Elections Ontario Belatedly Replies to our December 7, 2012 Letter but Again Leaves Key Questions about Accessible Telephone and Internet Voting Unanswered

If you don’t now receive our updates directly from us, sign up for AODA Alliance e-mail updates by writing to our new email address: aodafeedback@gmail.comFollow us on Twitter and get others to do so as well! Twitter.com/aodaalliance
Learn more at: www.www.aodaalliance.org
UNITED FOR A BARRIER-FREE ONTARIO

June 6, 2013

SUMMARY

There will be by-elections coming up soon in Ontario. With a minority Government in the Ontario Legislature, a general election could happen any time. Over a million Ontario voters have a disability. In the next Ontario general election or by-elections, will all Ontario voters with disabilities at last be able to mark their own ballot in private and verify their choice, for the first time in Ontario history?

Elections Ontario is the independent provincial government agency responsible for administering provincial elections in Ontario and for making sure they are fully accessible to voters with disabilities. Elections Ontario is required by law to submit a report to the Ontario Legislature by June 30, 2013 on alternative voting technology for voters with disabilities. In plain language, this means options like telephone voting and internet voting, as alternative options to the traditional paper ballot, marked in a voting booth at a provincial polling station during a provincial election.

Here is yet another instalment in the seemingly-unending saga of our efforts to make voting in Ontario fully accessible to voters with disabilities in Ontario. Back on December 7, 2012, we last wrote Elections Ontario to give feedback on our concerns regarding barriers facing voters with disabilities, and to ask a series of important questions. Read the AODA Alliance’s December 7, 2012 letter to Elections Ontario.

As the latest news, Elections Ontario wrote us back on March 11, 2013. We set out that letter below. With that letter, Elections Ontario sent us the PowerPoint presentation it presented to the AODA Alliance delegation that met with senior Elections Ontario officials back on November 19, 2012. We also set out the contents of that PowerPoint presentation below. Elections Ontario’s latest letter leaves several important questions unanswered. The long road to accessible voting for voters with disabilities seems to have no end in sight.

MORE DETAILS

1. Our Unanswered Questions

As has too often happened in the past, Elections Ontario leaves a number of our important questions unanswered, even though it took fully three months to compose its response to us. This is especially troubling since Elections Ontario has repeatedly claimed to be committed to accessible voting for voters with disabilities, and to be open, accountable and transparent in its work.

* In our December 7, 2012 letter, we asked Elections Ontario for informative specifics on why it has been so reluctant to test telephone and internet voting during an Ontario by-election. The Elections Act gives Elections Ontario a mandate to test this accessible voting technology in a provincial by-election, starting in 2012. Elections Ontario doesn’t answer this important question. In our December 7, 2012 letter, we wrote:

“On reflection, we believe that it is not sufficient for Elections Ontario to simply keep invoking the non-specific refrain that telephone and internet voting is “complex” and that there are “risks.” This was known when the Legislature enacted the recent amendments on this topic in the Elections Act. Operating an election is complex and has risks. We, voters with disabilities, and indeed all the public deserve to know specifically and in detail what risks and complexities have been addressed, and which ones remain outstanding. We deserve to know why telephone and internet voting can successfully work in Cobourg and Markham, Ontario for municipal elections, while voters in those same communities cannot be allowed to bear the same complexities and risks during an Ontario election. What exactly did you learn during the RFP process that you did not know before it, that led you to decide not to proceed with a test in 2012?”

* We asked Elections Ontario to explore other options for testing telephone and internet voting, apart from trying it out during a provincial by-election, and to let us know what conclusions it reached about this possibility. Their response says nothing about this. Our December 7, 2012 letter included:

“Beyond the foregoing, we understand that you face a practical problem when trying to test telephone and internet voting in a by-election. By-elections are typically called at the last minute. Elections Ontario is not necessarily given prior notice about them before they are called. To get around that problem, we therefore asked you at our meeting to explore other ways to test internet and/or telephone voting, other than at a by-election. Even if an alternative avenue for testing telephone and internet voting would not meet the Elections Act’s precondition for lifting the ban on these accessible voting technologies, it could help you break this ongoing log-jam. You agreed to consider this.

While we would prefer not to wait to 2014 for this technology to be tested, one avenue for Elections Ontario to test it could be during the next municipal elections. Some Ontario municipalities have already successfully used these options for voting at the municipal level, and expect to do so again in 2014. Elections Ontario could monitor that to gather all the information it needs. A second option would be for Elections Ontario to stage a fictional by-election at a date that Elections Ontario chooses, and to invite the public to use telephone and internet voting, to see how it works. A third option is to offer these voting options in one or more ridings in a general election, if one occurs in 2013, for example, but not as an actual way to cast one’s vote. It would instead be a test, to run in parallel to the actual vote. By that option, anyone who takes part in the pilot must still cast a vote using existing avenues under the Elections Act at their polling station, at an advance poll, or by mail-in ballot.

By any and all of these options for testing telephone and internet voting, there would be no threat to the actual security of the vote in an Ontario election. There must be other ways beyond these three options, to test it. We welcome a response from you on what options you explore and what conclusions you reach.”

* We asked Elections Ontario to investigate what banks do to protect against hacking into IVR systems i.e. automated telephone answering systems that give you options to press 1 for one option, press 2 for another option, and so on. Elections Ontario did not answer this. We wrote: “We suggested at our meeting that you should explore the measures that those organizations use to protect IVR systems from being hacked or disrupted. Those organizations would be an obvious source of expertise and experience. Hackers would likely wish to steal a bank account well before they would want to steal a vote. Moreover, those organizations’ IVR systems operate 24/7 all year, and not just during the short window of an election campaign.

We understood from our meeting that you will look into this. We are eager to hear back from you whether that investigation adequately addresses this concern.”

* We asked Elections Ontario when it sets the preliminary voter’s list for an election. It didn’t answer. Our December 7, 2012 letter stated: “Fourth, your business case indicates that telephone or internet voting would only be available to those who are already on the preliminary voters’ list. We were concerned about this restriction. We asked you to clarify when that list is set. “

* We asked Elections Ontario to make a preliminary report on telephone and internet voting in February 2013, well before the final June 30, 2013 deadline. Its letter does not answer this. Moreover, Elections Ontario has not told us why it wouldn’t make the requested interim report. Our December 7, 2012 letter stated:

“At our meeting, we noted that your final report on telephone and internet voting must be delivered to the Legislature no later than June 30, 2013. We have been urging Elections Ontario for over two years to get that report delivered well before that end date. The Elections Act does not require Elections Ontario to wait until June 30, 2013 to deliver this report.

At our meeting, we noted that a spring election is a real possibility in Ontario. There may not be a sitting Legislature to receive your report on June 30, 2013. We asked if Elections Ontario could release an interim report in February 2013, in the event that your final report is then not ready. Voters deserve to know as much as possible about the state of your work, and about any final or tentative conclusions you have reached, at that time. Whether that interim report is formally delivered to the Legislature or is simply posted on your website, voters with disabilities and all voters would be able to learn where you stand at that time on this issue.”

* We asked Elections Ontario to re-constitute its Accessible Elections Advisory Committee that it had set up at public expense, and later disbanded. We have no answer. Our December 7, 2012, letter stated: “We recommended at this meeting that you reconstitute your recently-disbanded disability advisory committee. Elections Ontario only formed that committee in 2010. You confirmed that that committee was very helpful and constructive. We confirmed that its work is not yet finished.”

* We asked for a follow-up meeting in January or February 2013, to give further feedback on information that Elections Ontario gathered, before it formulated its final report. Elections Ontario appears to be prepared to meet us after it renders its final report. While that is helpful, it means we have lost out on the critical opportunity to give final input at the vital stage when Elections Ontario already has in hand all the data it needs, and is deciding what to recommend in its final report. Our December 7, 2012 letter stated:

“At this meeting, we asked for a further meeting with you in January or February 2013 after you have gathered most or all of your input on telephone and internet voting. We would welcome this to have a further discussion before you finalize your positions.”

2. Further Reflections

* We are deeply troubled that Elections Ontario took the full three years it was given by the Ontario Legislature back in 2010 to prepare its report to the Ontario Legislature on telephone and internet voting. This leisurely pace for preparing such a report is unjustified. It simply delays a workable solution to the barriers which too many voters with disabilities face when trying to exercise their fundamental democratic right to vote.

* Where Elections Ontario did opt to answer some of our questions in its March 11, 2013 letter to us, one answer sticks out like a sore thumb. Back in 2010, when we were pressing the Legislature to amend the Elections Act to ensure that voters with disabilities could enjoy the full right to accessible voting, the Ontario Government hitched its horse to one option alone, namely a stand-alone accessible voting machine in a polling station. It adopted that option on the recommendation of Elections Ontario, and after Elections Ontario tested it in a by-election. Clearly all the information that the Government got about this device came from Elections Ontario.

During those 2010 debates in the Legislature, the Government several times threw around the price of $10,000 per accessible voting machine or $15,000 per voting machine plus related training. It did so to support its decision to only require a minimum of one accessible voting machine per riding. It argued that at that high cost, it was simply too expensive to put one such machine in every polling station. In sharp contrast, the City of Chicago and the related Cook County has one accessible voting machine in each of thousands of polling stations.

As but one illustration of how this came up, during Standing Committee hearings at the Legislature on March 31, 2010 into proposed accessibility amendments to the Elections Act, Liberal MPP Greg Sorbara (who earlier chaired the Legislature’s Select Committee on Elections) asked the following to a representative of the Canadian National Institute for the Blind:

“Mr. Greg Sorbara: My question is along the lines of my friend Mr. Prue. Let me put it this way: In the work that we did in the select committee, we found that, currently, the voting machine of the type that Ontario is going to be using costs, with the training associated with it, about $15,000. As Mr. Prue said, there are about 15,000 locations—polling places—in the province of Ontario for a general election. If you do the math, that’s $225 million to supply all polling places with this kind of capacity. Does the CNIB realistically suggest to the government, which is looking at $20 billion in red ink, that to assist the disability community its first priority should be to spend $225 million on voting machines for the next election?”

During earlier proceedings of the Legislature’s Select Committee on Elections, chaired by senior Liberal MPP Greg Sorbara, Chief Electoral Officer Greg Essensa appeared and gave a detailed presentation about the proposed accessible voting machine.

Since those legislative debates, and as a result of our pressing this issue, Elections Ontario has released costing data. It showed that that $10,000 per machine figure was a gross overstatement. It is obviously troubling when a Government uses inaccurate and over-inflated costing information to claim that more effectively meeting an accessibility need costs too much.

We asked Elections Ontario where the Government got the $10,000 per machine figure that it threw around back in 2010. In its March 11, 2013 letter to us, Elections Ontario surprisingly claims not to know. It stated:

“You inquired about the differences in actual costs and costs that were discussed during Bill 231 debates. I understand that at one point during the debate a figure of $10,000 per machine was indicated. I am not aware of the source of that information.”

This is simply hard to believe. Back in 2010, it was obvious that the Government was relying exclusively on the advice and experience of Elections Ontario in adopting that accessible voting machine. When Elections Ontario invited us to take part in a test of that machine, it was clear that Elections Ontario was dealing directly with the company that provides it. We know that Elections Ontario, and especially the Chief Electoral Officer Greg Essensa, was in regular direct contact with Liberal MPP Greg Sorbara, who had lead carriage of this issue for the Government in the Legislature.

Elections Ontario’s March 11, 2013 letter to us was from The Deputy Chief Electoral Officer, Lauren Wells, not from the Chief Electoral Officer Greg Essensa. It was Mr. Essensa whom we personally wrote and asked about this costing information, back on December 7, 2012. Ms. Wells simply says “I am not aware of the source of that information.”

That leaves glaringly unanswered the key question whether chief electoral Officer Greg Essensa, the key player at Elections Ontario on this issue, is himself aware of the source of that inaccurate costing information. This is quite disquieting. Elections Ontario owes us and the public a complete answer to our inquiry.

* In sharp contrast to Elections Ontario’s lax approach to action on telephone or internet voting, Australia appears to be moving ahead in this context much more quickly. It is about to use a form of telephone voting in an upcoming election. Learn more about Australia‘s plans for telephone voting. Ontario voters deserve nothing less. Many Ontario voters without disabilities would no doubt welcome and enjoy the option of voting via telephone or internet. Our proposal for telephone voting would go further than Australia’s, but Australia still appears to be far ahead of Ontario.

* We hope and trust Elections Ontario will at least meet the legal deadline of delivering its final report to the Legislature by June 30, 2013, as the Elections Act requires. We remain deeply troubled that the Ontario Government remains in direct violation of the Accessibility for Ontarians with Disabilities Act. It still has not appointed the mandatory Independent Review of that law. The legal deadline for making that appointment was back on May 31, 2013. Learn more about the Ontario Government’s violation of the Accessibility for Ontarians with Disabilities Act.

* We also hope and trust that Elections Ontario will make its entire final report on telephone and internet voting, including all supporting appendices and documents immediately available by June 30, 2013 on its website in an accessible format, and not only in PDF format.

* We know that by-elections will soon be called in Ontario. We very strongly urge that Elections Ontario deploy telephone and internet voting in these by-elections. Elections Ontario committed to us many months ago that it would be ready to test these options in a by-election in 2012. It has now been three years since the Legislature gave it that mandate. There are no excuses for missing this next opportunity to test these accessible options in an Ontario by-election.

* Learn all about our multi-year campaign for fully accessible voting in Ontario elections and about election promises that have been made to us on this issue.

We always urge you to send us your feedback. Write us at aodafeedback@gmail.com

Please “like” our Facebook page and share our updates:
http://www.facebook.com/pages/Accessibility-for-Ontarians-with-Disabilities-Act-Alliance/106232039438820

To sign up for, or unsubscribe from AODA Alliance e-mail updates, write to: aodafeedback@gmail.com

Learn more at: www.www.aodaalliance.org

TEXT OF ELECTIONS ONTARIO’S MARCH 11, 2013 LETTER TO THE AODA ALLIANCE

Loren Wells
Deputy Chief Electoral Officer
Elections Ontario
51 Rolark Drive
Toronto, Ontario
M4G 3E8
416-326-6300

March 11, 2013

Mr. David Lepofsky, CM, O.Ont.
Chair, AODA Alliance
1929 Bayview Avenue
Toronto, Ontario M4G 3E8

Dear Mr. Lepofsky,

The Chief Electoral Officer, Greg Essensa, has asked me to respond to questions you asked at our meeting of November 19, 2012 and in your letter addressed to him, dated December 7, 2012.
First, I would like to thank you for meeting with us as part of our consultation on alternative voting technology. We appreciated receiving your feedback at the meeting and the additional comments and questions you provided on the subject in your subsequent letter. We have reviewed your feedback in tandem with feedback we have received from other stakeholders.
As you are aware, the Chief Electoral Officer will be submitting a report on his review of alternative voting technology to the Legislative Assembly by June 30, 2013. The report will include themes that have emerged from the consultation process. After our report on the alternative voting technology review is tabled, we would be pleased to have the opportunity to meet with you again.
As you requested, we enclose a Word version of the power point presentation we gave on November 19th. With respect to your request for our concurrence that the AODA Alliance should post the Business Case on your website, I note that we plan to include the Business Case as an appendix to our final report in 2013. Once the report is tabled, the Business Case will be available online as well as in the Legislative Assembly’s library. As such, we prefer to continue following our current approach of providing copies upon request.
With respect to your comments about the accessibility of our website and documents available there, we are exploring your recommendations for addressing website accessibility issues. Thank you for your advice that Word documents are more accessible because they can be opened using a variety of electronic devices. We will continue to endeavour to post documents in Word instead of, or in addition to, PDFs.
You asked for some additional information about Assistive Voting Technology. I can tell you that the actual cost of using Assistive Voting Technology during the 2011 General Election was $1,052,140. This cost includes the equipment, ballots, delivery, training and support.
In an email to you just prior to the provincial election in 2011, responding to your request for this information, we advised you that the estimated cost was $875,000. The actual cost is higher than the estimated cost because of elements of the process, many of which allowed us to provide more service, such as an increase in training locations, ballots, four extra pieces of equipment, onsite user acceptance testing, and other elements. For example, in the Electoral District of Algoma-Manitoulin, Assistive Voting Technology was available in four offices across that district.
You inquired about the differences in actual costs and costs that were discussed during Bill 231 debates. I understand that at one point during the debate a figure of $10,000 per machine was indicated. I am not aware of the source of that information.
You also asked for a clear statement regarding problems with the Assistive Voting Technology that were reported to Elections Ontario about the use of the equipment.
As background, Elections Ontario invites feedback from all electors. During the 2011 General Election, Elections Ontario received more than 20,000 customer service feedback forms and a few hundred emails or phone calls from among the nine million electors who were eligible to vote. Of these, approximately 700 electors provided feedback specific to accessible customer service.
While a relatively small amount of feedback was provided about access by persons with disabilities to election processes, Elections Ontario takes this feedback very seriously. Where negative feedback was received, Elections Ontario typically used the information to fix problems that could affect subsequent voters or is making use of the information to review policies and procedures.
Of the approximately 700 pieces of feedback on accessibility issues, 42 people provided feedback pertaining to the Assistive Voting Technology and related ballot tabulators. Of these, 22 provided positive comments and 20 provided advice or negative feedback.
The Assistive Voting Technology was used by 161 voters, and the related ballot tabulators counted these ballots as well as all others cast in returning offices during the advance voting period and the final five days of special ballot voting before election day.
With respect to the 22 positive pieces of feedback we received, electors liked the electronic system. Some commented on the informative and courteous service they received from election staff and some thought the Assistive Voting Technology and tabulators were a step forward. They found the process quick and smooth. One person indicated they only voted because Assistive Voting Technology was available.
The following is a summary of the 20 pieces of advice or negative feedback we received regarding Assistive Voting Technology and related tabulators:
•           Five voters appreciated the availability of Assistive Voting Technology but would like to see it made available during area advance polls or on election day.
•           Three electors switched to a paper ballot voting method because of technical issues, issues related to customer service delivery, or the elector’s time constraints.
•           Another voter recommended that we retain pencil and paper voting because of unidentified computer problems he or she experienced.
•           Four voters suggested improvements to the Assistive Voting Technology equipment, based on their experiences using the equipment. This includes recommendations to explain instructions at the beginning while avoiding lengthy instructions, raise the pre-set volume level and increase the pre-set speed, and provide key pads instead of paddles.
•           One voter had a poor voting experience because of the placement of the Assistive Voting Technology in the returning office and noise related to other voting activities at the same location.
•           Six voters had concerns related to the tabulators used to count all of the ballots that were cast at returning offices when Assistive Voting Technology was in use. For instance, one voter recommended that voters be allowed to insert their completed ballots into the tabulator instead of this task being done by the Deputy Returning Officer. It is not known whether any of the six voters used the Assistive Voting Technology to vote or whether they marked their ballots by hand.
Election staff was usually able to quickly remedy problems, sometimes by receiving support through our office.
Where comments were received about customer service or instruction-related issues with regard to the Assistive Voting Technology, Elections Ontario provided detailed directions to Returning Officers and staff supervising the equipment. This included directions to advise Assistive Voting Technology users about the overall voting process and outline the required steps to receive audio verification of their choice of candidate after the ballot was printed.
We will continue to consider ways of improving the use of Assistive Voting Technology as we plan for future elections.
In closing, I would like to thank you for meeting with us on November 19th to discuss alternative voting technologies. We look forward to being in touch with you after the Chief Electoral Officer tables his report later this year.
Yours truly,

Loren A. Wells
Deputy Chief Electoral Officer

Attachment

cc: Hon. Kathleen O. Wynne, Premier
kwynne.mpp@liberal.ola.org
Hon. John Milloy, Minister, Community & Social Services
john.milloy@ontario.ca
Dalton McGuinty, MPP
mcguinty.mpp.co@liberal.ola.org
Marguerite Rappolt, Deputy Minister, Community & Social Services
marg.rappolt@ontario.ca
Ellen Waxman, Assistant Deputy Minister, Accessibility Directorate
Ellen.Waxman@ontario.ca
Tim Hudak, Leader, Progressive Conservative Party of Ontario
tim.hudakco@pc.ola.org
Andrea Horwath, Leader, New Democratic Party of Ontario
ahorwath-qp@ndp.on.ca

*****

TEXT OF ELECTIONS ONTARIO’S NOVEMBER 19, 2012 POWERPOINT PRESENTATION ON TELEPHONE AND INTERNET VOTING

Slide 1
Elections Ontario
Alternative Voting Technology Consultation
Presentation to: AODA Alliance
November 19, 2012

Slide 2
Background
Election Act, Section 44.3
The Chief Electoral Officer shall conduct a review of alternative voting technologies, prepare a report of the review and, on or before June 30, 2013, submit the report to the Speaker of the Assembly.

Notes:
In line with the Chief Electoral Officer’s commitment to modernizing the voting process, Elections Ontario has done extensive analysis and research on alternative voting technology solutions such as online and telephone voting. There have been a number of changes over the past few years, and Elections Ontario must constantly negotiate the balance of integrity and accessibility in every decision we make.

With any changes to the electoral process, we need to be able to balance the responsibility of delivering accessible elections to all Ontarians with a more flexible administrative model that ensures the integrity of the process. Voting in Ontario remains to be mostly supervised and on paper. Introducing an alternative voting stream would mark a significant change in the process, and it’s an ongoing challenge to improve access to the voting process, while underlining integrity.

In addition, it’s important to remember that this new stream would not be replacing any of our existing voting options. This would be an additional channel; voters could still continue to cast a paper ballot in person, using AVT, or remotely via a special ballot. The complexity of implementing an additional steam in parallel with our existing processes should not be underestimated.

Slide 3
Purpose of Consultation
Elections Ontario is seeking feedback on the network voting principles and research conclusions, which we will describe in detail later in the presentation, in order to help identify any remaining gaps.

This feedback will be used for further evaluation.

Themes arising from the consultation will be included in final report to be submitted to the Speaker of the Assembly

Notes:

Just to make sure we are all on the same page: Network voting is the term Elections Ontario uses for a means of both casting and counting votes electronically, based on the transmission of ballots and votes via telephones, private computer networks, or the Internet. Network voting is a subset of a broader range of alternative voting technologies.

As has been said, Elections Ontario is neutral about the findings of the review – we have not predetermined the results, and are open and receptive to feedback. We will work to incorporate the feedback received into our report and future approach, where possible.

The information we’re sharing with you here is gleaned primarily from the Research Summary and Business Case, as a way to ensure we’re all on the same page before diving into the discussion and comments on our principles and research conclusions.

Slide 4
Research
Initiated research in fall of 2010
Internal research / inter-jurisdictional consultation
Contracted external consultants with a high level of technical expertise
Developed principles
Ontario-focused research and risk analysis
Developed broad-based implementation options
Developed research conclusions
Determined that concluding the research conducted via the procurement process was not feasible prior to the report submission deadline.
Public consultation process underway

Notes:
Beginning in fall of 2010, Elections Ontario sought out and reviewed extensively the best-practices in network voting being implemented worldwide.
Elections Ontario conducted internal research and consulted with jurisdictions that have utilized or studied network voting methods. We held a summit where we brought together people from different jurisdictions, many from across Ontario, who have run elections using online or telephone voting.
We contracted with external consultants with exceptional technical expertise to conduct research and analysis regarding the suitability of network voting technologies for Ontario.
Over the course of several months and comprehensive research and analysis, they helped the organization to develop a set of electoral principles to be applied against all proposed alternative voting solutions.
Then, they developed a broad range of networked voting options to evaluate – from internet voting, to voting via text message, to biometric authentication.
From the above, we developed our current research conclusions. This research and analysis was the basis of our Network Voting Business Case.
The Business Case was the source for a second procurement process we undertook to determine if there was a vendor who could help us develop an end-to-end solution that would match our principles and fit with our election processes to be piloted in a by-election.
Throughout the research and procurement processes, there were several critical pre-determined points in the project at which Elections Ontario would make a decision whether to proceed with a pilot. “Off ramp” points were based on substantial risk assessment, depending on technical, cost, testing, or other assessments or reasons, including assessments of the specific electoral district and its geography where a by-election would be held.
In the spring of 2012, taking into consideration the key electoral principles and the possible off-ramp points, the Chief Electoral Officer determined that a pilot was not feasible in 2012.
The procurement process was a key learning experience and the lessons learned were invaluable.
It was decided to cancel the procurement at that time, in order to do further consultation and research before moving forward.
Our research shows that, above all, we need to move forward in measured steps, and this additional consultation is part of that effort.

Slide 5
Principles
Accessibility
One vote per voter
Voter authentication and authorization
Only count votes from valid voters
Individual verifiability
Voter privacy
Results validation
Service availability

Notes:
After carefully researching the best examples worldwide of network voting principles, our consultants helped the organization, over the course of several months, to arrive at these eight principles. These were chosen with the context of Ontario and Ontarian voters in mind. They reflect what we believe to be the most critical factors in the successful implementation of a network voting solution, and provide the best possible balance of access and integrity.

We want to emphasize that “security is the watchword.” We must have a secure system; these principles will ensure we protect the security and integrity of the electoral process, and provide us with a framework for measurable outcomes.

We define the principles as such:
Accessibility: The voting process is equally accessible to all eligible voters, including voters with disabilities. In any case, the voting process will be performed by the voter without requiring any assistance for making their selections. We chose not to restrict the process to any individual stakeholder group.

One vote per voter: Only one vote per voter is counted for obtaining the election results. This will be fulfilled even in the case where the voter is allowed to cast their vote on multiple occasions (in some systems, people can cast their vote multiple times, with only the last one being counted).

Voter authentication and authorization: The electoral process will ensure that before allowing a voter to cast a vote, that the identity of the voter is the same as claimed, and that the elector is eligible to vote.

Only count votes from valid voters: The electoral process shall ensure that the votes used in the counting process are the ones cast by valid eligible voters.

Individual verifiability: The voting process will provide means for the voter to verify that their vote has been properly deposited inside the virtual ballot box.

Voter privacy: The voting process will prevent at any stage of the election the ability to connect a voter and the ballots cast by the voter.

Results validation: The voting process will provide means for verifying if the results clearly represent the intention of the voters that participated in the voting process.

Service availability: The election process and any of its critical components (e.g., voter’s list information, cast votes, voting channel, etc.) will be available during the whole election period to voters, election managers, observers or any other actor involved in the process.

Slide 6
Extensive Broad-Based Research
As a result of preliminary research, we started with:
7 network voting mechanisms (i.e., internet voting, voting via SMS, digital television voting)
6 means of voter authentication (i.e., biometrics, passwords, identification cards)
This produced 10 feasible network voting options, which were measured against Elections Ontario’s principles, and evaluated for cost, complexity and convenience factors. This created a shortlist of four options.

Notes:
Preliminary research identified seven basic network voting mechanisms which have been used internationally for election delivery, such as on-site or remote online voting or remote mobile phone voting through SMS. Research also identified six means of voter authentication, such as passwords, biometrics, and physical identification. Some of these were completely out of scope at this time due to their associated complexity and cost, such as biometric authentication. Every option was evaluated thoroughly against their ability to uphold our identified principles.

The intersection of these options produced ten feasible network voting scenarios. These combined scenarios were again assessed for their ability to support our electoral principles, and were also evaluated against an assessment of their cost, complexity, and convenience for electors.

Slide 7
Short List of Four Scenarios
1. Onsite online voting with supervised authentication,
2. Onsite telephone voting with supervised authentication,
3. Remote online voting based on password authentication, and
4. Remote telephone voting based on password authentication.

Notes:
From this really broad research, we were able to arrive at a short list of four scenarios qualified as suitable for a possible by-election pilot.

Slide 8
Research Conclusions
Recommended Approach:
remote online voting based on password authentication, and remote telephone voting based on password authentication.

There are three key parts of the election process impacted by this approach:
Registration & Authentication
Online Voting Process / Telephone Voting Process
Tabulation Process

Notes:
From the four options presented, it was narrowed down to two: (1) remote online voting based on password authentication, and (2) remote telephone voting based on password authentication.

The two options were chosen due to their ability to uphold the principles previously outlined. In addition, they are reflective of the capabilities allowed by Elections Ontario’s current technological solutions and authentication options. Finally, they also reflect the network voting technologies that were available at the time of the business case development.

There are three key parts of the election process impacted by this approach that we will now review:
Registration & Authentication
Online Voting Process / Telephone Voting Process
Tabulation Process or counting process

Slide 9
Registration and Authentication For Online and Telephone Voting
Flow Chart
Online or Telephone voting scenarios
Online or Telephone, Voter Receives ID
Voter registers online or, by telephone, Voter registers by IVR
For both online and telephone; Voter is validated and sets up voting password in NVS
Online voter signs in to voting web site
Telephone voter signs in to voting IVR

Notes:
The registration and authentication for both online and telephone voting is very similar, using different technologies. All electors would receive a network voting registration letter that includes a secure numeric Elector ID and instructions for accessing a remote network voting registration web site and a telephone number to call.

Electors who choose to register for online voting would visit the web site and enter their Elector ID and their date of birth to register. A second card would be mailed at this stage to provide the voter a secure second PIN before proceeding to the next step.

Once authenticated, the system would validate their eligibility and allow them to set up a secure password to use for voting.

Alternatively, electors who do not have easy online access could then call a toll-free number to perform the same steps. This would use an interactive voice response (IVR) system interface that connects to the same backend system.

Once the advance poll period begins, voters who had registered for remote voting could log in to either the voting web site or the voting IVR system using their Elector ID and password.

Slide 10
Online Voting Process
Online voter casts ballot on computer
Ballot encrypted with public key
Encrypted ballot is digitally signed
Ballot is sent to digital ballot box on servers

Notes:
Once a remote voter has been authenticated on the voting web site, he or she would cast a ballot by making a selection from an online screen. Voters who use the telephone would make their selections using an automated menu system. Both options would need to be optimized for usability and accessibility in order to provide the best user experience.

Slide 11
Telephone Voting Process
Voter dials a toll free number
Selects preferred language
Types predetermined password via keypad
Access gained to audio ballot
Completion and submission of ballot via keypad

Notes:
With the telephone voting option, electors could vote from any location, provided that they have access to a telephone, which can be a conventional phone, a mobile phone, or voice over IP (VOIP) from a computer or a telephone. Voters would dial a toll free number, select their preferred language, and then type a previously created password using the keypad. The system itself would authenticate the voter and determines eligibility to cast a ballot. If the authentication is approved, the voter would gain access to an audio ballot, which they would cast via audio instructions.

Slide 12
Back End Processes
Vote cast
Voter struck from the voter’s list
Voter receives a receipt
May then verify the inclusion of their ballot
Vote is stored in a secure server environment
This environment would be subject to stringent physical and application security measures

Notes:
After voting on one of these channels, the voter would be struck from the voter’s list and receive a receipt that would allow them to verify the inclusion of their ballot in the final election results.

After a ballot has been cast on either the telephone or online channels, it will be stored in a secure server environment that is subject to stringent physical and application security measures, as well as availability and performance requirements.

Slide 13
Tabulation Process
Polls Close
Digital Ballot Box is opened
Does each encrypted vote match a voter?
Decrypt ballots
Separate digital signatures from votes
Count unencrypted ballots
Publish results to Returning Officer

Notes:
The tabulation system, which would be unlocked by security keys held by no less than two Elections Ontario officials, will securely decrypt and tabulate valid ballots. It will then distribute the combined network voting results to the Returning Officer, who will include them in the official vote count.

Slide 14
Risk Identification and Mitigation
Risks were measured according to their complexity, impact, and residual risk, meaning the risk remaining after mitigation strategies.

There were three subsections of the risk analysis:
Security Risk Assessment
Operational Risk Assessment
Voter Risk Assessment

Notes:
Risk identification and mitigation played a key part of our review and comprises a substantial portion of the Business Case: over 1/3 of the Business Case outlines the risks and describes the planned mitigation strategies. It is an example of rigorous, thorough research and analysis.

While the approach I have outlined was based on our research because of its ability to support accessibility, integrity and security, there are risks. A network voting model based on a combination of online and telephone voting is potentially vulnerable to several types of risk, with the Security Risk Assessment having identified the highest number of threats.

Slide 15
Necessary Challenges in Current Circumstances
Available only to those on the preliminary list of electors
Two mail-outs
Offered only in the advance polls
Integration of new channels in parallel with all existing streams

Notes:
Our research conclusions have some challenges, but ones that are necessary for upholding the integrity of the process. Again, these challenges are as a result of the need for balancing accessibility and integrity. Changes to current circumstances – technological, legislative, or other changes – could affect these challenges in the future.

1. That the network voting option would be available only to those on the preliminary list of electors – this is unavoidable given our current processes and available technology, in order to ensure that the list and strike-off are managed in such a way that prevents electors from casting multiple ballots.
2. That there are two mail-outs – it was determined that the risk associated with only one-mailout was too high for the Chief Electoral Officer to accept. The two-step mailout process offered an acceptable, low level of risk that upheld the principles.
3. That network voting would only be offered during the advance polls – in the risk analysis, it was determined that the risk of a failure of the system on Election Day was far too high for the Chief Electoral Officer to accept.
4. In addition, it’s important to remember that this new stream would not be replacing any of our existing voting options – voters could still continue to cast a paper ballot in person, or remotely via a special ballot. We have to look at the complexity of adding another steam in parallel with our existing processes.

Slide 16
Other Considerations
For example, although we addressed the issue of the pilot in the presentation, you may have additional queries as to why we chose not to pilot at this time. Section 4.1 of the Election Act gave us the opportunity to pilot, but does not require that Elections Ontario pilot before we submit our report on alternative voting technologies. It is imperative that any solution we choose to pilot upholds the integrity of the vote, meets all requirements outlined by our principles, and is within the organization’s capacity to deliver. As you have seen, the process is extremely complex, and it is critical that all of our identified principles be upheld.

We are also consulting with the Ontario Human Rights Commission and the Accessibility Directorate of Ontario to get their feedback on our principles and research conclusions.

I note that the Integrated Accessibility Standards were not published when the Business Case was in research and development. Elections Ontario would ensure that any implementation of Network Voting was in compliance with the standards, both in the web architecture and in the physical processes.

Notes:
Some other considerations, for example, although we addressed the issue of the pilot in the presentation, you may have additional queries as to why we chose not to pilot at this time. Section 4.1 of the Election Act gave us the opportunity to pilot, but does not require that Elections Ontario pilot before we submit our report on alternative voting technologies. It is imperative that any solution we choose to pilot upholds the integrity of the vote, meets all requirements outlined by our principles, and is within the organization’s capacity to deliver. As you have seen, the process is extremely complex, and it is critical that all of our identified principles be upheld.

We are also consulting with the Ontario Human Rights Commission and the Accessibility Directorate of Ontario to get their feedback on our principles and research conclusions.

Slide 17
Discussion
Has Elections Ontario identified the most appropriate principles?
How should they be ranked, if at all?
Do the research conclusions successfully reflect the principles?
Should we consider narrowing eligibility if that would allow us to reduce some of the accessibility challenges?
What are your thoughts on an identity card for all Ontario electors?
Have you identified any gaps in our research or analysis?
Other comments regarding Network Voting?

Notes:
Which brings us to our discussion.
Has Elections Ontario identified the most appropriate principles?
How should they be ranked, if at all?
Have we accurately identified the categories of risk? What are your thoughts on their prioritization?
Do the research conclusions successfully reflect the principles?
Since we can’t require a driver’s license, because not all electors have one, and the Ontario Photo Card has a cost associated with it, what would be the implications of proposing an identity card for all Ontario electors?
Have you identified any gaps in our research or analysis?
Are there any other comments regarding Network Voting?

Slide 18
Principles
Accessibility
One vote per voter
Voter authentication and authorization
Only count votes from valid voters
Individual verifiability
Voter privacy
Results validation
Service availability

Notes:
Accessibility: The voting process is equally accessible to all eligible voters, including voters with disabilities. In any case, the voting process will be performed by the voter without requiring any assistance for making their selections.

One vote per voter: Only one vote per voter is counted for obtaining the election results. This will be fulfilled even in the case the voter is allowed to cast multiple votes (in some systems, people can cast their vote multiple times, with only the last one being counted).

Voter authentication and authorization: The electoral process will ensure that before allowing a voter to cast a vote, that the identity of the voter is the same as claimed, that the elector is eligible to vote, and that he or she has not exceeded the allowed voting intents.

Only count votes from valid voters: The electoral process shall ensure that the votes used in the counting process are the ones cast by valid eligible voters.

Individual verifiability: The voting process will provide means to the voters for verifying that their votes have been properly deposited inside the ballot box.

Voter privacy: The voting process will prevent at any stage of the election the correlation between voters and the contents of the ballots cast by such voters.

Results validation: The voting process will provide means for verifying if the results clearly represent the intention of the voters that participated in the voting process.