Another Stolen ElectionFollow

Varus reincarnate?

That's what I thought as well. I was waiting for him to quote of one us, to see if the brilliant puns would come out.

Wait and see...

I think there are a few bad apples at ACORN but at the same time, it's an organization that has done alot of good work for years with a large staff of volunteers who work with the community. I hate to see them trashed to gain political currency, because that's what it really is. Otherwise, the investigation would be about looking at their own system of accountability and not to slam Obama's possible involvement with this organization. I hate this quasi-McCarthyism where republicans are throwing daggers at these ancillary people, hoping that something will stick without regard to the consequences. I'm not all that caring about how Bill Ayers feels but otherwise, it is really problematic.

When your get out the vote efforts result in a roughly 50% fraudulent registration rate (much higher in some areas), it's not a "few bad apples", it's something systemic within the organization itself.

You would need a remarkable coincidence for the scanner to be misaligned in a way which still provided a valid vote for every person/issue on the ballot. If a choice anywhere on the ballot is invalid then the scanner will immediately alert the voter and give them the option to spoil the ballot and try again. Since the polling judge on scanner duty is the one to spoil the ballot, you'd catch on pretty quick that something was up.

No one ever mentioned this. So you made something up in your own mind and then called it perception and not reality...nice.



Don't you think that if they programmed the machine to switch the vote, that they would also program the printer to do the same thing???



WRONG!!!

WITHOUT SEEING THE COMPUTER CODE THERE IS NO WAY TO VERIFY THAT THE SELECTIONS THE MACHINE SHOWED YOU ARE THE VOTES IT ACTUALLY REGISTERS AND PRINTS ON THE RECEIPT

Cheers to the Adams County Clerk in Colorado for impounding a Diebold vote-flipping machine.

http://www.aurorasentinel.com/articles/2008/10/29/news/metro_aurora/doc4907ee63d0553766738959.txt

This one happens to be from Democratic to Republican, but we're still waiting for any clear patterns to emerge >.>



Edited, Oct 30th 2008 6:02pm by soulshaver

No. I'm attempting to put a bit of sanity into the issue. It's utterly pointless to rail about the flaws in the new system if you can't show how they are worse than what it replaces.

The fact that you didn't mention it doesn't change the fact that this is a valid comparison to make. When you talk about how an electronic system can be rigged (with the presumed argument that we shouldn't use them), you are by default making an argument that the alternative is less easy to rig. You may not actually say that, and you may not even be bright enough to realize it, but it is exactly what you are advocating.



And the machine that reads the paper ballots can be programmed to do that to.

We're comparing one method to another, right? Keep that in mind.


Also. Absolutely nothing in the tests linked to earlier in this thread showed anything of this sort happening. While what you say is possible, it's no more possible with the touch screens or paper scanning systems. In both cases though, the physical paper will not match the electronic results (if it electronically records one thing, while printing a report showing something else). That's trivially easy to spot.

Again. I'm much more concerned with methods that can't be easily caught with a simple sampling check of the paper versus electronic results. And those are much more likely to occur if you *don't* have the electronic system at all. With both in place, you can check one against the other. If you only have one, then someone only needs to falsify one thing.





Writing that in caps doesn't make you any less incorrect.


Let me repeat this. And I'll make it really easy for you:

Let's imagine there's an election with just one thing. A choice between A or B. Let's imagine that I've got a touch screen system. There are a variety of ways that this could go wrong:

1. I touch A, but B is checked on the screen.

I can see this and make sure the right choice registers before I submit it.

2. I touch A, the screen shows A, but the system records B instead.

I have a printout that's based on what the computer records. I can see that the printout doesn't match what I selected and can bring this to the attention of an election worker.

3. Someone's been really sneaky. I touch A, the screen shows A, the system records B, but still prints A out just to fake me out...

The paper result will not match the electronic one. So when an election worker counts a random sample of votes (which they do) and realizes that the ratio isn't correct, they very quickly catch this.


Let's see how this compares with an on-site optical scanner system (like they use up where Joph lives):

1. I mark A, but the scanner records B.

If I check the receipt (and that's part of the process, which may not be the case), I may catch this. If not, the paper record will not match the electronic one, so if this is by design, this should show up just as in option 3 above. In this case we've got the original paper ballot saying one thing, but the receipt and electronic total saying another, so it's pretty easy to see where the flaw or tampering occurred.

2. I mark A, the scanner records B, but the printed receipt (if there is one) still shows A (a fake out).

This is also the same as number 3 from above. The paper total will not match the electronic one and it'll most likely be caught. Again, since there's both the paper record of my ballot and perhaps a receipt, it's still relatively easy to spot where the problem occurred.


Note that this only has three cases, since you don't have a chance for spotting it directly on the screen as with the touch screen. I suppose in the case of the scanner actually just being broken, it may raise a similar alarm right there (whether there's a receipt process or not). But while that protects you from scanner miss-alignment, it wont help if it's actually programmed to falsify the vote.


Let's compare this to a traditional paper balloting system where you punch or mark your selection and drop it in a box. The ballot is then taken to a central location where it's read through some sort of scanning system.

1. You punched A. Your ballot arrives at the scanner, but the scanner is rigged to record B.

Same as above. This will show up as a discrepancy between the tallied result and hand checked sample tallies. So no better, but no worse than above.


2. You punched A. In between the polling place and the counting place, your ballot box is replaced with one with a different set of ballots in them with other votes. Those are scanned into the system and stored in the official record.

You're screwed. There's no paper record kept at the precinct to refute what is now stored at the central location. There's no electronic trail to verify either. While this isn't easy, if someone can do it, they can do it without any way for anyone to know it's been done. Period. End of story.


The point is that electronic system provide an additional trail of data. We can check what the voting machines at each poling site recorded. We can compare that to the paper receipts at those same sites. We can compare those to the electronic results after they get transmitted to a central counting location. If there are discrepancies, we can backtrack them to their source. We can know which precinct's results didn't get counted correctly. We have the paper ballots themselves and the electronic results *and* we also have paper and electronic trails that are recorded and kept on each machine at each polling location.


There are many more checks when using an electronic system as compared with not. Yes. There is the potential for failures and flaws. But IMO the likelihood of actual large scale tampering is much much lower when using an electronic system than when not. When we were talking about early variants which didn't print out paper receipts, you'd have a point that we were replacing one set of flaws with another set of equally flawed replacements. But those were replaced very quickly. Now, we have two separate paper trails and in most cases, two separate electronic trails as well. It's a much much better way of doing things.

The is a logical fallacy called a false dilemma. You are presenting two choices and saying that I have to pick one. You are wrong. I am not making any argument by default. I am making the argument that we need a NEW system, different from the old and different from the new, one with TRANSPARENCY and ACCOUNTABILITY.



You completely missed the point. We need a NEW METHOD, one with TRANSPARENCY and ACCOUNTABILITY, read the posts above and some examples are provided.



This is where you're argument completely falls apart. YOU DON'T GET TO SEE THE PRINTOUT!! Do you think that people can actually check the printout of the machine as it is processing votes??? Hilarious. It is locked up inside the machine. Now you are just making things up.



If you want an expert opinion, older machines with hardware ROM based encoding like ScanTrons are much more reliable and less capable of being tampered with because with an sort of kernel OS where there is user input and multiple software layers there could be all sorts of hidden code.

I am advocating paper ballots that are easily read and marked read by verified tabulating machines in addition to being counted by hand which will provide data for an online database where each citizen can confidentially verify his or her votes so we will have 3 different measurements of the vote (machine count, hand count, online accessible database) and the citizens can be satisfied that their votes are recorded and counted correctly.

This is completely absurd.

Yes. And you test a touch screen at the beginning of the day as well.

Like I said. It's exactly the same. Except that if somehow the mechanism does get out of alignment during the day, you'll see it right on the screen when using a touch screen, but will only notice it on the scanner if you look at the output and it doesn't match.

My handler example from earlier is relevant here. We'd been using this thing to debug a software problem for a few hours. Then, completely out of the blue, it decided to offset itself 3 inches to the right. Inside the device, it thought that this was the physical location correlating to where the part it was to pick up was. Now in this particular case, we were looking at it through a transparent screen and could see that it was trying to pick up a part from the wrong location. Imagine if the only way we could tell something was wrong was to read off the results of the test it was supposed to run? We might know something was wrong, but it wouldn't be immediately obvious what.

Same deal here. A touch screen provides you immediate feedback on your selection. You touch a part of the screen, and your choice appears checked on the same screen. If the wrong choice appears, you can immediately try again. If it persists, you can call someone over to help you. If the optical scanner fails in a way it can detect, it just tells you that something is wrong. If it fails in a way that results in the wrong choices being recorded, you may or may not know depending on whether you double check the output.

In all cases, a touch screen is better as an input device. Again. For exactly the same reason why we type stuff on a computer screen that shows us that what we typed is what we wanted to type rather than punching a card, feeding it into the computer and then hoping that everything worked out.


Optical scanners really are the electronic voting equivalent of the computer punch card. It's a nice in-between step, but there are much much better ways of inputing data and there's no real reason not to use them.

Edited, Oct 30th 2008 5:03pm by gbaji

No. You're not. You're arguing that electronic voting systems using touch screens are prone to error and manipulation and might lead to a "stolen election" (you wrote the title, right?).

At no point in your OP did you recommend that we needed a new and better way of voting. You simply argued that touch screen electronic voting was "bad". It's not a false dilemma in this case to assume that this is an implied argument for the alternative, and it's certainly not invalid for me to argue that the alternative is *worse* than what you're complaining about.



Sorry. I was responding to your original point and the blog you linked to. It spent all it's time showing the flaws and possible vulnerabilities of electronic voting systems, but failed to compare those to traditional paper ballot systems. So excuse me if I thought it relevant to point out that the electronic systems, for all the possible flaws, are still a step in the right direction.

I'm also not sure what kind of "transparency" you think is needed.



This is where you're argument completely falls apart. YOU DON'T GET TO SEE THE PRINTOUT!! Do you think that people can actually check the printout of the machine as it is processing votes??? Hilarious. It is locked up inside the machine. Now you are just making things up.[/quote]

Sigh. Some older systems do not generate a paper trail. Some only generate a printout at the end of the day, or when the storage media that set of tallies goes with is removed or replaced (or the system is reset for some other reason). The newer ones all generate a printout at the time of voting which the voter can check right there (essentially a voting receipt). These changes have occurred exactly because a lack of paper trail has been recognized as a potential problem with electronic voting.

Had you gotten your information from a more reliable tech source, you might have stumbled on the reality that much of the hype and scare of this story is overblown. Two specific bits:



I'm not just making stuff up. Perhaps if you actually took the time to learn about the subject instead of blindly believing some crazed blogger? Just a thought...




What's funny is that the same guy even said that people make mistakes with these things more often than the machines themselves do. And he kinda proved it when he thought it was some kind of miscalibration that it didn't change his vote when he clicked the "all republican" button. Turns out that's the correct behavior of the system...

Also. As always, you can see the immediate results of a change *and* you see a review screen *and* in newer systems you see a paper printout of your selections.

What more do you want?




I'll take the word "expert" with a grain of salt if you don't mind. ;)

They are indeed more reliable and harder to tamper with. They're also harder to program in the first place, and don't provide any protection at all against the most easy form of election fraud: stuffing a ballot box.

The whole point of electronic voting systems is to have the voter himself input the data into an electronic media. We've used various paper scanning devices for decades. But due to the simplistic nature of such things, they contain no ability to check against fraud.



Huge gaping flaw there. Voting is anonymous. So there's no way for a citizen to verify that his particular vote was counted properly. The only way to do it without removing the anonymous requirement is to check off each person's name as they go in to vote (which means you have to have accurate registration rolls). You then need a system that is present right there in the voting place to allow a voter to ensure that his vote was counted. Once his ballot is mixed in with everyone else's, it's "lost". So verification needs to be done right there.

Electronic voting systems are basically just an extension of the counting machines used in the past. The difference is that instead of putting your vote on a piece of paper, dropping it into a box, then having someone else later take it out of the box (along with all the other votes inside) and feeding it through some sort of counting machine, you are inputting your vote directly into the counting device itself.

IMO, that's a huge step in the right direction, and for all the complaints and suspicions surrounding them, they don't actually add any new ways for someone to steal an election. Voting machines could already be rigged or programmed to benefit one side. False ballots could already be run through said machines. But it's harder to do that out in the open in front of the public than it is to do that in a back room somewhere.

That's the whole point. You demand transparency, and while I'll admit that the current generation of electronic voting systems are not perfect, they are farther along in that direction than anything we've ever used before. There's an expression in the tech world: If you wait for the perfect solution to a problem, nothing ever gets better. Basically, no piece of new tech starts out "perfect", and it's wrong to expect it to be. It just needs to be better than what it's replacing. A step in the right direction is better than refusing to take that step because it's still not exactly where you want to be.


Or. As one of my early mentors used to say: "Publish or Perish". He was talking about writing programs and/or pushing out changes to a computer environment, so it's a slightly different meaning than the phrase usually has. He was basically saying that at some point you have to decide that what you're working on is good enough to put out there in the world. And once it is, you should do so without hesitating. Progress occurs faster over time by putting your first attempt out there, getting feedback, and then making changes, than by attempting to guess what the best product would look like and making that. You'll find that things you didn't think were important are, and things you thought were important really aren't. And along the way you may think of new things you'd never have thought of as long as the product was just sitting in the development stage.


Same thing with electronic voting. It was far from perfect when they first started using them. People complained. That caused changes. People pointed out yet other problems, and that prompted more changes (paper trails are a great example of this). The point is that if we sit there in the dark trying to invent the perfect voting system, it'll take us longer to get there than if we put out the best thing we can today and then keep making improvements along the way. I just think it's unfair to insist that we do it the other way around. Trust me. It doesn't work. You'll spend forever working on it, and then unveil your "perfect" system, only to have it turn out it's got just as many flaws as any other first release product.



Yup.

Oooh! What do I get? The toaster?... ;)




No. I'm describing a couple of different types of errors that can occur. There are a number of ways that a scanner or a touch screen could improperly read input. We can go back and forth listing them all day long, but ultimately the core difference is that you can see immediately on the screen when the touch screen registers the wrong thing, whereas you have to rely on the very system you're using to scan your ballot to tell you if the input it read is wrong in the case of a scanner.


Like I said. It's the difference between inputing with a punch card, and inputting via a keyboard and wysiwyg interface. Touch screens offer the same advantages.

Edited, Oct 30th 2008 6:13pm by gbaji