Interesting.

more dirt please? this is kind of bland and could be more descriptive on how other members of the voting team felt about this process outside of your precinct. also how about the lady that drove you home. how did she feel about the process and did anyone get dispatched in other precinct's? good article though, but you could get in a little more depth.

I wonder how often the SanDisk 128MB PCMCIA flash cards fail? Couldn't someone who wanted to wipe out one of the ballot memory cards slip it into a Pocket PC and format it?

Yeah, I agree with the above poster. Why leave us hanging at the end? You tell the story well, but some more analysis and conclusions would be welcome.

I have a question. You state in your review that "I think that current electronic voting systems are better than the traditional systems in terms of security, and also in terms of usability for most people" and then go on to present a number of examples that I would say show them to be less secure and less usable than a paper ballot. None of the problems you describe seem at all likely with a paper ballot, and some - such as the case of trying cheat by claiming they got the wrong card - are just not possible with a paper ballot.

Could you please post a follow-up to your review, covering the areas in which you believe the Diebold machines offer better security and usability than a paper ballot? So far, you've made a pretty good case for the other side, while giving an unsupported claim in favor of the Diebold gear. Maybe the stuff is actually pretty good and all of us who don't trust it at all are wrong, but you haven't made your case.

I moved from San Diego to the LA area last fall and am delighted that here in Torrance we still have paper ballots, with ink stampers rather than punches. They solve any possible problem with hanging chads, and provide the important paper trail that Diebold's machines lack. In my opinion, paper ballots do it better, faster, cheaper, and more securely than electronic voting.

I am a Canadian, so forgive my lack of understanding of just how your voting system works.

But it sounds like people either register, or tell you which basic group they would like to vote for (independent, democratic, republican), then you give them a balet that will present the people running under that group, they then pick one to cast their vote. Is this correct?

Here in Canada, we tell the people running the poll nothing, then are presented with a page of variable size (depending on the riding and the type of election) and choose whomever we wish from any political party. It just seems odd to me that you would be restricted to the group you told a person working at the poll (ruins the anonimity it seems)...

I'm a voter from San Diego. I blogged about my experiences recently in Oceanside. Sorry,... too late in the evening to recount. Read the whole soted mess here:

http://www.spencerlindsay.com/archives/000114.html

As dj mentions, how hard would it be for a voter to use a VAC and then format it? (claiming it didnt work) Or even copy it, provided they bring their own SanDisks?

Although you did say VACs had to be reprogrammed before using again, so perhaps the above attemps would fail? Good news is, formatting would only work once, maybe twice, before those in charge catch on or raise eyebrows. (hopefully)

And please elaborate a bit more on JB's response. What does the new system excel in?

But it sounds like people either register, or tell you which basic group they would like to vote for (independent, democratic, republican), then you give them a balet that will present the people running under that group, they then pick one to cast their vote. Is this correct?

This is correct, but perhaps you have missed the fact that this was a primary election. That means that the major task of the voter that day is to help his party choose a candidate or candidates who will run later in the general election. In most US states you must be registered as a Democrat to vote in the Democratic primary; apparently this is not so in California, and if you are non-affiliated you are allowed to vote in the primary. But like in my state, you cannot vote for both a Democratic and a Republican candidate.

Think of it as two private elections combined into one pseudo-public election.

Thanks for the comments Ross

Hmm, I think I am starting to get it now, this isn't actually a political election (those who win will not be granted office), more of a party election to determine who gets which seat within each party.

Here, joining or registering a political party costs money, and all the parties run their elections seperatly. The Conservative party ( http://www.canadianalliance.ca/english/index.asp ) is having an election in the next few days. I think here it is actually possible to join multiple parties and vote for the leadership in each.

I am a two-bit sales guy and I could violate this voting system with the equipment I currently own and little bit of planning and deception. Basically we are just trusting that no one will want to do such a thing?

I live in Minnesota where we use a paper ballot and you use a felt-tip pen or sharpie to complete an arrow that points to your candidate of choice. This system works perfectly, is cheap and easy to setup and administer and is definitely impossible to mess with.

I agree with JB, if you think this Diebold (and die it will, bold or otherwise) system is so great and yet end your story with an allusion to the infantile skill required to violate the system what are we the reader to think?

Also not a US citizen (Australia), so for any actual election you really have to do multiple votes? What happens if after the primaries you decide the party originally selected is full of idiots, can you change? I think our system is alot like the Canadian one, at the end of they day, there is a record of my vote, and a record of the fact I voted, but no record/link betweeen them, I really do not like the idea of telling anyone what party I voted for/support, its not their business, no matter who they are or what trust is placed in them

In Canada we simply have a pre-compiled voter list before each election. Some elections officers go door to door and register everyone of legal age for the election. When you go to vote, you show your ID, they match it up and then you get to vote. Even if you weren't on the list you can bring ID and proof of your address and you can vote instantly. We mark an X or a check next to the name we want and the votes are counted. Other than theft, or natural disaster there are very few ways to cheat the system. There's many checks in terms of counting and large penalties for defrauding the election. To my knowledge we've never had a problem, certainly nothing that wasn't easily confirmed or denied by a recount.

What's the problem in the US exactly? Why do these machines need cards and removable memory anyway. Why would they even need to be networked? It seems everything could be pre-programmed since it's a one-time event.

I would like to have heard more about other people's reactions too, since it sounds like these machines may be even more untrustworthy than the studies suggest, and I don't get the resistance out there to requiring a paper trail as a back up system, which my site advocates. Even if the electronic machines could be made consistently reliable, that doesn't negate the possibility of system failures, programming glitches, power failures etc that could cause data alteration or loss, all of which could have occured in many of the described situations.

Andrew (the Aussie), the primary is the only one in which you "pick a party" to vote for.

In the general election, you may vote for any of the candidates, or in certain cases, write in a candidate.

So, yes, it's a multi-step process.

In the case of early primaries, the person you voted for may have been eliminated due to poor response in the primaries, so you get to make your final choice in the general election.

Also note that you don't have to vote in the primary to vote in the general election.

I'm from eastern Arkansas in the US and most elections here use paper ballots that are marked with a No. 2 pencil that will register in optical mark readers. I really like this method of voting and will not like to see it replaced by electronic voting machines. Voting on paper doesn't require electrical power to cast the ballot and can even be hand counted if the need comes up.

I cannot imagine the technical nightmare of using electronic voting machines with the volunteers who staff our polls. Most of these very dedicated people are well past retirement age and not that technically inclined.

My thoughts on the US primary voting method mirror those of others on this thread. In Arkansas you do not have to be a party member to vote in any party's primary. However, one can vote only in one party's primary election. Until the mid-1990s, primary elections were financed by the political parties. However, as the Republican party has grown in influence and enough interest in fringe parties has grown, the state has taken on the funding responsibility. The only problem that comes about during the primary elections is in run-off elections. You can get in a lot of trouble if you vote in one party's main primary election and then a different party's run-off election.

Another odd election tidbit in Arkansas is that until the 2002 federal elections, we did not have a true "secret" ballot. It took an amendment to our constitution to change the rules governing the traceability of ballots.

I think all the non-US visitors seem to miss the whole point of computerized elections. Here in the US there is no requirement that elections be honest or verifiable. Elections are simply for show, not to count who the public really wants. The winner is predetermined, just like in professional wrestling. It's all scripted in advance and we are all just playing along like we really believe. It's great fun and we end up with the most misunderestimated presidents on earth.

I think all the non-US visitors seem to miss the whole point of computerized elections. Here in the US there is no requirement that elections be honest or verifiable. Elections are simply for show, not to count who the public really wants. The winner is predetermined, just like in professional wrestling. It's all scripted in advance and we are all just playing along like we really believe. It's great fun and we end up with the most misunderestimated presidents on earth.

So true. Well said.

Why Electronic voting? Because it is NEW IMPROVED and of course its an easy way for a company to make lots of money selling new equipment in exclusive contracts. Remember HANGING CHADS ? Whoever sold those systems made a lot of money, and it sure didn't improve on the # 2 pencil used to mark paper ballots.

It is apparent to me that there will be a lot of problems with this system. There is no paper trail, and the president of Diebold has promised Bush that he would deliver electoral votes to him.

"Third-world" countries have a better system than we do. They circle the picture of the candidate of their choice with a permanent marker. A no brainer.

Until people start to demand a paper trail, I strogly urge voters to become absentee. At least there is something to refer to in the case of another Florida type fiasco.

If not, then welcome to police state 2004.

here, in france, you have to go to register yourself at city hall.
once you get to the poll station, you have a little enveloppe and as many different stacks of 2.5" by 3" bits of paper as there are people/lists to be elected. then you go hide yourself in a booth with a curtain. you put the paper that you select in the enveloppe, discard or keep the others in your pocket, and exit the booth.
then you show your id, sign on the register next to your name, and cast your vote by dropping the enveloppe into a plexiglas ballot box so that people present can see that the box was totally empty in the morning.
when the poll station closes, 3 people and a notary public are assembled to count the votes (3 people so that it's harder to cheat). counting the votes consists of the first person taking one enveloppe from the box, extracting the contents and showing to everyone present. the notary public takes note, and the person passes the contents to the next person that records the vote. this person then passes it to the next one that keeps stacks of the various votes.
counting takes about half an hour for the average voting station.
of course, as no chads are involved, re-counting takes about 5 minutes...

Wow, quite a few comments. I'll try to cover everything.

give me some more dirt please: Sorry about the blandness. I haven't kept up with all of the details of e-voting as much as I'd like to, and thus don't feel entirely qualified to draw firm conclusions. A lot of the problems speak for themselves. My opinion of the situation as a whole is similar to Avi Rubin's, except that I found San Diego's poll worker training to be less adequate. I didn't get much free time to talk with the precinct's Inspector, but she was similarly concerned about security.

dj radon: The failure rate of the memory cards is most likely low, but someone certainly could format one. However, they would need to be a poll worker (or close to one) during closing time, or break open the compartment without a poll worker noticing. I don't know if a copy of the results is kept inside each station, other than in the ballot memory card.

JB: Ouch. Good points. It's hard for me to pick out specific areas in which one system is better than the other. I see a huge potential for fraud while counting paper ballots, which (in the areas I'm familiar with) is done at each precinct, often with minimal supervision. In this case, anyone who wanted to rig a precinct would need very specific technical expertise, and the centralized counting is more carefully scrutinized - although I grant that there are some extremely worrying issues with the counting here, too. I also didn't want to sound too pessimistic in the article. Thinking again, I would prefer old-fashioned paper ballots with stricter procedures, but that seems a faint possibility at best. I think you may have me as far as the unsupported claim goes.

Paul Reinheimer: No problem; see Ross Presser's comment. In most cases, people were given ballots based on their registered party, as listed in the roster of voters. Voters who were registered as Independent (non-partisan) were the ones who were able to select a party's ballot on the spot. We made up a sheet of options later in the day, so that voters could indicate a party without saying it out loud. We did have a few items on the ballot that were "real": some changes to the law that required voter approval, as well as elections for some local govermental positions. Those were the only things that Non-Partisan Non-Partisan voters had on their ballots. All of the other things were just in preparation for the General Election this fall.

Spence: It sounds like your poll worker wasn't quite up to speed, as you may have guessed. The printers in the machines do not record each voter's vote at all, even with paper available. They're only used for printing the "zero report" in the morning and the totals report in the evening. They could be adapted to print individual vote records, except for the fact that they have a tendency to jam frequently.

CD: We would have inserted the VAC into a machine to check. If it popped out (indicating that it had already been voted), I think that we would have reprogrammed the VAC only if it had been issued to the voter a short time before, without enough time for them to have cast a ballot. But this could be a problem in precincts with less attentive or paranoid poll workers. The VACs themselves are smart cards - only the memory cards (which are in locked compartments) are SanDisks. The VAC smart cards are supposedly encrypted.

Blank: I'm from Minnesota as well, and have fond memories of that system. :) Simple and elegant, although counting does take a while.

Andrew: Yes, you can change your voter registration to a different party between elections. Your party membership is, unfortunately, more or less a matter of public record here.

Gunnar Van Vliet: Most things in the machines are in fact pre-programmed. My impression is that the cards are removable in order to save the hassle of having the central counting staff open (or power on) each machine to extract the votes. They don't trust the printed report alone.

I forgot to mention a few things. The "Public Relations For Pollworkers" sheet, which I posted along with the manual, is incorrect on at least one item. At the end of the first page, it states that "there is no port someone could use to plug in a keyboard." I believe that this is correct for the voting stations, but the PCM does have USB and Ethernet ports, although they were unused. Time alone with the machines could have proved interesting indeed.

I believe some other site mentioned that the seals on the memory card compartments weren't secure, and sometimes weren't applied properly. I can verify that - a seal came off by itself (intact) on at least one of our stations.

I feel much less positive about the system than I may seem in the article. I certainly do spend most of it on tearing down the system, based on various problems. Despite all of these, the basic fact is that it still works. It absolutely should work better, and there are very few excuses for its failings - but, nevertheless, it could be worse. This isn't to say that I won't be very unhappy if Diebold keeps its contract.

Thanks for your comments!

In soviet russia system was much simpler...

insert own funny bit here...

Hey, if in Soviet Russia, elections voted in me, I'd be a little nervous..

Here is the manual in .pdf format instead of those .jpgs:

To the non-USians here:

Primaries in some states are closed to those in the respective parties, and some are "open". If they are closed, then it is party policy that you only get to vote among the candidates in YOUR party (only makes sense). If it is "open" then anybody can vote for any candidate. Naturally "open" primaries lead to the possibility of one party trying to "corrupt" the other's primary by voting for the worst candidate in the other party. Anyway, in the case of closed primaries, the poll workers obviously have to know which party you belong to, to give you the correct ballot. In "open" primaries, I guess it doesn't really matter, but I think "open" primaries are still the minority and probably it is just easier to follow the same procedure even though the vote is not correlated to you. It might be useful to keep that data around to compare, for instance, pre-election ratings of candidates against the numbers of independent voters who voted - for instance to be able to detect that 233 Bar Party voters requested a Foo Party Ballot, and lo and behold, Candidate CrackHead in Foo party rose by 233 votes, which would lead to suspicion of hijacking/corrupting the primary.

Anyway, in the GENERAL election, any candidate is of course open to anybody regardless of party.

Spence: It sounds like your poll worker wasn't quite up to speed, as you may have guessed. The printers in the machines do not record each voter's vote at all, even with paper available. They're only used for printing the "zero report" in the morning and the totals report in the evening. They could be adapted to print individual vote records, except for the fact that they have a tendency to jam frequently.

WoW... If they are used twice per day (beginning and end of the day) how the heck can the "jam frequently" ?!?
Even Microsoft has better quality standards for their products! Methinks something isn't quite right.

Here is a detailed report on the security of the Diebold voting system:

http://www.raba.com/press/TA_Report_AccuVote.pdf

I voted on one of them in a tech demo they did here in GA. The whole time I was expecting:

"Good afternoon Dr. Falken, how about a nice game of chess?"

JB: Thanks for your post. We were also prompted to act as poll inspectors to monitor the electronic process. I am convinced that since both elections I worked on were "called" as or before I left the drop-off site, counties could save a ot of money by using scantron paper ballots. THe expected longer count time might deter those candidates so eager to admit defeat until real votes were finally tallied. All remarks you listed were consistant with our experiences. TJ

no requirement that elections be honest or verifiable

Here in Oklahoma people have gone to jail for messing with the election system. Judges are the ones that cheat and try to change the law on or after election day.

• hold polls open longer
• mess with counting chads
• change rules on how many recounts are allowed.
• let unregistered voters vote

What happens if after the primaries you decide the party originally selected is full of idiots, can you change?

Nope. You have to live with their own specific brand of idiocy (versus the other party's brand)
which is mostly expressed in who do they sell out to and for how much.

Thanks for the article. It's interesting to compare to when I worked as a polling clerk in Australia.

We have paper ballots with a box next to each candidate's name. To vote you write a "1" in the box of your top preference, a "2" next to your 2nd preference (if you want to indicate it) etc. In the US I believe this is known as "instant runoff" voting.

So you receive your ballot paper - having shown ID to the polling clerk who checks that you're on the roll for that area. The ballot papers have the order of candidates randomised. You then go to a booth, mark in pen your choice, fold it over, then walk to the exit and place it in the box. Sometimes you have 2-3 papers, depending on how many elections were occurring simultaneously (Senate, house of reps, local etc).

My days as a clerk were lots of fun - we'd shut the doors, collect all the boxes and empty the papers out of each one. Each clerk was watched by "scrutineers" appointed by each party/candidate. As a clerk you'd sort the papers into piles depending on the first choice candidates. Scrutineers would be watching carefully, sometimes querying decisions. After sorting you'd count the piles. Generally you'd have someone recount each pile as a double-check. Repeat for each of the preferences. Ballots would then be packaged up and delivered to district counting area where they'd be recounted (I think) as a double-check.

So, secure but reliant on people. The whole process would typically take around 1-2 hours to do a full sort and count.

The time it got painful was in the first local election in the Australian Capital Territory, which had the largest ballot paper ever (at the time) - over 1m long with 117 candidates, and something like 5 separate ways to vote (for party, for candidates directly etc). But event that was only a 5 hour marathon. The recount at HQ no doubt took longer (official results took several days, but that included postal and absentee ballots).

The ACT has recently moved to computer voting (owing to multiple recounts at a recent election with candidates 3 votes apart). I'm still not convinced it's a good thing. Sometimes it takes a while to do it right by hand, but that doesn't mean it's not worth taking a while to do it right.

wouldn't this be a good thing, in that it lets us vote early and vote often? how else are we gonna get the shrub out of office?

Sounds like you were close to my voting precint. I voted in Oceanside (practically Vista), and like you said, Party affiliations were announced aloud. I distinctly remembering every single person who voted while I was there was Republican, with one exception. A fella in his late 60s/early 70s was a democrat, but wanted to vote republican. In a upper, middle-class, bedroom community with the average home around $400k, this didn't surprise me.

An engineer who works for me is a former ATM technician, so we have talked about the Diebold systems and reliability. I didn't mind voting electronically, however as a manager over a Technical Support group, working on a software engineering MS degree, yada yada yada...I really am uneasy about relying on software for such a critical application. Especially now that I know it's running a Microsoft product. I would have felt better (but still uneasy) if it was remotely mounted from a HP or Sun machine. I've supported thousands of terminals, connected to hundreds of Sun servers with about a 99.9% reliability rate. I can't say the same about the M$ servers/workstations (PCs) I've administered.

John

What I found interesting was that if you inserted the Supervisor Card a 2nd time, you were presented with some options that a poll worker should not have seen. IIRC, there were modem xfer options and clearing vote options.

Another problem is that even though there were counts/records of voters via the sign-in sheets and the counts on the voter machines, there is absolutely nothing that can be done when there is a discrepence.

Here is how I would have designed a new voting machine: These machines would have to print the black dots on a scan-tron card the voter put into the machine after signing in. Based on the voters selections just like the current machines, the voter would push a CAST BALLOT button and the card would get the appropriate dots filled in. The voter could verify that before putting the cards in a clear box too. Then there would have been a paper trail, electronic voting where you see what you voted for and can change it if you want, AND a paper trail that can be counted/recounted quickly.

And I came up with this 2 years ago after thinking about it for about 10 minutes. These current bozoos, didn't even implement any kind of paper trail for this system and look at how many cities/counties purchased them...

signed,
a poll worker(sys insp) in San Diego

It's exactly these sorts of clear, well-written, and unemotional accounts that we need in order to make people open to hearing criticisms of the new electronic systems - rather than just labeling the critics as techno troublemakers.

I am gathering first person experiences from San Diego County and other counties where touch screen or other electronic voting systems are used. I especially need reports from poll workers, voters who may have had problems voting, and those who may have been the poll workers allowed to take the TSx and VCE home with them prior to the election.
It has been stated that much of the encoder problems were due to a low battery, yet the encoders were supposed to be plugged into 110 volt power at the polling location. I need more information on that subject. Also any experiences where the encoder initially worked and then quit or had problems later. In Alameda County one poll worker said the encoder worked for 1 1/2 hours and then had problems. Please send any information to:
conquip at ix.net.com.com

When was the last time you saw a register jam? Give me a freaking break. That kind of thinking belongs in the 80's.

I live in Brazil and would like to share my (little) experience with voting here.

First, there is no such thing as nation-wide primaries (parties select their candidates anyway they want, tipically by an open voting in a meeting). Second, for mayor, governor and president voting can take one or two turns. If a candidate gets more than 50% of the votes in the first turn, he is ellected. If not, the first two most voted candidates go to the second turn and ho gets more votes is ellected. No such thing as getting more votes and loosing the election... Also voting is obrigatory.

I work once at the elections, back when it was done on paper. Basicaly we used a piece of paper with the names of the candidates, with a box besid each one. The pieces of paper are signed by the pool works and carefully controled. There is a list of the registered voters in the precint. The voter come, an id is checked, he signs the list, get a ballot go a courtined desk, make an X in the box besides the candidate, fold the paper and put it into a bag. (For jobs with lots of candidates, there is a number assined to each one, the voter has to write this number). At the end of the day the bag was sealead and taken to a central location where the votes where counted. Nothing fancy, just a lot of paper work and a few days for the final results.

We have beem using eletronic vote machines for a few years. A big difference from US is that the same machine is used in all Brazil, and its design was (carefully, I hope) monitored by the goverment and the parties. Parties have access to the source code of the voting software (of course, this is not enough but is a good start). With the machines, voting is faster, it is easer and less error-prone. Results come in a few hours instead of a few days. The model used today have a printer inside, each vote is printed in a piece of paper that drops in a bag, so votes can be manualy recounted. An external printer that allowed the voter to check the vote was tried in the past, but it was too expensive, two troublesome and most voters did not took the time to look to the printed votes. There still are concerns about fraud (specially on the software side) and secrecy (the voter has to identify itself at a console before voting, you have to trust that the id is not stored alongside your vote).

Daniel

I have a question:

You said:
"We had another technical problem later in the day, when a voter reported that his summary screen (the last step before casting the ballot) was blank. I confirmed that it was; everything else seemed normal, and the boxes were checked next to each candidate on the ballot..."

So you saw who this guy voted for? I thought all elections in the US were secret ballots?

John: Oddly enough, I saw a lot of young people in my precinct voting Republican, and a number of old ex-military types voting Democrat. It might have been an oddity of the area, since older, long-time Solana Beach residents were probably there back when it was a less prosperous surfer colony.

Ken: The ballots are "more or less" secret, but I did see a number of peoples' votes. Doing so was more or less inevitable when helping them with problems. Most voters were fairly comfortable with this; I helped the few who weren't without approaching their screens, although it did take longer. (Poll workers have had to help voters with paper ballots as well, although not as much.)

also a worker: I didn't know that about the Supervisor Cards...that's very worrying.

I have a personal rule: Never be the first persont to try a new product. Let some other poor sap work out the bugs first.

That said as a San Diego voter I took the absentee ballot option. It work very well and I think I will be staying with that method in the future. Also there is something quite satisfing about holding a paper ballot -- it makes my choice seem some how tangible.

The Diebold people at training did resets using the supervisor cards but I figured it required a special password or something. I guess not. What caused me to go back in was that one of the printers got stuck and kept printing over the same line when doing the final summary.

I too was quite surprised/concerned to see those options. It therefore won't take but one person to clear out quite a few votes if they wanted to. Do that in a few well placed precints and a close race could be won. Kinda like Florida and the butterfly ballots. ;-)

Just want to pitch my idea for a hard-to-tamper-with voting system.
1. You sign in at the voting station, and your name is added to a list of people who voted.
2. You are given a unique number where the number is covered in some way that it cannot be peeked at, like a scratch-off lottery ticket.
3. You go in the booth, cast your vote, also providing the number, and the vote and number are registered. Then you can dispose of the ticket, burn it, keep it, whatever you feel like.
4. At the end of the voting, both the list of people that voted is made public, and a seperate list of numbers and votes is made public (csv file). You can now make sure they recorded your vote accurately (if you kept or memorized the number), total the votes yourself, and make sure your name is on the other list (or not on it if you didn't vote).

Certainly still some things to be worked out with this, but I think it would be fairly harder to tamper with votes undetected.

I guess if we all accepted the obvious fact that our political, corporate and legal systems are all corrupted to the hilt, it would be easier to accept that the electronic voting machines are intentionally designed with the intent to simplify "fixing" an election. After all, why else would they be so resistant to the addition of a paper trail. The reader who stated in an earlier post that political winners are pre-ordained scores "nothing but net" on this issue. The corporate media plays it's role in pumping up the pre-ordained winner simply to sucker us into accepting the outcome.

I am wondering if the situation in Flordia was not planned this way , So that they whoever they are could introduce computer voting ....

Nonthing makes sense anymore , not even the most private of rights we have is safe ( maybe it never was )

Give us the chads back !!!! at least we all now know to make sure the chad is out ......

Many of our recent elections use electronic vote counting of paper ballots, with very simple and straight forward marking of the ballots with a #2 pencil.

The ballots are marked in private and then handed to an official who slips the ballot into
the machine face down (so the votes are hidden).
They are slid out of a sleeve so that there is
NO way anyone other than yourself knows your vote. Further, there is checking done to verify
that only one vote was cast for each position.
If there were too many votes or not enough, the
appropriate message was displayed and the voter was given another chance to correct their vote,
by having the paper ballot returned to the sleeve.

If the machine didn't find any problems, or the voter didn't care about the problems (they were given a choice), the ballot was automatically sealed in a case beneath the machine, ready if
a recount was requested.

Pretty slick, and there is a paper trail, and counting happens super quick.

Our last few elections were fully counted within an hour of the close of the last ballot.

This is for Canadian elections - we really didn't understand why it took many days to decide the 2000 'election' of George Bush.
We were stunned at how broken an election system
could be if it took weeks to finally declare a winner....

It looks like the only way the Democrats can beat these Republican voting machines is if we all use absentee ballots, thats what my family will do.

Did anyone follow commenter Spence's link? (Comment #7) What he describes is MAJOR. It means that the eVote terminals in his pricinct were never zero-summed before the machines were used for voting. (i.e., if some test votes were cast to check the machines out before the elections, those votes were actually COUNTED as valid votes during the primary.)

A coalition of activists in Georgia has gotten a bill passed in the Georgia Senate to mandate a paper trail for the Diebold Accuvote-TS machines in this state. We need some mainstream media coverage to get force this bill through Georgia's House and get it implemented by Nov. 2004. Check out the URL below for more info. There's not very many days left for action before the legislature shuts down and the bill dies.
The "Democratic" Secretary of State Cathy Cox wants the bill dead, and Georgia's "Democratic" legislators are being partisan and supporting her position. The Georgia's Republicans are using this bill as a tool to attack the Secretary of State here, but they still have the moral high ground on this issue. Any of you who can pass this story onto the media, or activate contacts in Georgia on this issue, Thank you.

http://atlanta.indymedia.org
http://www.voterchoice.org

To a poster above: Why don't you ask the disabled community what they think of VVPATs? You probably aren't at all concerned about their privacy now are you? Thought about vote buying? Thought about a system that prints one thing and records another? VVPAT is a joke. You think it takes a long time to vote now, wait until those are required! And what if the printout is wrong? What if someone just 'claims' it is wrong. You people need to think this through. Computerized voting systems have been in use for over 20 years in this country. Why is it an issue now? I will tell you why, some clever lobbyists that work for vendors that sell Voter Verified Paper Audit Trail solutions are stirring things up. And many of you are getting sucked into their BS. And there is not ONE VERIFIED (key word here kids) case of an anomolie with these systems that wasn't due to HUMAN error. Prove one case that was the systems fault. All the problems you speak of have existed as well with lever, punchcard, paper, and every other system of voting (yes and marbles too). And for you geeks out there, shut up. Of course a system can be hacked if you have total access. What are the chances of that? I have hacked at a number of these systems both in and out of the processes in which they are used. For the most part, they are secure. In the cases in which a problem was found, the risk assesment was so low it was hardly worth mentioning. No less safe than an ATM. AND who the heck is going to reformat a compact flash card during election day? PLEASE! That is about as likely as someone coming in and peeing on one of the voting devices.

The process described you described with System Inspectors, Voter Access Cards, and Supervisor Cards is much more complicated than necessary.

We should vote on plain paper ballots, drop the ballot into a transparent box, and be allowed to watch the votes counted in the same room that evening.

It's simple and transparent. More security, less complications.

I write about this here:
http://www.moveleft.com/moveleft_essay_2004_03_21_how_elections_should_be_conducted_hand_count.asp

WOW paper ballots!!! You are a genius! Nobody has ever thought of using paper ballots before! They are so error proof(not), secure(really not), and quick and easy to tally (really really not). Who is willing to wait 2 to 4 weeks for a result? Have you ever been an election official? If you don't know the process then you should educate yourself before coming up with a so-called solution. There is a reason the equipment has evolved in this direction.

All this ill thoughtout hysteria reminds me of the year 2000 angst. It reminds this guy of just the same thing:

http://seattletimes.nwsource.com/html/opinion/2001898724_lance09.html

Lets fix a problem that only exists in the media!

How vulnerable are SANDISK cards to external magnetic sources?

Is it possible to bring a strong magnet with you when you get close to the machine...like when voting...and wipe the SANDISK?

"TheSkyIsFalling!" is incorrect on several points:

1. An accessible voter-verified paper audit trail (VVPAT), as will be required in California by 2006, will not compromise ballot secrecy for disabled voters. Such accessible VVPATs must be independently verifiable by disabled voters in a manner (e.g., an automated audio-interface that reads the VVPAT contents back to the disabled voter via headphones) that preserves ballot secrecy.
2. A VVPAT does not enable vote selling, since the VVPAT is deposited into a ballot box; it is not retained by the voter and thus cannot be used for vote-selling schemes.
3. Regarding a system that "prints one thing and records another", that is precisely why most legislation that mandates VVPATs also mandates mandatory, surprise recounts in randomly selected precincts. If a voting system is routinely printing one thing and recording another, that discrepancy will be caught via such mandatory recounts. Such mandatory recounts are an essential part of any VVPAT solution.
4. If the printout is wrong, then that ballot is marked as spoiled, and the voter is given another opportunity to vote. This is no different than the existing case of a voter spoiling a paper ballot.
5. While computerized vote tabulation systems have been used for over 20 years, direct recording electronic voting systems have only been in widespread use for the last few years.
6. There are many documented cases of malfunctions of electronic voting systems. For details, please see:
http://verifiedvoting.org/resources/documents/ ElectronicsInRecentElections.pdf
7. You pose the challenge: "Prove one case that was the systems fault." That's easy. In the document referenced under item 6, read about the malfunction of the ES&S iVotronic electronic voting machine used for early voting in Wake County, North Carolina. ES&S admitted that their firmware was faulty, admitted liability, and paid damages to Wake County. We have signed letters on file from the Director of the Wake County Board of Elections that document this incident.
8. While some of the problems inherent with electronic voting machines (e.g., lack of a paper trail) have existed with lever machines, lever machines cannot be easily reprogrammed, nor can one easily embed malicious instructions inside a lever machine, or a paper ballot or a marble, for that matter. Thus, it is inaccurate to say that "all of the problems you speak of" have existed in these other systems.
9. You state that "the risk assesment was so low it was hardly worth mentioning". That was certainly not the conclusion reached in any of the security studies that have been conducted regarding electronic voting machines: a) The Johns Hopkins/Rice University report, b) The Maryland/SAIC report c) The Ohio/CompuWare report d) The Maryland/RABA report. All of these identified extremely serious security risks that they felt were extremely worthy of mention.
10. The risks posed by electronic voting machines are not simply a "problem that exists in the media". Rather, this is a problem that potentially impacts the integrity of our elections, and one that deserves careful study and analysis based on documented facts. It is not one that should be glibly dismissed via a series of ill-informed opinions that are asserted without any corroborating evidence.

Bob K,

Where is your 'corroborating' evidence?

1. Why don't you speak to some disabled advocacy groups about that issue? They will beg to differ with you. And, if a blind person’s only choice is to have their ballot read to them, why isn't that good enough for everyone else?

2. So what you are saying is that you can guarantee that all receipts are placed in a ballot box? So you will trust the same voters that could potentially sabotage a voting system to place their receipts in a ballot box? Of course it can be used to facilitate vote buying. That is why this country doesn't use it! That is why you have never voted and recieved a reciept! Hmmmmm….

3. That may not be a bad idea concerning the use of VVPATs. However, that doesn’t make them necessary. Did you have the same processes with other forms of vote recording devices? Why now?

4. Yea and who is to say that the voter won’t do this just to disrupt an election? We have some features available to us with these new systems that allow a voter to review and correct every choice they make. Why complicate and lengthen the process with a paper receipt? It will only serve to confuse the voter.

5. Not true. DRE have been in use for 20 years. Sequoia, Shoup, and Microvote have all had systems out since the early 80’s.

6. You really need to check the facts on those accounts. Many of them didn’t mean that votes were lost or otherwise improperly manipulated. Most of those issues had to do with PROCESS. Those that didn’t could be problems that result from any voting process, regardless of the technology used. For instance, did you know that it many states it is perfectly legal to cast a blank ballot? That is what happened in Broward. Guess what else? The system worked because they knew about them! What would you say if the ballot was paper? Someone erased the marks? Someone stuffed the ballot box? Please come up with irrefutable evidence.

7. OK so you have some irrefutable evidence here. At least it appears as such. So what? One system had a firmware problem. And there have never been miscounts or other mechanical system malfunctions? And once again the system worked, the jurisdiction was able to determine what happened and rectify the situation. That is the best you could ever hope for not matter what process or equipment you use. So a lever machine or punchcard device has never malfunctioned. Paper ballots have never been manipulated. Right, and monkeys are flying out of my butt as I write this.

8. The problems indeed have existed; just the manner in which the change can be affected is different. Nice try. You really need to look into this issue before speaking on it.

9. You didn’t understand what I wrote or I wasn’t clear. Most of those reports did not find any showstopper issues. Avi Rubin is a very smart man but he is out of his league. He knows not of what he speaks. And don’t bring up that crap about him being a poll worker for a DAY, it means nothing. I have read in detail all of these reports, have you? Doubtful as you are misrepresenting their conclusions. Especially when you consider the process involved surrounding the use of voting equipment. Many of the recommendations where process-related. And there where vendors that came away relatively problem free. I don’t know else how to say this but I will try again. If a system is used properly the chance of anomalies occurring are minimal. If you can’t get to the machine, you can’t hack it. If the vendor doesn’t have access to load ‘upgrades’ then it can’t be done. Security is only as good as the persons implementing it. All voting system software is certified. It is rigorously tested. And if the jurisdiction using the software and hardware do so properly, the results obtained are no less accurate than any other system currently in use.

10. Why can’t I be any glibber than you? You toss out supposed ‘facts’. You have ill thought out solutions. You probably read Vanity Fair for all your technical information. You are making something out of nothing. Fear is a powerful force, especially when it is used to deceive. The media and its minions are doing a very good job at pushing their agenda on an under-educated public. Perhaps they will win. But in the end, we will be no better off than we are now. And we probably will have spent billions of dollars in the process fixing something that just isn’t broken.

The fact is no system is perfect. Democracy isn’t perfect. And obviously the ability of Americans to rationally and logically think through an issue isn’t perfect. And don't get me started on how stupid voters can be. You wouldn't believe what happens on election day. The fact that we even allow some folks to vote in this country is a bigger problem that the equipment we use (tongue somewhat in cheek). The sky is indeed falling.

Oh and by the way, looks like Wake County had some troubles following the process now didn't they? Acceptance testing is an integral part of the process. Had they been paying attention, the whole incident would never have occured. But of course that isn't the issue is it? We have to blame the big bad voting machine companies. The republican controlled voting machine companies, full of their specially hidden computer code that ensures a republican win! The sky is indeed falling! Run for your lives!!!!

Firstly, Canada uses paper ballots for federal elections, and they usually find out the results the same night.

Secondly, the reason there have been security problems with paper ballots in the past is that the box is stuffed beforehand or moved and stuffed before the counting.

Using a transparent ballot box, which is kept in one place from the start of voting until the end of counting, and allowing the public to watch the ballot box from the start of voting to the end of counting, avoids those problems.

Have you ever voted before? Do you realize how many candidates/issues can be on a ballot? I am afraid that in any reasonably sized jurisdiction, especially on during a general election, there would be very little chance to get results in a timely manner. The idea sounds good but isn't practical by any stretch of the imagination. If we can't trust our technology, lets just all go back to living in mud huts.

Have I mentioned that nothing happened when Y2K hit? Many thought the end was near, what a waste of energy. We have the same situation here too.

Also, have you noticed that no one is holding election officials responsible for any of the problems in CA or Ohio? If someone was checking behind the company, like any reasonable organization should do, many of these problems would be averted. It's called PROCESS, TESTING, DUE DILIGENCE!!! Check the version of the software before you run it. Check the system before you use it. I am not suggesting that Diebold isn't culpable, but the jurisdicitons involved should share the responsibility. Most of the issues we are seeing in the press would have been avoided if the election officials knew what they were doing. (Also there haven't been any accuracy issues in CA!)

The size os voting precincts isn't set.

Big voting precincts can be made smaller to make public hand counts easier.

With regard to Y2K, millions of dollars were spent addressing the problems, and we don't know what would have happened if that money wasn't spent. Y2K isn't an example of a problem which was just ignored.

Also, Y2K involved an unintentional bug.

With voting machines, the issue is intentionally rigging an election, and how do we design a voting system which gives people confidence that their votes were counted accurately.

Havng people vote on plain paper and watch the ballot box through to the end of voting and then through to a public hand count would accomplish that.

The votes could be counted in one night with voting precincts the right size.

Wow, it is becoming quite evident that most of you know nothing about this topic. The chances of precincts being resized are about a billion to one. Guess who controls the size of precincts? Politicians do! And they won't make them smaller just because it is the right thing to do (not that it would help paper balloting). Smaller precincts won't reduce the size of a ballot. Most issues that fill up a ballot span many precincts, making precincts smaller would just increase the work involved in creating and managing an election, already greatly under-funded and under-manned as it is.

And yes, Y2K is a perfect example. It wasn't the government, loopy liberal organizations, or so-called experts that made the changes. It was business. It was a practical matter for the companies that stood to lose business as the result of any anolomies that occured. Would a company risk everything to create and market a system that, when properly used, would fail? Critical thinking here people.

Just what are we trying to fix? All systems have shortcomings. Life ain't perfect. There is NO conspiracy! There is no evidence that any electronic voting system (remember we have been using them for 20 years) has ever been hacked. Have these systems been incorrectly deployed by election officials? Yes, in some cases. Do these election officials know much about Information Technology? Barely anything. What do you think happens when a county or state installs a brand new computer system with little or no IT knowledge? Exactly what is happening in California, Indiana, Ohio, and Florida.

And do you really think a clear ballot box hasn't been thought of before? Come on people, use your brains, stop reacting and THINK first!

Making the precincts smaller wouldn't reduce the number of races per ballot.

It would reduce the number of ballots per precinct.

In the Georgia election for the US Senate in 2002, many people think the machines were rigged.
Uncertified software was used.

You can say they can't prove it was rigged, and they can say you can't prove it wasn't. There was no way to a recount.

Plain paper ballots, or optical-scan ballots, or voter-verified paper ballots printed by electronic voting machines, can be recounted.

Eric Eric Eric....

Your just not listening now are you? You are also posting rumours.

The Georgia patch issue is NOT fact. It was an accusation made by a FORMER employee of Diebold. There is NO proof that uncertified software was used. And once again, if it happened, it is just as much Georgia's fault as it is Diebolds. There are ways to prevent software from being altered. This is still America, at least the last time I checked. Proof is important in a Democratic Republic. Checks and balances buddy. You know, all that democratic stuff you are trying to protect.

There are other voting machines, that AREN'T electronic, that DON'T have a paper trail, what about them? You don't need a recount on an electronic system. Do you have a bank account? Do you have every receipt from every transaction on every account you own?

You can recount anything, even computer results, the difference is the computer will always come up with the same result. The counting of paper ballots, ballot reciepts, and optical scan ballots will almost always give different results. You might want to check into the statistics before getting behind a particular solution. READ ABOUT PAPER BALLOTS!!!! THEY ARE BEING REPLACED FOR A REASON!!!!

Read this....

http://www.americandaily.com/item/5551

And of course, you obviously could give a rats ass about persons with disabilities. Who cares if you can't see or write. We don't need your vote, and you certainly don't have the right to vote in private!

The AutoMark from Vogue Election Systems helps people with disablities and people who don't speak English to mark a paper ballot:

http://www.vogueelection.com/products_automark.html

Regarding that article saying the machines will be tested on Election Day, firstly, how do we know that will be done in a particular state, and secondly, are they put into a testing-mode for that? If so, how do we know the testing-mode is the same as the real thing?

Well I am not sure about the actual privacy afforded by the system you mention. It doesn't appear to be completely private as there is a possibility that either someone would look over the voter's shoulder or read their paper ballot. If you were blind, I think you might have a problem with that system. As for systems that are already being used, many of them blank the screen during audio ballots to ensure privacy.

You also bring up a good point. How do we know that the poll officials are testing? GO WATCH! You can request to observe. I also find it quite offensive that we hold our election officials in such low esteem. There has to be some amount of trust involved in any system. The folks that run our elections aren't crooks. They deserve to be given the benifit of the doubt. That goes for all the people that volunteer to assist on polling days. These aren't sub-70 IQ dolts that organize and maintain our elections. We need to STOP and THINK about the entire system. It isn't just the machines, its the people, laws, and processes.