More controversy over voting machines, but this time it’s Diebold that’s complaining. According to a Globe article, Diebold is suing Galvin’s office for choosing AutoMARK machines over theirs.
The really ironic part is that Diebold “wants a review of the internal records showing how Galvin’s office came to select AutoMARK earlier this year.”
– Dan
Please share widely!
Chuck Hagel was the CEO for ES&S (who makes AutoMARK) I believe he’s still a part owner. ES&S counted his votes in Nebraska. Nice. I don’t see much of a difference between the two companies.
<
p>
BTW, wasn’t Galvin supposed to have this done by 2006?
Better late than never. Or something.
AutoMARK takes touch-screen input to produce a paper ballot that gets sent through the optical-scan reader just like any other paper ballot. This avoids the problems of other touch-screen technologies such as Diebold’s.
<
p>
Bill Galvin, for all his faults, really did make the right choice here. Progressives should support him on this.
This is solely a ballot marking system and it does not tabulate votes in anyway? If this is the case, and towns can tabulate results as they do now, then I agree that Galvin is correct to choose AutoMARK.
here’s the manufacturer’s web site:
<
p>
AutoMARK
<
p>
and ES&S’s:
<
p>
ES&S AutoMARK
You already gave the info….tx.
It is totally marking and the traditional counting is done on the town’s original system..Yes, Automark is best, if problems that were found are correct by election day.
but there’s still a difficult potential problem.
<
p>
If I’m using the touch screen and I create a scanned ballot, I still have to look at that ballot to make sure it was created correctly. Even if I’m blind, functionally illiterate, or not a native (or otherwise) English speaker.
<
p>
There are also potential problems of running out of paper* or ink, printers jamming or printing incomplete ballots, etc. Additionally, on-screen layout problems can result in undervotes as seen in FL 2006. It’s more complex than the system we have now, so there are new potential problems.
So, it ain’t perfect… but it’s soooooooooooooo much better than a touch screen that tabulates.
<
p>
* Do you ‘feed it’ a regular ballot, or does it have it’s own feed system?
So right on. The blind are the one’s who lose. if anything goes wrong with the system. Which is originally why this system was created. so let’s hope the SoS’s office gets it right by Election day.
The AutoMARK creates a paper ballot. Blind people aren’t going to be able to read a paper ballot, no matter what. What system would you suggest instead?
but so are the current disability-voters.
<
p>
1. Braile printers. Expensive, not all blind people read braile, but it would have some effect.
<
p>
2. Voice record votes. The “counters” would require multiple folks, just like when they were squabbling over voter intent in Florida, 2000. This has problems when names are similar, and has plenty of potential technical pitfalls, but it could work. It also has the potential to lose the “secret” part of the ballot — but to a lesser extent, so does the current system (since the ballot is printed, it’s clear it came from that voting booth. If there are 1 or few voters using that booth, it could be extrapolated…)
<
p>
I’m not suggesting that either of these could be implemented — there will always be edge cases and these edge cases will always be more expensive to deal with. It’s easier to point out a potential trouble than to solve that trouble…
As been said, better late than never.
<
p>
I was the first in the City of Worcester to try it on Election day.
<
p>
It has some tweaking to do. The ink wasn’t dark enough on the printout ballot, thus it kept saying I HADN’T voted. It took 20 minutes For Staff at the voting site to catch on. And I still had to vote on a conventional ballot! Luckily I am NOT blind, and could do that! Usually I’m out of there, in 2minutes. So MR. GALVIN has some work to do, AND SOON!
<
p>
Barb
If anything calls for open source, it is voting machines. How could any company could win a contract without letting a state see the code? It’s ridiculous.
<
p>
I have a question about voting. How important is a secret ballot these days? Seems to me it makes it unverifiable and that it would be better to see your vote recorded, to make sure it counted. But maybe there are ways to have a secret ballot and make it verifiable, perhaps with cryptography or intermediate codes that could be a verified by the public to be a voter, but not a particular voter, and verified to be a particular voter by that particular voter.
<
p>
Also, how important is having a single election day, rather than a period of a couple weeks when people could vote at city hall, as if by absentee ballot.
…’A Spectacle of Corruption’. It is about Georgian politics and voting practices. Some, like multi-day voting, were used then and the book gives you a pretty good idea why they were done away with.
<
p>
It’s funny, too.
PP is talking about “Georgian,” as in “in Britain under King George.” As in, really old. Amazon calls the book an “exuberant novel of 18th-century political chicanery.”
<
p>
The 21st century American state of Georgia actually does have early voting. Ga. Code. s. 21-2-380(b):
<
p>
Thanks Peter and David,
<
p>
I actually thought Georgian refered to that place in Back in the USSR, so thanks for explaining that. It sounds like an interesting book but I agree that’s “really old” and might not necessarily apply these days. But if it illustrates why the secret ballot is so important maybe I should check it out, because I don’t think most people are afraid to say who they voted for, indeed many would like to be able to prove it. But I know that it’s considered sacrosanct for democracy, especially violent fledgling ones. But in those places, people are afraid to put their candidate’s bumper sticker on their cars too.
<
p>
I have another question: how do we ensure “one man one vote”? How do we make sure people aren’t impersonating voters that are unlikely to vote? How do we verify that everyone on the voter rolls is a real person and not a placeholder for someone who impersonates that person, and 20 others? I’ve heard anecdotal stories about people voting in more than one city, or maybe it was just being on the voting rolls in more than one city at a time and being able to choose where to vote.
<
p>
Also, what do people that live in two places do? It seems to me they have a legimate right to vote on local issues and candidates in both places, but should only be able to vote for state and federal races once.
<
p>
how do we ensure “one man one vote”?
<
p>
Voter fraud can win you a fine of up to $10,000 or five years in the can or both (MGL 56.8).
<
p>
How do we make sure people aren’t impersonating voters that are unlikely to vote?
<
p>
In my town, there’s a voting booth for every 3500 citizens (roughly), and the same folks show up to run the polls. We’ve also got poll watchers. I’d be really hard to get away with voting for a person who isn’t you based on personal recognition — and if you tried it twice in the same booth, you’d get nabbed for sure.
<
p>
So — assuming the name you give didn’t already vote, is it worth that huge penalty to try to slip in one extra vote?
<
p>
How do we verify that everyone on the voter rolls is a real person and not a placeholder for someone who impersonates that person, and 20 others?
<
p>
When you register, there’s a field for “last 4 digits of drivers license or SS number”. If you provide that and it checks against state database, you’ve “proven” that you are you. If you are not, they ask for ID the first time you vote to make that confirmation. Again, is it worth that huge penalty to try to slip in one extra vote?
<
p>
I’ve heard anecdotal stories about people voting in more than one city, or maybe it was just being on the voting rolls in more than one city at a time and being able to choose where to vote.
<
p>
Sure. There are a few people registered twice in my neighborhood — somehow their name was spelled slightly differently and maybe they moved down the street. There’s also dead people on the rolls, and people who have since moved out of town. It turns out its easier to get on the registration list than to get off — which isn’t so bad, considering how removing people who still live in the neighborhood but haven’t voted in a while is a real democracy chiller.
<
p>
If you move from MA within MA, your old voter reg info should be removed by the state. If you move between states, it will likely hang around until they purge the names based on census data, etc, an artifact of voting being a states rights issue.
Ultimately, the penalty for voter fraud is worth far more than the value, and due to the number of local politicos hanging around the voting booth, your chances of getting caught are far higher than zero.
I see, that’s what I figured. Is there a way to tell how much fraud does go on? How often does someone get caught impersonating someone? How often do dead people vote? How often do people arrive at the polls to find their name already crossed off? Have there ever been cases of poll workers and watchers stuffing the ballot box with voters that haven’t voted?
My guess in recent MA times: zero.
<
p>
Why zero? Because if it wasn’t zero, I’d think there’d be lots of exposes and assorted different media angles, etc. But, that’s just a guess. Is there an easy way to query for convictions by statute number?
<
p>
To respond to “ever” — sure, both accusations, personal accounts, and convictions. But it seems that if you want to skew election results, its far more efficient to do it with other methods, like:
<
p>
“Democrats votes this Tuesday, Republicans this Thursday” – or –
“Elections delayed one week” in a particularly partisan neighborhood – or –
“If you have any outstanding warrants… even an outstanding parking ticket… and you vote, you will be arrested”
<
p>
Then there’s also tricks done by the government itself, like Florida purging the roles of any name which matched the name of a convicted felon without further consideration (like two people named John Lee Washington”, etc)… and nearly all of them happen to be black with few Hispanics. This happened recently in Florida, courtesy of Katherine Harris IIRC. Effective gamesmanship on voter reg issues seems to happen at higher levels than individuals casting a few fraudulent votes, and I suspect that’s because the penalty for casting a few fraudulent votes is so high when compared to the likelihood of swinging an election with those votes.
So as long as John Washington can go to another precinct and say he’s his cousin George who died 200 years ago, then it’s OK? 🙂
<
p>
But what you bring up is a good case for early voting, so that Mr. Washington could have time to resolve whatever trick was played on him. And heck, why not let people change their vote after an election for a week, too? With disconnected ballot id numbers that can only be traced back to the city precinct but not to the individual voter, except by the individual voter, people could not only verify that their ballot was counted, they could change it if they have voters remorse. Elections aren’t certified on election night anyhow.
can you get to in a day? There’s a natural limit to this kind of fraud — it takes too long and is too risky to log in sufficient votes to feel confident in swinging the election. I’m not arguing against safeguards — just that in terms of voting security risks, this one is way down on the list.
<
p>
<
p>
Nope. Not really. Since a vote isn’t attached to the voter (generally speaking), even if George walks up and finds that he (someone else) already voted, how can they take that earlier vote back if they don’t know which one it was?
<
p>
<
p>
Because that means your vote is attached to your name. That means that the government can look and see how you voted. That means you’re treading into totalitarianism very quickly.
<
p>
<
p>
So, you’re proposing that I cast my vote, it gets encrypted, and I get a key. I can change my vote later if I show up in person, enter my key, and it works. So, now there’s a whole slew of problems.
1. You must require it in person, other wise there’s no way to ensure that votes aren’t changed as the result of intimidation.
2. It must be electronic for this to work… but now, if I go back and change my vote, the space on the disk can be ferreted out with a diff. Now, the data on the drive can be mapped with me as a voter. Could the encryption be broken? Maybe. It’s a big risk, with what reward?
3. What if I lose my key? Remember, we’re talking about a 40 or 128 or 256 character string here, not a 4 number PIN or a password — those are too easy to break.
<
p>
What’s the value in allowing people to change their vote later? You’re opening up all sorts of security risks… are you getting much benefit for it?
…but I VIOLENTLY disagree with you about voting in two places.
<
p>
NOBODY has more than once legal residence, even though they may own or rent property in more than one town.
<
p>
We have MANY second homeonwers on Cape – why should somebody down for the weekend decide who our selectmen – that WE will have to live with 24/7/365 are? Does this mean I should be able to vote in Worcester elections, too, because I still own part of a house there?
<
p>
Uh-uh. No Way.
<
p>
My FAVORITE discussion along these lines was with a Gentleman from Lincoln, who was upset that he couldn’t vote against a local overrride for schools. I asked him – if he’s so concerned, why doesn’t he just change his legal residence to his Cape home? He replied that his accountant had told him that the tax consequences of doing so would be more than $500,000 – but he REALLY wanted to lower that tax rate on Cape, too! I wished him good day, as my sympathy had vanished like a dank puddle in the sun.
you’re talking about “early voting,” which is in effect in several states, and which I think is an excellent idea that should be adopted everywhere.
<
p>
On your secret ballot question, that’s an interesting idea — making ballots identifiable but encrypted, presumably to be examined in the event of an unusually close election or credible allegations of shenanigans … worthy of further consideration, it seems to me.
A Republican ousting an incumbent Democrat?
<
p>
I’m sorry, David, but a secret ballot is a sacred thing. It allows people to vote as they truly want, without the threat of future repurcussions. And as far as close elections go – Barnstable County has had three with a margin under 20 votes in the last three years (one leading to a Court case where the House seated the incumbent even after a judge had ordered a new election because of town clerk ‘ineptitude’). And I STILL want a truly secret ballot!
Hadn’t thought of that — but it’s pretty good! 😉
The problem is this:
<
p>
If you can verify your own vote, then you can be induced to “verify” your own vote — by someone bribing you, by your spouse, by your employer, etc. Besides, just because the screen says:
John Howard voted for: Clintobadwards
you don’t know if it was actually recorded, or instead tallied
John Howard voted for: Rudy McRomney
and, for that matter, now the government knows who you voted for — and that’s no good either. For that matter, what if it is “wrong?” Now what? How does anyone know you don’t just have voter’s remorse?
<
p>
There’s no need for encryption if the data management techniques are improved. Massachusetts is pretty good — there’s a paper hardcopy backup of your vote* that you created, not a machine. Counts of actual ballots cast is pretty straightforward.
<
p>
Sure, there’s other potential problems ranging from running out of blank ballots to losing bags of cast ballots. But encryption doesn’t really solve any problems, and creates many more.
As for multi-day voting, I’m iffy. On the one hand, increased turnout is good and we have limited multi-day via absentee now. On the other hand, a campaign has a clean ending: roughly 6:00pm the night before the election. After that, it’s all GOTV. With multi-day, there’s no “bang” ending… more of a trickle. Ultimately, voters who vote more than 24 hours early are effectively voting before all the information has been released, and it seems to me that this kind of scenario should be avoided where reasonably possible. Additionally, multi-day elections are much harder to secure, because there aren’t a number of differently-interested parties (plus police) hanging around the votes themselves for their entire lifecycle.
<
p> * I don’t know about disabled voting issues, since I don’t vote disabled.
(i) Regarding encrypted ballots, at some point there has to be some association between a decryption key and a voter–this to make sure that the voter doesn’t have two decryption keys–and so the encrypted ballot may actually be even less “secret” than normally cast ballots.
<
p>
(ii) I don’t know how the blind cast ballots now or before, but it seems to me that this issue regarding computer voting is a red herring. I actually don’t know how they get to the polling place. If they have a trusted companion (other than a guide dog, of course) who can read the ballot to them and fill it in according to the blind voter’s instructions, that’s fine. If they have a computerized voting machine, it would be a simple matter to provide a screen reader and headphones that people could use to describe what’s on the screen and help them cast a ballot.
<
p>
How did they vote before? Or did they?
seems to run without problems. We go down to the Town Clerk’s office, he checks us off, we vote in the conference room, the votes go into the vault until the election day. Several times this meant we could vote inspite of work, travel, family responsibilites. In Vermont we can vote as soon as the ballot is printed.
My parents vote ‘absentee’ – which is really the same thing – because they are quite frail and elderly. In their retirement community they get together with friends and discuss candidates and vote about a week ahead.
<
p>
secondly about secret ballots: I live in a small town. I know and work with the people who are running for office. When I vote, I do not want them to necessarily know that I am voting for or against them. And I also don’t want them to know if I split my ballot or vote a straight Democratic or Republican ticket. I don’t want to have to justify my vote to anyone but myself.
My mother’s vote cancelled my father’s vote for about 50 years. He never knew she disagreed with his vote, although her children did.
All well and good to slag the devils at Diebold, but aren’t they open;ly considering getting out the election business? If anything, this looks like the seting up of an excuse to quit the venture.
I’ve been involved in government contracting issues, and Diebold’s requests for information regarding the contractor selection is not out of line.
<
p>
I have other issues regarding Diebold’s refusal to release source code (number of issues, pro and con), but I do believe that chastising Diebold on this issue is out of line.