Paper Vote Canada

1. Reconstructing votes.

It occured to me that it may be possible to reconstruct who voted for whom with optiscan as well as with the paper-roll based trails.

If the optiscan takes the large ballots and stacks them one by one on top of each other, as each person votes, then you can do the same as with the paper roll: if you know the order in which people voted, you can then match each ballot to each person and find out who they voted for.

2. Precious Memories.

Paper ballot boxes are big. They're just physically big objects. That makes them difficult to alter if people are watching.

DRE memory cards are small, and given the industry, will only get smaller. That means in theory, once voting is done, if you're physically deliverying the memory cards...

a little "hey look over there" slight of hand and...

you've swapped one or more cards for ones you have pre-coded, thereby replacing thousands of votes in an instant. Think this is unlikely?

Bev Harris has recommended that due to the insecurity of the modem dial-in setups used to report votes to the central tabulator, people should Simply drive the vote cartridges from each polling place.

Avi Rubin has reported

One by one, we removed the memory cards from the machines. I held them in my hand as chief judge Marie was ready to load them into one of the machines that we designated as the accumulator. How fragile. All of the votes from the entire precinct in my hand. Substituting those cards with five identical looking cards, one could replace all of the ballots that were cast with bogus ones.

Canadians have not been exposed to electronic voting. Elections Canada has no official position on electronic voting. A 2003 recommendation was made to allow eligible voters to register to vote online, but has not been adopted by Parliament.

The London Free Press - October 30, 2004
TODAY'S BUSINESS LAW: Electronic voting debated

People need to understand that electronic voting (primarily optical scanners, but some touch screens as well as Internet voting) has been used at the municipal level in Canada, and it may be coming at the provicial level as well.

Wikipedia entry on Electronic voting.

Also see earlier posting on c2.com discussion.

There are a bunch of stories in Risks Digest 23.58

Just some thoughts.

In a US election they are typically voting for many different offices plus propositions. This is one of the reasons that pushes them to automated counting. It also makes a challenge for this "machine generated paper trail" because each voter is going to generate a lot of votes for different things that all have to get printed.

Also, I read an important point somewhere (Evoting Experts I think). If the paper trail is on roll paper, all you need to know is the order of who voted, and you can reconstruct who voted for which candidates. So much for a secret ballot.

Meanwhile, in Canada, we have dead simple elections. We often are voting for only one office.

WE DON'T NEED AUTOMATED COUNTING.

Did those controversial electronic voting machines properly record Americans' votes yesterday? We'll never really know. Without a paper trail, there's simply nothing to check against in order to verify accuracy.

ComputerWorld - November 3, 2004
Opinion: E-voting 'success' unverifiable -- and that's the problem

Defend Democracy Canada

To defend the heart of democracy by challenging the use of electronic voting systems until such time that the idea of converting to electronic voting systems is abandoned or that a system exists that will allow for the accurate, safe, and reliable counting of votes that will uphold the basic required elements of the electoral process.

...

At this time this site has the specific goal of exploring and defending the integrity of elections in Ontario. More specifically, in the counties of Stormont and Dundas.

A fairly generic article from CTV. Includes video though (fairly slow to load).

About 50 million Americans are expected to vote on the touchscreen machines on Nov. 2. Officials insist they have been tested, and can indeed be trusted.

...

"I'm not sure which is a greater concern -- that the machines will fail, or that they'll be rigged," Rubin said.

As a result of the controversy, record numbers of people are choosing to vote in advance, on paper ballots.

CTV.ca - October 14, 2004
E-voting vulnerable to attacks, fraud: expert

CTV News: Alan Fryer in West Palm Beach, Florida (Windows Media, 2:32)

I may not be able to keep up with US e-voting news, but these guys probably can.

www.evoting-experts.com

I also found another site linked from theirs that looks good

www.votersunite.org/news.asp

There is no way I will be able to keep up with the torrential flood of e-voting reports I expect post-US-election.

The important thing for voters is to be aware of your rights, and to ask for a paper ballot if possible. This is allowed in some areas of the US.
However, you should still vote even if the only option is electronic.

The only way to be 100% certain that your vote won't be counted is not to vote at all.

The US election is Tuesday November 2, 2004.

Each voting machine must be protected, both during the election but more importantly during the long periods before and after the election.

An appropriate level of security would have a secure facility, access card control, video surveillance, etc. A secure vehicle would transport them, never letting them out of authorized control.

But instead, these things are, I am not making this up, shipped around in people's cars and stored in people's garages.

Joanne Kartchner goes out to her garage every morning to check on the electronic voting machines that were dropped off early last week.

The garage is locked and the machines are sealed in special bar-coded tamper tape, but Kartchner, a longtime election worker who volunteers her Palo Alto home as a polling place, wants to be double sure. ``I make sure they are all there,'' Kartchner said Saturday.

The registrar of voters began delivering electronic voting machines to 822 polling places in Santa Clara County last weekend, among them at least 20 community or senior centers, 50 apartment complexes or mobile home parks, and 80 private residences.

The Mercury News - Oct. 31, 2004
E-voting in homes raises concerns

US NPR (National Public Radio) has had lots of stories on electronic voting.

Here's a search listing them:
search.npr.org Query=electronic+voting

Some selected stories (with audio):

The Tavis Smiley Show - August 17, 2004
E-Voting Reliability in Doubt as Elections Near

Talk of the Nation - August 6, 2004
Electronic Voting Still Controversial

Votergate is the investigative documentary feature film uncovering the truth about new computer voting systems, which allow a few powerful corporations to record our votes in secret.

...

This 30 minute Special Edition is designed specifically to help viewers navigate past the fear and spin being thrown at this critical issue.

www.votergate.tv

30 minute video available for viewing.
Found via blackboxvoting.org

It's worth watching, it's well done.
One of the things they walk through is the Diebold central tabulator ("GEMS").
I tend to mostly talk about the individual machines, but the central tabulator is obviously a very attractive target. And Diebold, master computer security experts, have done GEMS... on Windows. In Microsoft Access. Oh, and as a bonus, with banks of modems dialing in to it.

So here are the great things about voting machine testing.

Most of it is based on mechanical testing concepts.
So they do shake and bake tests, to see how the machines survive harsh conditions.

In terms of testing the actual machines, at the most they might do some functional testing in ... wait for it... test mode.

So that's great. The advanced machine software testing consists of tapping onscreen buttons in test mode and having the machine say "I counted a vote".

All it would take would be a few hours with Visual Basic and you could fake that up pretty well.

Anyway...

The three companies that certify the nation's voting technologies operate in secrecy, and refuse to discuss flaws in the ATM-like machines to be used by nearly one in three voters in November.

Despite concerns over whether the so-called touchscreen machines can be trusted, the testing companies won't say publicly if they have encountered shoddy workmanship.

They say they are committed to secrecy in their contracts with the voting machines' makers -- even though tax money ultimately buys or leases the machines.

"I find it grotesque that an organization charged with such a heavy responsibility feels no obligation to explain to anyone what it is doing," Michael Shamos, a Carnegie Mellon computer scientist and electronic voting expert, told lawmakers in Washington, D.C.

The system for "testing and certifying voting equipment in this country is not only broken, but is virtually nonexistent," Shamos added.

Although up to 50 million Americans are expected to vote on touchscreen machines on Nov. 2, federal regulators have virtually no oversight over testing of the technology. The certification process, in part because the voting machine companies pay for it, is described as obsolete by those charged with overseeing it.

The testing firms -- CIBER and Wyle Laboratories in Huntsville and SysTest Labs in Denver -- are also inadequately equipped, some critics contend.

...

"Four years after the last presidential election, very little has been done to assure the public of the accuracy and integrity of our voting systems," Rep. Mark Udall, D-Colo., told members of a House subcommittee in June at the same hearing at which Shamos testified.

"If there are any problems, we will spend years rebuilding the public's confidence in our voting systems," Udall said. "We need to squarely face the fact that there have been serious problems with voting equipment deployed across the country in the past two years."

In Huntsville, the window blinds were closed when a reporter visited the office suite where CIBER Inc. employees test voting machine software. A woman who unlocked the door said no one inside could answer questions about testing.

Shawn Southworth, a voting equipment tester at the laboratory, said in a telephone interview that he wouldn't publicly discuss the company's work. He referred questions to a spokeswoman at CIBER headquarters in Greenwood Village, Colo., who never returned telephone messages.

...

Also in a sprawl of high-tech businesses that feed off Redstone Arsenal and NASA's Marshall Space Flight Center in Huntsville is the division of Wyle Laboratories Inc. that tests U.S. elections hardware, including touchscreens made by market leaders Diebold Inc., Sequoia Voting Systems Inc. and Election Systems & Software Inc.

Wyle spokesman Dan Reeder refused to provide details on how the El Segundo, Calif.-based company, which has been vetting hardware for the space industry since 1949 in Huntsville, tests the voting equipment.

"Our work on election machines is off-limits," Reeder said. "We just don't discuss it." He did allow, though, that the testing includes "environmental simulation...shake, rattle and roll."

Carolyn Goggins, a spokeswoman for SysTest Labs, the only other federally approved election software and hardware tester, refused to discuss the company's work.

...

critics led by Stanford University computer science professor David Dill say it's an outrage that the world's most powerful democracy doesn't already have an election system so transparent its citizens know it can be trusted.

"Suppose you had a situation where ballots were handed to a private company that counted them behind a closed door and burned the results," said Dill, founder of VerifiedVoting.org.

"Nobody but an idiot would accept a system like that. We've got something that is almost as bad with electronic voting."

CTV.ca - AP - Aug. 23 2004
U.S. e-vote machine certification criticized

In this months CACM ... some Yale students show that altering only a single vote per machine would have changed the electoral college outcome of the 2000 election.

Slashdot - 30 October 2004
More on the Dangers of eVoting