How to steal an election...How do we stop it?

Want to be a......President? Then follow this simple how-to formula for stealing an election:  

             ELECTION = TECHIE + MOTIVATION 

That's it. One reasonably clever techie. Period. More startling is the realization that from a single point of infiltration, potentially every....every voting machine in the country can be infected. That is every voting machine supplied by the same vendor.  Methods the techie uses will vary of course. But as long as he or she can get within a reasonable proximity to any strategic voting machine component, infiltration is possible. Hand-held computers, CD's, and flash drives used legitimately for initializing the voting machines and communicating ballot configurations present indisputable risk. Just one rogue hand-held programmed to infect one voting machine can take over the entire fleet. For as soon as the infected voting machine transfers the rogue program up-line to the server, the server can populate all the other voting machines with the next download. Other pervasive components include wireless networks, cable and phone lines, for those voting machines and ballot scanners transferring their vote tallies across networks or receiving software upgrades. 

A much easier voting machine coup can be achieved by an insider from within the vendor company. For example, take any versatile voting machine manufacturer. We'll call this company ElecTron, Inc. ElecTron manufactures, distributes and supports touch screens and ballot scanners in varying models. Their election consultants rank among the best. However, all ElecTron machines share the same core vote tallying software and basic upload-download processes. Every time ElecTron upgrades its software, the new software version populates all ElecTron's voting machines. Alarmist? Yes. Possible? Absolutely. 

As a former manager of election software and career information technology professional, I look at 2000, 2004, 2006 and now 2008 from a business perspective. In the world of computer software, it's tempting to jump to the quick fix without ever defining the real problem. So many intelligent, well meaning people are caught-up in this voting machine mix. But I suspect few know computer-speak. Nor should they. If that was their intent, their passion, they'd have gone to techie-geek school. Some of us are born geeks. Others for certain are not.

So, from an application development standpoint, we need to step back a moment. Stop looking for the perfect solution until we consider the whole problem. We want our votes counted...and we want our votes to count, every time. But computers will fail. Systems will fail. Consequently, the most glaring failure to date is not pregnant chads, butterfly ballots, or 18,000 missing votes, it's the failure of our election laws to provide failsafe recovery.

Until we fix our election laws to be on par with technology, to protect us from machine and human error, and HUMAN INTERPRETION of election results, our election process will continue to be broken. The courts should not decide the people's choice. 

As for the actual voting machines, if we are to achieve one voter, one vote... every time, we must go back to basics. Apply sound business practices to ensure the voting systems we purchase fully meet our needs for election integrity. And ensure those machines work...all of them. Not just a sampling.

We must implement and follow high-bar guidelines for voting machine providers and elections officials to uphold, no hedging, no exceptions:

• Demand scrupulous, independent and random pre and post election audits...in statistically relevant numbers and frequency.
• Rigorously quality-test more than a token number of machines prior to every election.
• Maintain stringent controls and accountability for all software and hardware upgrades to voting machines and peripheral procedures, including:

o   Parameter-driven upgrades such as ballot configurations.

o   External programs and processes applied to individual voting machines. These can accidentally alter a machines processing environment and thus affect voting results. For example, Sarasota's explanation of a "smoothing filter" sometimes delaying machine responses may indicate a lack of oversight and control. Not all systems were equal.

• The same stringent controls applied to voting machines should apply to external machines and devices such as hand-held boot up devices, flash drives, and CDs.
• Track all activities and inconsistencies on all voting machines and peripheral machines, such as hand-helds by date, time, purpose, results, and serial number.
• Audit all software and hardware upgrades as well as machine deviations and failures.
• As mentioned above, revise election laws.

Your eyes glaze over yet? Mine too. So for my part, I took my one small voice to the printed page hoping that fiction might spark interest where technical speak and logic do not. And while the heroine of my novel scoffs at the notion of a silent coup marching across the country in her fictitious voting machines.... It could happen more easily than any of us want to believe.