I’ve received a few notices recently of a new agreement that the State of Georgia struck with Diebold Election Systems to buy 6000 “ExpressPoll” electronic poll books, along with a service contract, for what has been quoted as costing the state $15 million. A story from WTOC in Savannah, Georgia provided details about how the electronic poll book will work:

The ExpressPoll-4000 will provide poll workers with immediate access to voter information at each voting precinct. Voters are quickly and accurately verified by poll workers who simply touch a few characters of the person’s first or last name and/or date of birth on the unit’s intuitive touch-screen interface.

The functionality of the ExpressPoll-4000 enables poll workers to immediately identify if a voter is not in the correct voting location, and can quickly provide the address of the voter’s designated precinct. A detailed street map may also be presented on the ExpressPoll-4000, providing visual instructions to the location of the voter’s designated precinct. The ability to rapidly verify whether a person is registered will reduce the number of provisional ballots issued in precincts.

…

Registration information for all voters within a jurisdiction can be stored on the ExpressPoll-4000 to dramatically reduce the number of calls placed by poll workers to voter registration offices on Election Day. A poll worker can quickly and easily locate the name of any registered voter along with their appropriate precinct location, increasing voter convenience.

The ExpressPoll-4000 can also eliminate the need for costly printed voter registration books because the names of all voters in a jurisdiction are duplicated from the official voter registration roll for the respective election and stored within the memory of the unit. In addition, the ExpressPoll-4000 will be used in Georgia precincts to create voter access cards used by each voter to activate the Diebold touch-screen voting stations. The ExpressPoll-4000 will identify the correct ballot style for each voter based on stored information from voter registration files, further automating and refining the voter check-in process.

It’ll be interesting to figure out methodologies to evaluate how electronic poll books work in practice, as this a concept that many have endorsed. In fact, in 2001 the Caltech/MIT Voting Technology Project recommended “putting the complete registration database for a county on a compact disk and leasing a laptop computer for each polling place. Where this has been done it has reportedly eliminated a majority of registration problems and reduced polling place bottlenecks” (page 29).