In the past, it appears that governments worldwide and their respective electorates have been quite concerned with any changes implemented in their voting systems. Unfortunately, this is not the case with the switch from paper ballot voting to electronic voting systems [8]. In the rush to appear technologically advanced, inadequate voting systems are being installed and used. Significant errors and failures in voting systems since Florida 2002 have been noted by Dr. Rebecca Mercuri [9], one of the leading thinkers in the field. And yet the Irish government have committed to introducing the Nedap/Powervote system for the whole country at the next general election [10]. This Dutch/British system appears to have been chosen without consulting Irish experts in the field of computer security. Claims by said experts that the system may be insecure are often dismissed as paranoia [11] by both politicians and the public, despite evidence that elected officials do not deserve our unmediated trust.
It is true that an EVS could offer several advantages over manual systems. It could tabulate the results more quickly, eliminate the human error which sometimes occurs in manual vote tabulation, expand the franchise to those currently unable to vote because of special needs, and improve the fairness of the Irish count system.
From a traditional standpoint, however, it is questionable whether or not the Irish electorate needs a faster count. Many active party members have expressed their disappointment that they will no longer be able to enjoy the excitement of a long count. The anticipation involved in waiting for the result is an integral part of our political culture [11].
If the system is going to eliminate human error from counting, it must be developed formally. Human error in the design or implementation of the count software might have considerably more serious, and more far-reaching, effects than human error in the manual system. Formal development is discussed further in the final year project report accompanying this document [12].
Many people with special needs can currently only vote with the aid of an election official, which means that they do not have the same privacy in voting as other citizens. The potential exists to create a system accessible to people with visual, auditory, and movement disabilities.
The form of Proportional Representation (Proportional Representation - Single Transferrable Vote - PR-STV) used in Ireland gives a very accurate representation of the will of the people [13]. Unfortunately, this is at the cost of a very complicated count procedure. One part of the count procedure involves a compromise between efficiency and fairness. Since it is difficult to determine manually which votes should be transferred, votes are taken from the top of the pile - effectively randomly. This does not mean that voting at a particular time makes your vote more likely to be transferred - votes are mixed thoroughly at the beginning of the count process - but it does mean that the late preferences on some votes are given less significance than others. An electronic system could fairly determine which votes should be transferred, removing the need for such a compromise. It is worth noting that this adjustment could only be made to the counting procedures if we make it across the whole system. Which means it could only realistically be introduced once the electronic system was the main medium for vote-casting in the country.
It is a commonly held belief that the change from hand-counted paper ballots to an EVS is inevitable. In that case we must examine carefully what we expect from such a system, and how it should be developed. We must use all the tools at our disposal to ensure that the introduction of an EVS does not put the integrity of our democracy at risk.