Paper Vote Canada

A coalition of concerned residents is urging voters to boycott using the state's new electronic voting machines and to stick to the standard paper ballot option when they go to the ballot booth for the general election.

Members of Safe Vote Hawai'i, which describes itself as "a coalition of Hawai'i's leading technology professionals, community activists and official election observers," said they have concerns about the integrity of the electronic machines because they do not produce an immediate "paper copy" for each vote cast.

"We feel that paper audit trails are mandatory for a safe e-voting election and we don't have any way of getting that for this election," said Jason Forrester, spokesman for the group and a senior systems analyst for SAVVIS, formerly Digital Island.

Honolulu Advertiser - October 8, 2004
Integrity of e-vote machines questioned

Atomz searching (the search in the sidebar on the right) has been fixed.

It had forgotten about this entire search index, for some reason.

This is the scope of the efforts of people in the United States just to try to monitor the e-voting machines and issues associated with them on voting day.

computer scientists focused on thoroughly evaluating the technical aspects of e-voting machines. This task will be left to more than 1,400 techies participating in a domestic monitoring project of the Election Protection coalition, organized by almost 60 U.S. groups.

The coalition will disperse 25,000 monitors to voting sites in 17 states around the country. One of the coalition's projects is TechWatch, run by the Verified Voting Foundation. TechWatch has recruited technology professionals to observe the operation of voting technology during tests before the election and on Election Day itself.

The Verified Voting Foundation has also developed a web-based software application that will allow coalition volunteers to respond to voting incidents in real time. The coalition runs a hot line that voters can call to report problems. When they do, hot-line operators enter the details into the Election Incident Reporting System. The system maps the problems and sends an alert to a monitoring team leader in the area, who can then dispatch a mobile response team of volunteer lawyers, techies or others to observe how the problem is handled.

The groups tested the system during the Florida primary election, when 296 hot-line calls were entered into the system, of which 18 were voting machine problems.

Another group, Votewatch, will collect statistical data on the election. Among other things, Votewatch will compare the vote tallies posted in polling places at the end of the day with cumulative unofficial counts for an entire area reported later that night. If the numbers differ, it would indicate that the counting software at individual polling places failed to count some votes. Volunteers would make note of the technology that was used at those polling places to gather statistics on the accuracy of systems.

The Election Protection coalition also includes groups that will monitor issues besides voting technology, including voter intimidation and disenfranchisement.

Does that sound to you like an improved, less-expensive voting system that people trust?

Wired News - Oct. 05, 2004
U.S. Elections Under a Microscope

Kwesi Addae, a fifty-three-year-old former political-science professor from Ghana and the founder of Pollwatch Africa, has monitored elections in half a dozen shenanigan-prone countries, including Togo, Nigeria, and Guinea-Bissau.

...

Addae, as one of twenty foreign election observers brought here this month by the left-leaning human-rights group Global Exchange, is inspecting the creaky mechanics of American democracy at close hand. He signed up for the two-week mission, he said, “out of fascination.”

In Africa, Addae had seen phantom polling stations, ruling-party rent-a-mobs, ballots bought with beer. But nothing in his experience had prepared him for the iVotronic touch-screen voting system. The iVotronic is the latest in electoral technology, a light, flat-screen computer that will be used this November in parts of Florida, among other places. ...

The observers crowded around. “You’re quite fortunate to be able to touch and feel the same thing that those voters will be using on Election Day,” a salesman from Election Systems & Software, the machine’s manufacturer, said. ...

“Let’s vote,” Addae said. ... Addae was next. He bellied up to the machine and poked at the screen. Nothing happened. Someone handed him a small cartridge—a “personalized electronic ballot,” or P.E.B. Addae examined the cartridge quizzically, then fumbled as he tried to insert it into the machine. “Put it back in for just a second,” an observer named David MacDonald, a former member of Canada’s parliament, said. “Now remove it.”

A ballot appeared on the screen. For the purposes of this demonstration, the election was a local one, and the candidates for mayor were all actors; for city council, hockey players; and for school board, musicians. For some reason, they were also all Canadians. “O.K.,” Addae said, peering over his glasses. “I want Michael J. Fox.” Passing over Neve Campbell, John Candy, and Dan Aykroyd, he touched Fox’s name, and a little green check mark appeared. Addae ran his finger down the list of candidates in other races: Gordie Howe, Ken Dryden, Bryan Adams, Shania Twain, Gordon Lightfoot. “I want Bobby Orr. I want Neil Young,” he said. When he was done, Addae punched a flashing red button marked “ vote .” The machine beeped twice, and a message appeared on the screen: “Your ballot has been cast. Thank you for voting.”

Addae shook his head. “It’s not very easy,” he said. Back home, Ghanaians still vote the old-fashioned way, by writing an “X” on a paper ballot and dropping it into a box. “Thereafter, they are counted one by one, and the figures are written down and announced,” he said.

...

After he was finished with the computer demonstration, he ambled across the room, where some observers were voting with paper ballots. He grabbed a ballot and filled in the bubbles next to his choices. The salesman collected twelve ballots and loaded them into a optical scanner. Whoosh —the ballots shot through. But there was a problem. The machine rejected five of the ballots, including Addae’s. The observers had used a pen instead of a No. 2 pencil. This election observer was ready to deliver his verdict: “I am not impressed at all!”

The New Yorker - 2004-10-04
The Talk of the Town - Visiting Dignitaries - Dry Run

This November, Eileen Rivera Ley, 41, will vote by herself for the first time. Blind voters in Maryland and several other states will use electronic voting machines equipped with technology that allows the disabled to vote independently.

It used to get crowded whenever Rivera Ley voted. Blind, Rivera Ley had to rely on someone else to read the ballot aloud, then vote for her. That meant as many as four people -- Rivera Ley, the person who pulled the levers and election judges from both major parties as witnesses -- huddled in the voting booth.

In Canada, only one person assists you (if needed) in the voting booth; I don't think anyone observes.

Wired News - AP - Oct. 03, 2004
E-Voting Fans: The Disabled

Imagine your bank teller accepting a deposit and then saying, "Oh, you don't need a receipt. It's all in the computer." On Nov. 2, that's essentially what millions of citizens will be told when they cast ballots on new electronic voting machines. Forty-two states are poised to use this latest technology, but with only 28 days left until the presidential election, some states are still debating whether to provide a paper confirmation of each voter's choices.

Potential problems with electronic voting — and very real mishaps — have gained more public attention in recent months, and manufacturers and election officials have tried to play down concerns. But some state officials have also chosen to build in more safeguards to ensure that the electronic vote data, if corrupted either accidentally or maliciously, have a backup. That means one thing: a paper record.

USA Today - Christian Science Monitor - October 3, 2004
Observers remain uneasy over e-vote machines

this year, the City of Edmonton has rented 21 audio-electronic voting machines, which will allow citizens who can't see well to cast their ballots without help. And everyone else who votes in advance will do so electronically, too -- traditional paper ballots will be available only on election day, Oct. 18.

...

Electronic voting machines have been severely criticized in the U.S., where they were used for the first time in 2002. In July 2003, professors at Johns Hopkins University in Baltimore released a scathing report that said the machines were open to tampering and made meaningful recounts impossible because they leave no paper trail.

But operations manager Steve Thompson, who led the project, is confident the electronic voting terminals the city is renting will perform without a hitch. He points out that the machines are independent of one another and store voting records in three separate places, including on a removable pack that can be taken from a broken machine and read in another.

"We don't believe that the failure of a single or even two or three machines would in any way jeopardize the ability to count the votes that had been cast on that machine," Thompson said.

...

There are no plans, however, to make electronic voting the norm. He estimates the cost of mainstream electronic voting would be about $1.25 million, a price tag he thinks council and taxpayers will find prohibitive.

For heaven's sake, the point isn't whether the votes can be read, the point is whether the votes are accurately recorded.

Edmonton Journal - September 22, 2004
Ballot machine boon for blind voters

This story points out another trend which I saw in the reports from Toronto.
In order to "save money" they like this idea of renting machines, whether it's loaning their own machines out, or using someone else's machines.

So that's great. As if the security problem wasn't hard enough, they're letting the machines change hands repeatedly.

Can high-tech voting machines prevent a repeat of America's 2000 electoral fiasco—or will they make things worse?

IT IS not often that the dry subject of voting technology makes the headlines. It famously happened in America's presidential election in 2000, when the previously obscure differences between hanging, pregnant and dimpled “chads”—the small flakes of paper punched out of cards by mechanical voting machines—suddenly became a national crisis.

...

Touch-screen machines have problems of their own, as another Florida election vividly illustrated last January. In a local election held on January 6th in parts of Broward and Palm Beach counties, 10,844 votes were cast, and Ellyn Bogdanoff won by 12 votes. Under Florida state law, a result this close triggers a manual recount. But no recount was possible, because there was nothing to count: the voting machines' only paper output is the final tally.

The lack of a paper trail has made new touch-screen voting machines hugely controversial. As Robert Wexler, a Democratic congressman, likes to point out, “a reprint is not a recount”. Critics also complain that there is no way to tell if the machines are faulty, insecure or rigged.

The Economist - Sep 16th 2004
Electronic voting: The trouble with technology

Link from crisispapers.org - Electoral Integrity.

Using the RFP number 3412-99-01464 I have found some more Toronto documents.

I love this analysis.

4.0 Direct Recording Electronic (Touch Screen)

...

Recounts using these machines have proven to be one of the closest systems to 100%.

ha ha ha ha ha

That's hilarious.
Of course the recounts are always the same. THERE'S NOTHING TO RECOUNT.
They are saying, roughly "every time I ask this machine to produce its vote count total, it produces the same numbers".

Hmm, when I store a number in my pocket calculator's memory, it produces the same number each time too. Maybe we should replace our paper voting with pocket calculators.

Plus which, "closest to 100%"?
Are you saying sometimes the machines report different results from their memory banks?

(Sorry, in sarcastic mood today.)

Anyway, the report is
May 6, 1999

To: Corporate Services Committee
Budget Committee

From: City Clerk

Subject: Voting and Vote-counting System - Municipal Elections
www.city.toronto.on.ca/legdocs/1999/agendas/committees/cs/cs990520/it005.htm

It examines the following types of voting systems:
1.0 Mechanical Lever Machines
2.0 Punchcards
3.0 Optical Mark Reading (Optical Scan)
4.0 Direct Recording Electronic (Touch Screen)
5.0 Telephone
6.0 Mail
7.0 Automated Terminals or Kiosks
8.0 Internet

Conclusion:

After reviewing all of the current voting/vote-counting systems available to the City and examining them in relationship to the principles established, it is recommended that the best voting/vote-counting system for the City of Toronto would be a combination of optical scan - voting place tabulator and direct recording electronic (touch screen). The optical scan - voting place tabulator provides the best practice available given its reliability, integrity and similarity to paper ballot. The electorates' familiarity and acceptance with this type of vote-counting equipment ensures the greatest likelihood of continued success utilizing this type of system.

The advancement in technology makes direct recording electronic (touch screen) a viable alternative in a limited variety. Its portability provides greater flexibility to employ these units for specialized needs (i.e. institutional voting, physically challenged voting and an advance voting program). By employing the use of these units the advance voting program could support an additional 500 potential advance voting locations.

Request for Proposal:

In order to comply with Council's request to provide updated costs of a recommended voting and vote tabulation system, and to ensure sufficient lead time for the delivery of the system in time for Election 2000, a Request for Proposal (RFP No. 3412-99-01464) was issued for an Optical Scan Voting System and Touch Screen Voting System in February 1999. The purpose of this Request For Proposal was to provide updated figures for the Capital Works Program and determine the most effective Optical Scan Voting System and Touch Screen Voting System in the marketplace.

Subject to the approval by Council of this report, the results of the Request For Proposal together with recommendations on the award of contract will be forwarded to the Administration Committee for consideration in July, 1999.

This is the analysis they provide for DRE machines

Advantages:

a) easiest of all voting systems for the voter;

b) results are quickly available;

c) no extra costs for materials (i.e., ballots);

d) recounts have proven to be extremely accurate; and

e) most technologically advanced form of voting in the marketplace.

Disadvantages:

a) notable large capital investment required; and

b) new voting method for voters - could result in long lineups.

I don't even know where to begin.
What difference does it make how "technologically advanced" it is?
Two disadvantages? How about
c) Massive possibility for incorrect results due to fraud or incorrect programming
d) Complete loss of transparency in elections process

Who wrote this analysis, a high school student?
Sorry, that's an insult to high school students.

Earlier on in the report, there is a very telling presentation of background

Historically, composite paper ballot elections have been prone to human error, particularly:

a) subjective discretion applied when deciding valid votes;

b) tallying votes when the deputy returning officer is communicating votes to the poll clerk orally; and

c) transposition errors when carrying figures from tally sheets to final statements.

In addition, the operating costs associated with composite paper ballot elections is as much as three times that of an automated election. The increase in costs include;

a) the requirement for more voting subdivisions with fewer voters to ensure the manageability of the counting process;

b) the employment of staff at the voting places and additional staff to count ballots at the close of voting; and

c) the associated recount costs.

Summary: Humans and paper are unreliable and expensive. Computers are better. Subjectivity is eliminated, all is perfect objective computer logic. All hail our robot king!

There's only one problem with this presentation.
It's wrong.

I would rather trust humans than machines.
Voting is not about achieving the maximum cost-effective calculational efficiency.

It's about having a system that people can understand and trust, so that they have faith that the election reflects the will of the people.

A paper system with humans counting has explicit, obvious workings and checks and balances.

An electronic system has nothing. Invisible electronic pulses. Silent hidden programs.

You might as well take all the paper ballots into a locked room and say "trust us, we have a wonderful machine in this magic room that will count everything, sorry, you can't see it, we'll tell you the results in a moment".

This is a huge chunky extract of Toronto Administration Committee stuff.
And in case you're wondering,
1) yes, the John Hollins who was the Director of Elections for Toronto has now moved on to be Chief Electorial Officer for Ontario.
2) Election Systems and Software is a US voting equipment vendor

I wonder if they promoted this as "$13 million Canadian taxpayer dollars go to US vendor, to help our elections work as well as they do in the USA".

I'm guessing not.

How good is ES&S? Well, I think this article from Black Box Voting sums it up:
ES&S Trying to Compete With Diebold In Race for Crappiest Voting Machines

Anyway,
THE CITY OF TORONTO
City Clerk's Division

Minutes of the Administration Committee

Meeting No. 3
Tuesday, July 13, 1999
www.city.toronto.on.ca/legdocs/1999/minutes/committees/adm/ad990713.htm

-6. Request for Proposal No. 3412-99-01464 for the Acquisition of 2000 Optical Scan Vote Tabulators and 100 Touch Screen Voting Units. The Administration Committee had before it the following:

(i) joint report (June 15, 1999) from the Chief Administrative Officer and Acting Commissioner, Corporate Services, the Chief Financial Officer and Treasurer and the City Clerk recommending that:

(1) the City Clerk be given authority to enter into a contract with Election Systems and Software Inc., being the proponent with the highest evaluated score, for the acquisition of the necessary optical scan vote tabulators and touch screen voting units, including all necessary support and service agreements, at a capital cost not to exceed $13.05 million ($1.0 million in 1999 and $12.05 in 2000), such contract to be in accordance with the Request for Proposal and the Proposal submitted, and on terms and conditions satisfactory to the City Clerk, and in a form satisfactory to the City Solicitor; and

(2) the appropriate City officials be authorized to take the necessary action to give effect thereto; and
(ii) report (June 28, 1999) from the City Clerk responding to a request by the Administration Committee on the possibility of Elections Ontario and Elections Canada cost sharing the purchase of the City of Toronto s vote tabulators and touch screen voting units; advising that at this time there is no interest at either Elections Ontario or Elections Canada to cost share the purchase of vote counting equipment with the City of Toronto; that Federal election legislation does not permit the use of equipment and Ontario's provincial election legislation currently only permits equipment to be used in by-elections; that should either of these jurisdictions choose in the future to use vote counting equipment, the City of Toronto could contract with the jurisdiction to rent the City's equipment; and recommending that this report be received for information.

(iii) communication (July 9, 1999) from Mr. Clinton H. Rickards, Director, Canadian Sales Global Election Systems Inc., expressing disappointment that his request to participate in the presentation of electronic voting machines to the Administration Committee has been refused.

__
______

The City Clerk, and the following officials from the Clerk s Division, gave a presentation to the Administration Committee during the lunch recess respecting the foregoing matter, and filed a copy of their presentation material:

- Mr. John Hollins, Director of Elections;

- Ms. Janet Andrews, Senior Elections Consultant and Co-Chair of the Evaluation Committee;

- Mr. Stephen Miller, Senior Elections Consultant; and

- Mr. Greg Essensa, Senior Elections Consultant.
________

The following persons appeared before the Administration Committee in connection with the foregoing matter:

- Mr. Jamie Aiello, t.e.s.t.;

- Mr. John Meraglia, Co-Chair of the Evaluation Committee;

- Mr. Bob Urosevich and Mr. Clint Rickards, Global Election Systems Inc.;

- Mr. Dan McGinnis, Vice President of Sales, Election Systems and Software;

- Councillor Michael Walker; North Toronto;

- Councillor Mario Silva, Trinity Niagara;

- Councillor Mario Giansante, Kingsway Humber; and

- Councillor Bob Davis, York Eglinton.

A. Councillor Minnan-Wong moved on behalf of Councillor Davis that the Administration Committee refer the foregoing joint report (June 15, 1999) from the Chief Administrative Officer, the Acting Commissioner of Corporate Services, the Chief Financial Officer and Treasurer and the City Clerk, to the Chief Administrative Officer for report to the September 7, 1999, meeting of the Administration Committee:

(i) on the various other options, including leasing and rental, in consultation with the proponents who submitted this as part of their proposal, such report to include financial comparisons to the purchasing options; and

(ii) including information on the feasibility of borrowing this equipment from other jurisdictions. (Carried) B. Councillor Moeser moved that the foregoing motion by Councillor Minnan-Wong be amended to provide that the Chief Administrative Officer also report on the ongoing cost of maintenance of the batteries used for these voting units. (Carried)

C. Councillor Bussin moved that:

(1) the City Auditor be requested to:

(i) review the business case and the financial and technical evaluation of the proposals, including a rent-to-buy option, where included as part of the proposals;

(ii) review and evaluate the systems used in Vancouver, Ottawa, Chicago, Cook County, Philadelphia and Seattle,

and submit a report thereon to the aforementioned meeting of the Administration Committee;

(2) the appropriate staff be requested to provide to the Members of the Committee and Councillors, this week on a confidential basis, additional information on the evaluation criteria, weighting and ranking, including copies of the resultant independent testing authority for compliance with the standards of the U.S. and Federal Elections Commission;
and

(3) the City Clerk be requested to:

(i) give consideration to and report on providing orientation sessions for candidates, agents and scrutineers; and

(ii) report to the aforementioned meeting of the Administration Committee on the option of using a non-composite ballot, and on the option of requesting a change in Provincial legislation to allow a ballot design based upon the Provincial ballot. (Carried)

D. Councillor Mahood moved that the report (June 28, 1999) from the City Clerk be received. (Carried)

The decision of the Administration Committee, therefore, is as follows: The Administration Committee:

(1) referred the foregoing joint report (June 15, 1999) from the Chief Administrative Officer, the Acting Commissioner of Corporate Services, the Chief Financial Officer and Treasurer and the City Clerk, to the Chief Administrative Officer for report to the September 7, 1999, meeting of the Administration Committee:

(i) on the various other options, including leasing and rental, in consultation with the proponents who submitted this as part of their proposal, such report to include financial comparisons to the purchasing options;

(ii) including information on the feasibility of borrowing this equipment from other jurisdictions; and

(iii) on the ongoing cost of maintenance of the batteries used for these voting units;

(2) requested the City Auditor to:

(i) review the business case and the financial and technical evaluation of the proposals, including a rent-to-buy option, where included as part of the proposals;

(ii) review and evaluate the systems used in Vancouver, Ottawa, Chicago, Cook County, Philadelphia and Seattle,

and submit a report thereon to the aforementioned meeting of the Administration Committee;

(3) requested the appropriate staff to provide to the Members of the Committee and Councillors, this week on a confidential basis, additional information on the evaluation criteria, weighting and ranking, including copies of the resultant independent testing authority for compliance with the standards of the U.S. and Federal Elections Commission;

(4) requested the City Clerk to:

(i) give consideration to and report on providing orientation sessions for candidates, agents and scrutineers; and

(ii) report to the aforementioned meeting of the Administration Committee on the option of using a non-composite ballot, and on the option of requesting a change in Provincial legislation to allow a ballot design based upon the Provincial ballot; and
(5) received the report (June 28, 1999) from the City Clerk.

(Chief Administrative Officer; City Clerk; City Auditor - July 13, 1999) (Clause No. 26(e) - Report No. 2)

Paper Ballots

Ladies and gentlemen, I give you the paper ballot: An idea whose time has come again.

from Tech Central Station - 11/05/2002 (probably US date notation)
by Glenn Harlan Reynolds

seen on c2.com wiki - Voting Machine Discussion

Jason Kitcat reports [New Media Knowledge]/IPPR are hosting 'E-Voting: Policy and Practice' in London [England] on 4th November.

This will get filled in over time as I learn more.
Feel free to provide any updates/links you may have.

Federal
- there is no e-voting at the Federal level

Provincial
- Ontario will be starting pilot projects
- NB is interested

Municipal
* Ontario
- Markham used the Internet
- Ottawa uses optical mark-sense
- City of Greater Sudbury uses optical mark-sense
- Toronto uses ?optical mark-sense?

I personally could care less about proportional representation.
Don't take offense, it's just not my thing. Feel free to pursue it if you want.

HOWEVER, do not try to solve proportional calculation problems by electronic expediency.

In A Made-in-Canada Proportional Representation System, Mohamed Elmasry writes

Electronic counting of votes will make the system easily implementable.

Bad idea.

Just to show you what fun people we are in Canada, here's an official Elections Canada Frequently Asked Question:

Is someone allowed to eat a ballot?

Short answer: No.

This is very cool: Global lessons in e-voting.

CNet looks at a few different e-voting systems from around the world.

Via j-dom.

Q. Why has Council authorized the use of Optical Scanning Vote Tabulators?

A. Council directed the City Clerk to review alternative voting methods for the 2003 Municipal Election. The City of Greater Sudbury was perhaps the last city its size in Ontario that had continued the use of a paper ballot. As a result of this review, Council authorized the use of optical scan vote tabulators for the November 10, 2003 municipal elections.

While still allowing the voter to mark their choice of candidate on a paper ballot, automated vote counting equipment offers several benefits to both voters and the municipality, some of which include:

* accuracy of vote tabulation
* consistency in vote tabulation
* rejection of overloaded and/or improperly marked ballots at the time of voting, therefore reducing voter disenfranchisement
* quick tabulation of individual voting place results following the close of the voting (results will be sent by modem from each location).
* the requirement for few voting locations and the need to hire smaller number of election staff.

Q. Are there safeguards in the Vote Tally System to Guard Against Fraud?

A. The procedures put in place by Council require the Returning Officer to conduct computer logic and accuracy tests before, during and after the vote on election night.

Prior to the advance vote, the Returning Officer and the Third Party Auditing Firm hired by the City of Greater Sudbury will attend the acceptance testing of the vote tabulation system, which will include:

* reviewing a sample of pre-audited test ballots prepared by the City of Greater Sudbury's equipment vendor;
* reviewing the reports prepared by the vote tabulation system once the preaudited test ballots have been entered into the system; and,
* comparing the results listed on the reports to the sample of pre-audited ballots and identifying any discrepancies.

On election night the City of Greater Sudbury's Third Party Auditing Firm will select at random seven vote tabulators - one for each of the six Wards plus one machine from the advance vote - to:

* compare the election results as listed on the reports prepared by the selected machines to the election results recorded at the City of Greater Sudbury's central election site and report on any discrepancies; and,
* compare the report produced by the selected machines following a re-entry of the votes cast to the original printout obtained at the time of the closing of the polls and report on any discrepancies.

Other security measures to ensure the integrity of the voting process and the security of the vote tabulation equipment and software have been provided by the Returning Officer and the City of Greater Sudbury's equipment vendor.

elections.city.greatersudbury.on.ca - It's as easy as 1-2-3-Vote

So if I understand this, they compare the results from the machine against... new results from the machine. Or they compare the results from the machine with the results transmitted by the machine. Does anyone compare the machine results against a manual ballot count?

The Netherlands is currently holding the election for the Regional Water Management Boards (my translation of "Waterschappen"). One can vote by mail or by Internet. The latter attracted my curiosity, and I poked around the 'net a bit to see what people thought about the idea.

It appears that a test election was held in order to test the procedure and get some feedback from test voters. An often quoted feedback was that "Only 26% of the test voters expressed concern about the possibility of fraud". ONLY 26%?? This response seems to be interpreted as a vote of confidence for the system.

Another nugget: "The secrecy of the vote is guaranteed. The relationship between the voter identity and his login code is removed from the file before the votes are counted".

The FAQ also has an interesting statement. An independent body (TNO) has investigated the security of the voting method. They concluded that "Voting by Internet is not less safe than voting by mail or phone". This formulation implies that the procedure is actually not very safe, and they know it.

from Risks Digest 23.55, 30 September 2004.

Brazil has recently become the world pioneer in electronic voting and registration. When it held national elections in October 2002, 91 million out of its 115 million registered voters turned out – more than 70 percent of those of voting age, and 3 million more than voted in the US elections that same fall. In terms of global electoral history, the number of votes received by the winner, Luis Ignacio de Silva (“Lula”), was second only to Ronald Reagan’s total in 1980.

To handle this heavy turnout, Brazil relied heavily on electronic voting. The Tribunal Superior Eleitoral (TSE) had been experimenting with electronic voting systems since the early 1990s, becoming a real pioneer in the use of “DREs” (“direct recording electronic”) voting machines. Brazil first used DREs on a large scale in its 1996 elections, with 354,000 in place by 2002. For that election, it deployed another 52,000 “Urnas Eletronica 2002,” a state-of-the-art DRE that had been designed by Brazilian technicians with the help of three private companies – Unisys and National Semiconductor, two US companies, and ProComp, a Brazilian assembler that has since been acquired by Diebold Systems, the controversial American leader in electronic voting systems.

Because Brazil has been willing to commit to such a large-scale deployment, each Urna costs just $420, less than 15 percent of the cost of the $3000 touch-screen systems that Diebold features in the US. The Brazilian system lacks a touch screen; voters punch in specific numbers for each candidate, calling up his name and image, and then confirm their selections. The numerical system was intended to overcome the problem of illiteracy, which is still a problem in parts of the country. To handle operations in remote areas like the Amazon, the machine runs on batteries up to 12 hours. Initially there were no printed records, but the Electoral Commission decided to retrofit 3 percent with printers, to provide auditable records.

Like any new technology, Brazil’s approach to electronic voting is by no means perfect. Indeed, significant concerns have been voiced about the system’s verifiability and privacy – especially about the TSE’s recent move to eliminate the printers, supposedly because they slowed voting.

Among the most important proposed improvements are a requirement that all voting machines produce both electronic and paper records, in order to leave an audit trail and increase voter confidence in the system; that system software be based on “open” standards and available for audit; and that the system for identifying eligible voters be separated from voting, to insure privacy.

Submerging Markets - September 29, 2004
Democracy in America and Elsewhere: Part IIIB. Campaigns, Voting, and Representation

... Ensuring the reliability, security, and verifiability of public elections is fundamental to a stable democracy. Convenience and speed of vote counting are no substitute for accuracy of results and trust in the process by the electorate.

from ACM Press Room ACM Recommends Integrity, Security, Usability in E-Voting: Cites Risks of Computer-based Systems.

Via Slashdot ACM on E-Voting.