFORE it goes into either the voting ballot box or spoiled ballot box. The voter should have 3 chances to discard a spoiled ballot (same as spoiled paper ballots are now.) The machine could issue a voting receipt, but NOT with any record of the actual votes cast.
The machine could keep an unofficial vote count, but any recount or audit would be based on the collected paper ballots.
I understand your point but, just as a matter of information, punch cards ARE unambiguous if your state properly defines what a vote is. The State of California provided legislatively that a vote consists of a chad with two or more corners detached. Florida had no such legislation, resulting in people attempting to define intent from smudges and dents in the cards. When Gore’s consultant tried to fiddle with the cards in California it was spotted immediately, because we noticed that they were handling some cards very delicately and flexing others. The ‘counters’ quickly became ‘observers’, and ROV staff replaced them.
So, I’m trying to picture how your proposed machine would work after a vote had been cast. If the voter isn’t allowed access to the ballot, then I guess there’d have to be some automated system whereby the he/she read the finished ballot, then pressed a button that either discarded it or moved it physically to the ballot box, or at minimum sealed it in an envelope and gave it to the voter to be deposited...?
In Alabama, we use electronic optical scanning machines.
We get a ballot, use a marker to fill in the -— > blank space, connecting the arrow.
So, you have a paper ballot after voting. If a machine dies, you have the paper. If a machine give erroneous readings, you can rescan all of the ballots from that machine.
IF needed, every precinct could rescan every ballot on every machine in just an hour or two.
I would love to see the Optical scanning machines made mandatory for every district. As close to foolproof as you can get.
Right the first time - after the voter checks the ballot, he would select either “VOTE” or “DISCARD” to send the ballot to one sealed box or the other. He gets to discard his first and second ballot and receive another if it is incorrect. But if he spoils the third ballot, he can either vote or discard it. If discarded, I suppose that he could vote a paper ballot - but I think only ONE try!
You could ask the machine's manufacturer, but would you trust the response? I am sure that Diebold's sales reps told their customers that the voting machine described in the video was perfectly secure.
Even if your verification software exists, how could you be sure that it had not itself been compromised?
High tech computerized voting machines just invite higher tech hacks. They should be viewed as a threat to democracy.
It is nearly impossible to hack a paper ballot that has been punched or indelibly marked. Paper ballots can be replaced with fraudulent ballots, and poll workers can intentionally miscount votes, but such actions usually entail substantial risk of discovery and prosecution.
There is nothing inherently wrong with using machines to verify ballot counts, but those machines should be as low tech as possible.
And you’d have to set it up sort of like those old casino keno machines, so that the path to the sealed box was behind something transparent....
...and the chain of custody throughout the entire process as secure as possible. Fraud prevention is in the system, not in the voting method you use. Any voting method can be compromised if a third party can get their hands on the ballots regardless of what form those ballots are in. But, yeah, I agree. I'm a big fan of the 'simplest tool' theory, i.e. the simplest tool that does the job is the best tool.
After 31 years with IBM in the field, plus 10 years consulting, I have a VERY detailed knowledge of punched cards, including the hand-punched versions. There are two problems - undervote and overvote, and I am certain that Florida 2K suffered from both. A number of ballots were disqualified due to overvotes for Buchanan - which would nullify votes for either Bush or Gore. And others were disqualified due to undervotes - partially punched chads, that might or might not still be blocking the holes. It is even possible that a fully detached chad could be placed into a punched-out hole and smoothed over enough to stay there.
So counting detached corners might not quite be up to the task. However, flat, uniform cards are a lot easier to handle in machines than flimsy, curled-up adding machine tape, so they might be a better medium, especially with serial numbers and security features to make counterfeiting difficult. Also, OCR has improved tremendously since I worked on some of the earliest versions back in the ‘60s, and particularly reading machine-printed letters. Printing the offices and proposals, plus the names or vote selected, would be quick and easy, and the machine printed card forms could be tabulated quickly and easily.
If you’re going to address undervote and overvote with your proposed system, then you’ll need to do some checking prior to permitting the voter to hit the ‘save’ button. You then get into an interesting balancing act because a certain percentage of undervotes is intentional or just because the voter doesn’t figure he knows enough about the candidates to make a decision. Even if all you do is ask the voter if he really wants to leave the race unmarked, you’ll be intimidating some people into voting on races they may not feel qualified to vote on. I don’t recall seeing any studies quantifying the percentage of undervotes that were intentional, but it’d be an interesting subject for somebody’s thesis.
There’s a bit of a philosophical question that arises in both circumstances. It’s a little less clearly drawn with undervotes where you’re asking whether you really want to force or encourage people to vote on issues they don’t feel they know enough about. It’s a little more clearly drawn in the case of overvotes, where you have to wonder whether you want your leaders chosen by people who can’t follow a simple direction to choose one and only one candidate. And, of course, you don’t have to get there with punch cards, where you assume that what the voter did is what the voter intended to do. The bottom line is voting is not a binary thing, there are more than two conditions that could accurately reflect the intent or the competence of the voter. Hand ballots and OCR ballots will, of course, also have over and undervotes.
The Buchanan issue was primarily a mis-design problem, and error-checking prior to acceptance in the system you describe would catch that. However, there’s a possibility that the individual coding in the error correction routine could get the conditions wrong and ‘send back’ a legitimate vote, sort of a false positive situation. That, in turn would result in voters ‘fixing’ their ‘error’, producing over or under votes on ballots that they originally marked correctly.
I take your point on the fully detached, smoothed over chad, although my experience was that the counting and handling process was pretty robust, and it was more common to have partially detached chads get knocked loose in the counting process.
The voting machines used here already show a ballot summary with undervotes highlighted in red, and direct the voter to select either “go back” or “cast ballot” from that point. Overvotes are not possible.
It cannot be proved, but it is possible that some of the Buchanan overvotes were caused by deliberate manipulation of the ballot cards - poking an object through a stack of them. That is why I favor machine printing rather than punching the ballots. No chads, no weak spots, and NO overvotes.
In my scheme, the voter would see a summary display with undervotes highlighted, and be offered an opportunity to return to make additions or corrections, or to print the ballot. Then he would be able to examine the printed ballot - under glass - and choose to either deposit it in the “VOTE” box or the “DISCARDED BALLOT” box.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.