When America goes to the polls for the presidential elections on November 2, almost 50 million people - or about 30 per cent of registered voters will cast their votes on electronic voting machines, subjecting these specialised computers to perhaps their toughest real-world test.
When America goes to the polls for the presidential elections on November 2, almost 50 million people - or about 30 per cent of registered voters will cast their votes on electronic voting machines, subjecting these specialised computers to perhaps their toughest real-world test.
In the wake of the cliff-hanging 2000 presidential election, whose result ultimately hung on the interpretation of a few hundred punchcard ballots cast in Florida, many US states have rushed to implement electronic voting systems.
But arguments over the accuracy and security of such systems continue to rage and if the battle between John Kerry and incumbent George W. Bush is anywhere near as close as it was between Bush and Al Gore four years ago, voting systems old and new could be under the spotlight again.
There are two main alternatives to traditional paper-based and mechanical voting systems optical scanners that "read" votes cast on paper ballots and electronic touch-screen voting machines which are also called direct recording electronic (DRE) voting systems.
The market for electronic voting machines is dominated by three companies: Texas-based Diebold Election Systems, Election Systems & Software of Omaha, Nebraska and Sequoia, a subsidiary of Britain's De La Rue.
Most of these newer machines work like Diebold's market leading AccuVote-TS touch-screen voting system. Voters whose names appear on the register at a polling station are given a smartcard to insert into the voting machine basically a PC with a touch-screen.
The voter selects candidates by touching the machine's screen. Manufacturers claim the electronic systems have virtually eliminated the problem of "undervoting" failing to select a candidate and "overvoting" selecting more than one candidate because most of the machines require voters to select one and only one candidate in a race.
Most machines also give voters the chance to confirm their choices on a summary screen and ask if any changes need to be made. Once the voter confirms his or her choices the vote is registered and stored on flash-memory cards inside the machine as well as on an internal back-up system. When polling closes, the memory cards containing the votes are taken to a central computer to be tallied.
While electronic voting systems have been used for a long time in some European countries and were widely used in recent elections in India and Venezuela (where they were the focus of opposition claims of voting fraud), their use in the US has, until now, been limited.
According to the Washington-based Election Data Services research firm, about 13 per cent of US voters used electronic machines in the 2000 presidential elections. The vast majority of the remainder used a hodge-podge mixture of paper ballots, punched cards and mechanical lever machines.
The race to switch to electronic voting systems in the US has been fuelled in part by federal funding. To help avoid a repeat of the Florida voting fiasco, where many voters reportedly ended up casting ballots for the wrong candidate on badly-designed punchcards, Congress passed the Help America Vote Act in 2002.
The bill earmarked about $3 billion for states to upgrade their equipment and many have done just that, investing in new machines that cost as much as $2,000 each, but are expected to pay for themselves over about 15 years by doing away with the need to print up ballots, according to manufacturers.
As a result, at least 32 states and the District of Columbia will deploy direct recording electronic voting systems mostly touch-screen based systems. Among these states, some such as Maryland and Georgia will use electronic voting systems almost exclusively.
In theory these new systems are supposed to record and count votes more accurately. But concerns have been voiced that electronic technologies can just as easily mean that more votes will go missing or be miscounted.
In addition, while all voting systems have been subject to abuse in the past, critics many of them computer science academics and voters' rights groups charge that DRE systems are unreliable, open to tampering, and have not been adequately tested.
They point to well-documented cases of equipment failure including problems and glitches in California and Maryland during the March primary elections, some of which led to votes being lost or mis-allocated.
Critics also complain that electronic voting system manufacturers refuse to allow open scrutiny of the software running their systems though in one instance the software from one vendor leaked out on to the Internet when the company left it on an insecure server.
But most crucially, critics worry that many DRE systems fail to produce a paper trail to allow results to be double-checked should they be challenged. Ideally, electronic systems, which typically ask a voter to confirm their choices on screen before recording their ballot, should also produce a voter-verified paper record that can be checked should a recount become necessary. "The lack of a paper trail is a serious issue for recounts," says Herbert (Hugh) Thompson, director of security technology and research at Security Innovation, an application security services company based in Boston.
Indeed, Dr Thompson claims that "hacking" e-voting systems is so trivial that almost any 15-year-old computer enthusiast could do it. "It can be done with just five lines of script," he says.
"The technology just is not ready to be implemented in this way," says Dr Thompson, and he is not alone. "The problem is these machines were not designed for security," says Avi Rubin, a computer science professor at Baltimore's Johns Hopkins University, one of the fiercest critics of first generation DRE machines.
Naturally the systems manufacturers and others strongly disagree. They insist that electronic systems are more accurate than most other voting systems and easier to use, particularly for those with disabilities. "We believe our systems are both accurate and reliable," says ES&S.
Many state officials and voters in opinion polls also report positive experiences with electronic voting. In Georgia, for example, officials say the new electronic systems they implemented after the 2000 election have dramatically reduced the vote-tallying errors that arose under the old systems.
The leading equipment makers also claim their machines contain many hidden security features. For example Diebold says its machines keep multiple copies of a vote in separate electronic storage devices including flash memory cards, ensuring it is impossible for votes to be "lost".
In a recently published detailed response to critics of its machines and the Global Election Management Software (Gems) it uses, Diebold said: "The allegations about the supposed manipulation of Gems rely on the assumption that someone is able to gain free and unsupervised physical access to the server.
"In reality, nothing is further from the truth. Election administrators, officials and regulators take the physical security of their voting systems very seriously. Regulations and procedures are in place that prohibit unauthorised access, complete with physical and password security."
Breakdowns
Still, some state and federal officials have acknowledged that mechanical breakdowns can pose problems, particularly when polling station staff have not been trained to fix technical problems.
Most importantly, however, they point out that despite dire predictions, there has not been a single documented case of vote tampering in the US since the first DRE machines were introduced 20 years ago. "Our systems are as secure as we can make them," says David Bear o
Sign up for the Daily Briefing
Get the latest news and updates straight to your inbox
Network Links
GN StoreDownload our app
© Al Nisr Publishing LLC 2026. All rights reserved.