Internet Election Systems are not a new subject in principle. For decades work has been done studying the possibilities in electronical systems and methods. Until now, no system has been used in actual elections on a scale that matters. Main reasons for this are mainly to be found in the fact that it is not so simple to fulfil all the requirements that a government election requires in order to be trustworthy and secure and admissible for a large percentage of the population. So far, systems that could live up to the requirements, simply required systems so complicated that the vast majority of the people at home still can't participate with the unmodified equipment in their own home or office location. Since not all voters can be forced to vote through internet, alternative election techniques should be offered. This implies an even greater complication: what to do with the combination of votes and how to ensure a flawless system? In the last quarter of 2004, 2.2 million voters of the District Water Boards Rijnland and De Dommel in The Netherlands have been able to go to the polls for the election of the board of representatives. All voters could cast their vote either by Internet or by regular postal mail. These public water board elections have been the largest formal Internet elections on the planet so far. In addition to the casting of votes through the Internet, all voters had the opportunity to verify the counting of their votes in the final outcome and to validate the election results. In spite of its sophisticated design, the Internet election system RIES did allow more than 99% of all voters to use the system without any change to their existing PC and Internet environment. 35 % of the voters cast their vote by internet. Only a very limited number of complaints have been issued, so one can safely say the system worked flawlessly. In both elections the RIES system lived up to all expectations. The system performed with a negligible number of complaints by voters and has been flawless in processing all postal mail- and Internet votes. Solely based on the published working principles and election files, independent parties have been able to fully validate the results of the (Internet) elections. In 2006 RIES is used by the Ministry of Interior for the registered Dutch voters living abroad for the Dutch parliamentary elections. Main purpose of this experiment was to enhance voting facilities. The results: 33 % of the voters cast their vote by mail and 67 % of the voters cast their vote by internet. In 2008 a combined election of all 26 national Water Boards will be issued with RIES. This will involve 12 million voters. In June 2006 RIES won the 2006 United Nations Public Service Award in the category of "Improving transparency, accountability and responsiveness in the Public Service".
Main results, benefits and impacts
Main purpose and goal of RIES was simplifying the ability to vote on one hand, and gaining more familiarity with Water Board elections on the other hand. Since for Water Board elections the turnout is traditionally low, the elections need to be organised within tight budget limits. One of the goals of the RIES project was to see whether cost reduction would be possible, both in back office and in actual election, combined with a higher familiarity and comfort for the voter. This has been very successful. Specially when taking into account the budgets spent on other internet election systems, reaching a far smaller group of voters. For a budget of a mere 400.000,- euros, a complete, flawless working system has been built, fit for use in any formal government election for any number of voters within any one country. This, set alongside the former attempts, where budgets of millions of euros were spent on elections reaching limited amount of voters, with far less sophisticated systems in the end, one can safely say RIES IS simple cheap and pragmatic. RIES has been built and designed completely from the users point of view. Great effort has been put in studying questions like: how people vote, why they vote, what information they need, what they find most important, what systems they use. All technical problems evolving from these answers are solved in a pragmatic, cost conscious way. A system has been built living up to all the requirements of any formal government election, and leading to a cost reduction in 2004 of 85.000,-, which implies 25% of the normal costs normally necessary for registration, validation, collection and tallying of vote. In 2008 the system is used for the national Water Board elections, enabling all 12 million eligible voters to use it. The expected cost reduction will be 750.000,- In this first trial of RIES, 35% of the voters used internet. The other 65% voted by mail. Both these results were processed by the same system, in the same back office. Despite its simplicity, RIES is based on a sophisticated scheme, in which no voter secrets leave the voter's PC browser and are never shared with any election server, nor stored in any form with any party, including the election authorities, during the election stage. RIES allows for the acceptance and processing of multiple voting entries from the same voter in the same election. This creates the possibility for the voter to vote once more by Internet or postal mail in case of uncertainty that the system processed the first attempt correctly, for example in case of a major network disturbance. The technology required for running RIES is simple, straightforward and therefore reliable and low-cost. For the large-scale RIES elections in The Netherlands in 2004 (in which 120,000 Internet voters took part) two sets of simple Intel architecture-based FreeBSD servers, with SSL offloading capabilities, offering an overcapacity of more then 97%.
Return on investment
Return on investment: Not applicable / Not availableLessons learnt
Most important learning point form the RIES project is that it is possible to create a system that lives up to specific requirements. It works It is simple and pragmatic It is beyond experimental phase It costs less than any other election system in case of the system itself and of the organization of the election Voters are highly content with the system Back-office can be simplified to a high extent Although it was hoped at the start that internet voting would also lead to a higher turn out, this is not realized in this pilot. Conclusion here is that the technique of internet elections in itself does not lead to higher election participation. No matter how satisfied the participating voters were while using the system, people that do not participate, still are not triggered. It does lead to a huge shift in how people that vote, do this, but people who don't vote, don't vote by internet. The reasons for participation need to be sought in the value of election itself. Nevertheless, there IS a huge shift in how people vote. 35% used internet, and the results in this are very satisfactory. The success of RIES is mainly to be found in usability, which is the key issue throughout the entire project. Visual handicapped people have great benefits from the system. They have been able to cast their vote without any help. This is a huge benefit for any election, since this group does rate elections as highly important and has a large participation rate. RIES can be used in any formal government election as it is now, with only minor detail changes. It is already fit for use in any comparable election with unlimited voter numbers. This is done with a budget of only ? 400.000,- This great success has been achieved mainly because RIES was managed by a very small and flexible team that was so enthusiastic, that they were able to gain the co-operation of their entire network without any effort. The RIES project team will be more than happy to organize a workshop for other governmental bodies in order for them to learn the implications of elections by internet. Of course, the project team is more than willing to present the case at the ministerial eGovernment Conference and to demonstrate it at the exhibition and participate in a workshop. There is already a version of the RIES called "RIES public". RIES Public is freely accessible to anyone who should require this. Most of RIES is based on open source technology. For a limited amount of time and a limited budget, the designers of the RIES can adapt this to any election so that it works without any problems in the same way the RIES did. Any governmental body in Europe (or outside for that matter) can use RIES for their elections. All they need to do is re-organize their back-office in a way that can deal with this type of elections.
Scope: National