Save Settings

Position paper on accessible electronic voting

contents:

Background

When the government announced in mid 2001 that a system to provide electronic voting was being planned I was immediately interested.
I realised that since an electronic system would be used to display the ballot paper and to allow the voter to cast their vote that this could open the possibility of visually impaired people being able to vote independently.
For many years software has been available to convert electronic data into either speech, large print or braille. I imagined that it would be a relatively straight forward process to incorporate this technology into the voting machines and thus allow visually impaired people to use the voting machines.
I contacted the department of environment, who were responsible for the investigation into the new system, and enquired if the new system would be fully usable by the visually impaired.
On the 9th of August 2001 I received the following reply from the department:

"Further to our telephone conversation of 9/8/01 in connection with the above, the Department of the Environment and Local Government has appointed a Dutch/UK company to deliver an electronic solution for the current manual voting and vote counting arrangements. Testing of the voting machine and system software is underway and should be completed early in 2002.

As part of the project we will be examining the possibility of developing a facility to allow people with a visual impairment to use the voting machine independently. This facility will likely involve voice generated feedback to a blind voter indicating to them their candidate choices, preferences etc."

Given this response I was quite optimistic that an accessible voting system would be forthcoming. However, in the ensuing months it emerged that no voice module would be developed until all trials of the new system had been completed. The dept of Environment cited the large cost in developing an accessible system which could not be justified since no final decision had yet been made to introduce the electronic voting system currently being tested. This was understandable if a little disappointing, so I contented myself to wait and see!

The current situation

In the current manual system of casting your vote I believe a visually impaired person has three options if they wish to partake in an election:

  1. They can apply to be placed on the postal register and thus vote by proxy. The advantage of this is that the visually impaired person does not have to locate a polling station. The disadvantages are that the ballot paper must be completed by a sighted person. The vote is then cast by proxy and possibility for error or, god forbid deliberate spoiling or tampering of the vote is introduced.
  2. The visually impaired individual can be assisted by either a family member or the presiding officer at a polling station to cast their vote. The disadvantage of this approach is similar to point 2 above in that the disabled person has no way of authenticating that their choice has actually been implemented. Furthermore, the secrecy of the ballot has been compromised.
  3. Finally, the person may decide not to vote at all. Anecdotal evidence suggests that choice is made by a large number of visually impaired people thus disenfranchising a proportion of the population.

Reasons why electronic voting is desirable for the visually impaired

If a fully accessible electronic voting system is in place then many of the disadvantages listed in the previous section could be removed. A person would be able to read or hear the list of candidates or options to be decided upon. they could then select their choices and cast their vote. Most importantly they could do all of this without the assistance of any third party. As a result the secrecy of the ballot is maintained and the visually impaired person can feel that their dignity as a citizen is intact.

Reasons why electronic voting should be fully accessible

Leaving aside that an accessible voting system is technically feasible and should be introduced as a matter of course, I believe that the government is obliged to do this by the laws of the land.

Firstly the irish constitution states in article 16 section 1.2:

i. All citizens, and
ii. such other persons in the State as may be determined by law, without distinction of sex who have reached the age of eighteen years who are not disqualified by law and comply with the provisions of the law relating to the election of members of Dáil Éireann, shall have the right to vote at an election for members of Dáil Éireann.

and more importantly article 16, section 1.4 states:

No voter may exercise more than one vote at an election for Dáil Éireann, and the voting shall be by secret ballot.

I think that this is pretty clear. Up to now it has not been possible to provide a system were visually impaired people could cast their vote in secret. However, the technology is now available to do this and thus should be introduced as quickly as possible.

Secondly, the Equal Status Act,2000 outlaw discrimination in employment, vocational training, advertising, collective agreements, the provision of goods and services and other opportunities to which the public generally have access on nine distinct grounds. These are:

I believe that the running of an election comes under the heading of a provision of goods and services. At present disabled people are descriminated against since they cannot vote independently.

Finally, the national Disability Authority has set out a list of accessibility guidelines for various areas related to information technology. One of these areas is the accessibility of public access terminals. Quoting from the NDA web site:

Public access terminals include (but are not limited to):

I would contend that electronic voting machines would come under the heading of a public access terminal.

Two of the guidelines for making these terminals accessible are:

1.7 Ensure that users with restricted or no vision can use all functions of the terminal
and 1.8 Ensure that all outputs can be perceived by users with restricted or no vision

Therefore the government is compelled by the guidelines of a body that it set up to ensure that the voting machines can be used by everybody.

FurtherMore, The Information society Commission states that the Government is committed to social inclusion. In the report entitled Building the Knowledge society, the ISC recommends that the NDA should carry out research to see how well their IT Access guidelines are being implemented. We can see electronic voting as an obvious example where this is not the case. If the Government is really committed to social inclusion in the Information Society then inaccessible electronic voting should never have happened.

Conclusion and Recommendations

After having my initial excitement about the opportunities that electronic voting could provide for the visually impaired dampened somewhat by the department of environment I remained in contact with them. The most recent correspondence I received from them was on the 26th of February 2004 and I quote it below:

The Department is examining the development of a facility for the future whereby visually impaired voters, who would otherwise require assistance to vote, will be able to use the new voting system independently. We will be consulting with the National Disability Authority and the National Council for the Blind, who have already provided constructive input into the improvements made to the voting machine, in realtion to the development of this voting machine feature."

This response sounds uncannily similar to the very first reply I received from the department over 2 and a half years ago. It reflects the slow rate of progress being made in providing accessible voting.
Having used the electronic voting system currently in place during the general elections in 2002 I can safely say that it is completely inaccessible. Currently the voter makes their selection by way of a touch screen. The only button that can be located by touch is the 'cast vote' button.
A blind person is no better off now than they were under the old system. Indeed, partially sighted people, in some cases, are worse off since the screens are more difficult to read as compared to the print on a paper ballot.

Given all this the following changes are imperative so as to allow visually impaired people to vote independently:

  1. The input device should be accessible to the totally blind and partially sighted. By this I mean that all buttons should be embossed or recessed so that they can be located by touch. Any labels or instructions on the device should be legible to partially sighted individuals.
  2. All outputs reflected via the visual display should be presented in a form that visually impaired people can use. This would mean that any text displayed on the screen is easy to read by the partially sighted. The size and contrast of text should be customisable so that individuals can change it to suit their needs. Blind voters should be able to read what is on screen and any selections they have made via an audible output device.
    Any error messages, help text and any other information that the fully sighted voter can obtain should be available to the visually impaired.
  3. Any audible output should be by way of headphones thus ensuring that the choices made by visually impaired voters are not heard by other people.
  4. Prototypes of any accessible system should be tested thoroughly by the end users of varying technical abilities.
    A full consultation process should be undertaken to ensure that the final system is truly usable by all.

To sum up:

Electronic voting provides a means, for the first time, in which disabled people can exercise their democratic right to vote in secret. This opportunity should be exploited as soon as is practicable.

The government is obliged by the constitution of Ireland, and by its own guidelines to ensure that all its services can be used by the entire population. If the same amount of money that has been spent so far on the electronic voting project had been used to build a public building, and that building had no ramps for wheelchair users, then the outcry would be considerable. The same principal holds true for electronic voting, and indeed is even more important given the nature of the issue.

Developing an accessible electronic voting system isn't a technical issue but rather a political issue. Technical problems can be solved. I am also convinced that this access problem could be solved very cheaply if done so from the design stage. The assumption made by the Department of the Environment that it would be very expensive is I believe and incorrect assumption. Of all people, the Department of the Environment should know that access is best considered at the design stage and not as a bolt-on fix afterwards. If accessibility proves expensive in this case, I think this is the fault of the Department for ignoring all the points made so far in this paper.

At this point there is no possibility that an accessible voting system will be in place for the local and european elections to be held in June 2004. However, a fully accessible system should be in place in time for the next general election which is expected to be in 2006.

In order to allow the visually impaired community, and the disabled community in general, to keep a watching brief on the progress being made towards a fully accessible electronic voting system I think receiving clear answers to the below series of questions would prove most useful. Therefore I would strongly urge that the below questions are submitted to the Commission on Electronic Voting:

  1. What is the design brief for the developers of the accessible electronic voting system.
  2. What are the terms of reference which the developers have been given in relation to accessibility?
  3. How much is being budgeted for it?
  4. What material has already been collected through consultation?
  5. How will the consultation be carried out (private individuals, small invited panel meetings, questionnaires, public listening meetings, etc.)?
  6. What test criteria have been set out for the proposed system?
  7. How is "accessibility" being defined?
  8. What level of accessibility will be deemed sufficient (All users, most users or some users)?
  9. What is the name of the system being considered?
  10. Why was this system chosen?
  11. What alternatives have been examined?
  12. Why were alternative systems rejected?
  13. Why is the accessible version not included in the main system?
  14. What form will the terminal take after accessibility has been
    included?
  15. What is the expected delivery date?