By Michael E. Miller
By Allie Conti
By David Villano
By Jose D. Duran
By Michael E. Miller
By Allie Conti
By Kyle Swenson
By Luther Campbell
Fifteen counties, including Miami-Dade, purchased touch-screen systems. These machines produce no paper ballots. In a close election, the only way to recount is to ask the machines to re-add the numbers they already produced. The success of ES&S in securing the contract for the largest county in the state is partly due to the shrewd choices the company made in hiring lobbyists. For instance, it hired former secretary of state Sandra Mortham, who was Jeb Bush's original choice of running mate in 1998, until she was tarnished by a spending scandal in her office. She lost her seat to Katherine Harris, that other darling of voting rights advocates. In 2001, Harris made ES&S's touch-screen system the first certified in Florida. Mortham got a cut of every deal the company made. The company's lobbyist for the sale in Miami was Miguel de Grandy, long a successful suckler at the county commission teat. In January, 2002, a Miami-Dade selection committee headed by David Leahy chose ES&S and the commission approved the contract.
After the elections problems in 2002, the county transferred Orlando Suarez, a division director, from the Enterprise Technology Services Department, where he was a kind of floating expert on various projects, to the elections department, where he was put in charge of the voting machine technology. In June 2003 he discovered bugs in the system from a May runoff election in North Miami Beach that made it difficult to match vote totals to a particular machine. He reported these problems to the vendor. The most significant conclusion Suarez drew was this: "The purpose of my review was to make sure that we could use these reports to 1) audit an election, 2) recount an election, and 3) if necessary, use these reports to certify an election. Unfortunately, if my observations are correct, we cannot use these reports in their present state for any of these purposes."
News of the bugs never made it out of the elections department, however, until the reform coalition began digging into the records of e-mails, memos, and reports. In December 2003 Dan McCrea was pawing through a box of records when he stumbled upon the Suarez e-mail. Later, an October 2003 e-mail from Suarez to Kaplan revealed that he had reviewed the audit logs for the October 7 primary in Homestead, and found that the activity record from five machines on which 162 votes were cast was missing. Big problem.
Connie Kaplan didn't agree. After accepting a "workaround" solution from ES&S, she transferred Suarez out of her department and back to his department. In his place, she promoted computer tech Donald Llopis. This later proved an embarrassment to her, when he "lost" important elections data. Suarez is, by some accounts, an exacting man. But he knows computer and management systems well. His personnel file is thick with commendations for the quality of his work going all the way back to his hiring as a computer programmer in 1977.
Llopis is a different story. The bulk of his career, from 1980 to 2002, was in sales. For instance, from 1998 to 2001, he was a general manager in sales for Café Bustelo, then had a short stint as a regional sales manager for Fedders International. In May 2002, his career took a sharp turn when he was hired by the elections department as a computer tech to work on the new iVotronic machines. An evaluation in his file gives him gold stars for his work ethic.
But why would Kaplan want to replace a competent fellow boasting 30 years of technical experience with a former sales manager who'd spent less than two years on the tech side? Kaplan says that Suarez asked to return to his old department. She adds that she put Llopis in charge of electronic voting because she reorganized the department for more efficiency and that's where his specific experience lies.
It was around this time, about December 2003, that the coalition's relationship with Kaplan took a turn. When the first Suarez e-mail was discovered, members met with her, but, they claim, she was dismissive of the problem. She acknowledged the glitches, but her attitude was that as long as the vote totals were accurate, the audit problems could be dealt with. Kaplan said as much to a reporter from the Daily Business Review, which broke the first stories on the issue, and the same again when New Times asked her. "The tabulation of the equipment has never been in question," she asserts. On her relationship with the coalition, she forces a tight smile. "It's my goal to have a good working relationship with them," she says. "My issues are if I've answered something, and these things that are not true keep getting brought up and brought up."
But Mahoney, the UM professor, had been consulting with computer experts around the nation, and in March 2004 she came to the conclusion that the audits absolutely had to be accurate. That's because there's no other way to determine whether the machines recorded every vote cast. There was, she reasoned, a real danger of undervotes, and no one would know without an accurate audit or a paper trail. There was some evidence supporting this conclusion. In January 2004, there was a special election for a state representative seat spanning Broward and Palm Beach counties. In Broward, about 10,000 people showed up to vote on the iVotronics that county also uses. There was only the one issue on the ballot, yet somehow, 134 of the people who signed in to vote that day, apparently, inexplicably, failed to do so. This was significant because the winning margin was twelve votes.