In a welcome outburst of practicality, the San Francisco Board of Supervisors voted 9-2 on Tuesday to approve a four-year, $12.6 million contract for electronic voting machines that will be online in time for the Feb. 5 presidential primaries. The timing was almost bizarrely appropriate, coming just three days after city elections officials were finally able to certify the results of the Nov. 6 election following a tedious five-week hand count of some 80,000 ballots.
We applaud the supervisors for taking time out from telling the world how to solve its problems, thus eliminating San Francisco’s risk of international embarrassment from repetition of the November election fiasco. Without these new machines, The City could easily have become the butt of worldwide jokes by delaying final recognition of California’s presidential primary winners for more than a month.
Oakland-based Sequoia Voting Systems has now guaranteed to deliver updated, state-certified machines next month and to pay $3 million in penalties if there are any malfunctions. This is essentially the same deal that the board unrealistically rejected last February, siding with voting-advocacy groups that insisted the contract must not be approved unless Sequoia offered open-source technology enabling public access to the source code that tallies the votes. Such a high-minded stance might be perfectly fine in an ideal world, but no voting-machine company today is willing to reveal its proprietary code to business rivals.
The supervisors’ earlier refusal to replace San Francisco’s discredited ES&S AutoMark machines triggered a chain of events in which The City ultimately was forced to tediously hand-count its ballots. California Secretary of State Debra Bowen had decertified ES&S over multiple statewide complaints, particularly involving charges that the Omaha, Neb., company’s machines could not properly read absentee ballots marked in lighter inks. At this point, The City and Bowen are independently suing ES&S.
We cannot resist saying yet again that there never seemed any valid reason to replace the rugged, proven optical-reader voting systems that were predominantly in use throughout the Bay Area. Fast optical-reader balloting had nothing to do with the Florida hanging-chad mess that sparked a national stampede for election technology upgrades after the 2000 presidential election. Counting votes by computer is intrinsically a more fragile and much less transparent process than the optical readers with their paper ballots, and some 17,000 problems with digitally miscounted elections were reported around the country in 2006.
Still, among the limited options available on San Francisco’s table now, going with Sequoia Voting Systems was clearly the best choice. And since the new four-year contract also specifies that Sequoia must switch to open-source code within a year if another company brings an open-source machine to the market, even the most hard-core voting transparency advocates ought to be satisfied.