Two recent online voting initiatives have been hailed as successes, but while officials in an Ontario city and a border state are satisfied with Internet voting, a project that would allow Americans to vote digitally has been cancelled. Even though voter turnout increased when the online option was offered, the United States government worries about ballot security. Delvinia Interactive Inc. of Toronto released Internet Voting and Canadian e-Democracy in Practice, the first of two reports analyzing last year’s municipal elections in the town of Markham, outside of Toronto (NL, Nov. 10/03). The second study, due to be issued next month, will look at the effectiveness of broadband technology in communicating election information to voters. The initial report found high-tech improved voting numbers. "In a municipality of more than 158,000 registered voters, over 11,700 citizens registered to vote online for the 2003 municipal election in the Town of Markham," Delvinia president Adam Froman writes. "Of them, 7,210 or 7.5% of the voting population cast their ballot online during the advance polls. This accounted for approximately 17% of the overall voter turnout of 42,198 in Markham and an increase of over 300% in advance poll voting." Froman found that the option of voting online appealed to people who had not cast ballots in the 2000 municipal elections. One-quarter of those surveyed online said they had not voted three years earlier compared with 18% of the in-person respondents. On February 7, the Democratic Party in Michigan held its caucuses to determine delegates for the presidential convention in Boston this summer. More than 46,000 people voted online, a great success according to the party’s executive chair. "We’re pleased not only with the number but with the security and integrity of the Internet voting system," Mark Brewer told a news conference in Lansing MI as he released results. Turnout was so high, he added, that if only the online and mail ballots had been counted, it still would have been the third-highest voter participation in state caucus/primary history. While those two test runs are encouraging for e-democracy advocates, the Pentagon still needs to be persuaded that the Internet is a proper voting tool. Earlier this month, the Defense Department scrapped the Secure Electronic Registration and Voting Experiment (SERVE), saying it could not guarantee the legitimacy of the procedures. SERVE would have allowed American expatriates, mostly military personnel and their families, to vote electronically. But a report released on January 21 raised the question of SERVE’s susceptibility to attack from outside sources. While conceding that traditional elections have not been free of tampering, the study’s four authors say the dangers are greater with online balloting. "When voting is conducted at physical precincts on mechanical devices or with paper ballots, vote manipulation, to the extent it occurs, happens on a far smaller scale: No single attack is likely to affect a large number of votes," the report states. "To make the comparison more explicit, a single teenager, hacker, or other malicious party could potentially affect tens or hundreds of thousands of votes cast through SERVE, while it is extremely unlikely that any single person could conduct vote fraud on such a large scale in existing non-electronic elections." Security concerns did not weigh heavily on the minds of Markham voters who opted to vote in-person last year. When Delvinia asked them why they did not vote online, only 9% said it was because they don’t trust the Internet. One-third of respondents replied that they had missed the registration deadline; another 11% gave a variation of the "wanted to vote in-person; like the social aspect of voting" response. When those who said they were uninterested in voting online next time were asked why, security issues came in a poor third, after computer literacy and in-person preference. But those who choose to shun electronic voting in 2006 are apt to be in the minority. Of the 994 in-person respondents, 42% said they were "very likely" they will vote online next time and a further 27% styled themselves "likely" to do so. Among the 3,655 surveyed online, 100% said they were likely to vote electronically in the future. The primary reason given for choosing the Internet option by respondents was the convenience of voting. Only 1% of online voters reported being unsatisfied with the process. Home was the preferred site from which to vote, with 79% casting their ballots on personal computers. Only 13% used their work computers to vote. Another 3% said they voted from other locations and 5% refused to state. That 13% may have risked their own privacy, the American study asserts. "If one votes at work, the employer controls the computer," the SERVE analysis concludes. "A study found that 62% of major U.S. corporations monitor employees’ Internet connections, and more than one-third store and review files on employees’ computers." Even voting at home could compromise the sanctity of the ballot, the report adds. Many family members may have access to one computer."It is not a security failure if your spouse uses your credit card with your consent; it is routine to delegate the authority to make financial transactions. But it is a security failure if your spouse can vote on your behalf, even with your consent; the right to vote is not transferable, and must not be delegated, sold, traded or given away." Regardless of where the computer is located, voters still run the risk of on-screen electioneering, the SERVE report authors note. While most locales have rules forbidding campaigning near polling stations, no such restrictions exist in cyberspace. "For example, an ISP (or browser company, etc.) may derive revenue from advertising, like AOL, and would thus have the ability to target ads based on the IP address a user is connected to at the moment," the study states. "These could take the form of pop-up ads or even ads within the browser window. The problem is that the moment when a voter connects to the SERVE vote server address (or a voter information site) he/she could be bombarded with all sorts of political ads. It is even possible that at least some forms of ads will end up being protected by the First Amendment, and then there will be no escaping them." Traditional media tended to fill the information gap for Markham voters, Delvinia found. Online voters were most likely to follow the election news through the mail or from community newspapers while those who actually showed up at polling places gleaned their information from radio and television.