This archived article was written by: Erik Falor
The controversy surrounding electronic voting booths has raised many questions about the role of technology in our government. In years past, daydreamers fantasized about the day when computers could take on the tedious and repetitive task of tallying ballots for their human masters.
Some of the earliest computer systems were designed for this purpose, and electronic voting equipment has been in use since ’30s. With the proliferation of computer technology in the ’70s and ’80s, it seemed like it was only a matter of time before computers would make it possible for every citizen to voice his/her opinion on each issue and we could replace our representative republic with a pure democracy.
While technology may never replace representatives and electoral colleges, it has impacted politics in our country. The initial success of Howard Dean’s campaign is attributed to his Internet campaigning. Online propaganda campaigns such as MoveOn.org have been wildly successful at getting citizens more involved in the democratic process.
Thanks to the biggest electoral scandal in American history that resulted in George W. Bush’s ascent to the presidency, electronic voting has been looked to as a viable alternative to using lever-operated voting machines for national elections in America. Advocates of electronic systems praise their intuitive interface and ease of use by people with disabilities. There are no levers to malfunction, and no “hanging chads” to get caught in machinery. Promoters of electronic voting also believe that the introduction of these systems will encourage higher voter participation, streamline the voting process, and ultimately reduce costs.
It should be noted that every voting system ever implemented in any country is vulnerable to its own host of problems. From one-party nations with only one candidate on the ballots, to stuffing ballot boxes with phony ballots, every system can be made to fail.
While technology has its place in politics, we should be skeptical of measures that would force us to completely trust in the software written by private companies; especially companies refuse to share the source code with the public. Using computers to count ballots is one thing. If there is a mistake, we can still do a hand-count. But using computers for the purpose of taking the ballots is another matter entirely. One can never be entirely sure just what is happening inside of the user-friendly black box inside the voting booth.
The supposed benefits do not outweigh the inherent risks in using sophisticated computer systems. Recent e-mail viruses afforded us a first-row seat to watch how quickly computer systems can be brought to their knees. Before any attempt to replace conventional voting systems with computerized systems is undertaken, computer security issues must be addressed. Unfortunately for democracy, all of the attempts have not taken much care.
It is a gross understatement to simply say that there are many security issues with electronic-voting machines. Not only are these so-called “design flaws” well documented by software experts outside of the companies, one company, Diebold, instructed its representatives and sales persons to deny the severity of the problems. The flaws in Diebold’s voting software gave users of the system the ability to vote multiple times, view ballots already cast on a machine, modify party affiliation on ballots, cause votes to be miscounted, create, delete and modify votes on voting machines, and tamper with audit logs and election results (voting rights).
Any one of these security flaws might be forgivable if it were the only problem present in the software. But the fact that there are at least seven different critical failings in the program is reason enough to doubt the competency of the entire division responsible for producing it. A version of software with that many known problems should never have left their research campus. That degree of crappiness is not tolerated in hand-held calculators, digital wristwatches or Furbies. Nor should it be present in the software that helps select our next leaders.
Diebold seems to be following the same “get it out to the public first” business model that commercial vendors of office suite software, video games and noncritical systems followed for years. This business model offers substantial economic rewards for a company that can rush a product to the shelves before its competitors can. However, it has taken its toll on consumer confidence. Ian Sommerville, head of the computing department at Lancaster University declares that it is ” … a sad reflection on the software industry that many users have low expectations of their software and are not surprised when it fails during use” (422).
But users should not settle for poor software controlling our voting booths. If heart pacemaker control software needed to be rebooted every 24 hours, it would never fly. A bank whose ATM debited accounts $40 every time $20 was withdrawn would lose its customers.
Because of the high social costs at stake, we must demand that all electronic election systems be held to the same high standards as banking and medical systems. Higher standards means more time and money be invested in the testing and developing.
It is not known how Diebold spent their time and money on the project. It is an ironic sidenote that at least one part of Diebold’s design process is concerned with security: the details surrounding the project are kept in confidence. There are special trademark laws protecting the source code of electronic voting systems. Special laws protect the source code from the scrutiny of anybody outside of the company. Because of these special measures, I dare say that these flaws are not simply the result of oversight, but are deliberate attempts to influence the outcomes of elections.
The theory is that details of the software should be kept secret so that any vulnerabilities that have escaped the development floor will not be discovered by crackers. But that contention is out of vogue. Paul Boutin, a writer for MSNBC, points out that Diebold should have gone with the flow and posted its code on the Internet for everybody to see. “It’s the only way security experts (real or self-imagined) will ever take the company seriously” (MSNBC.com).
Diebold and other companies have deliberately withheld their program’s source code from the review of software engineers. The code still made it out though, due to employees who leaked the booth’s source code on the Internet. This released source code revealed faulty code and poor software design. All of the experts who have reviewed the software would agree that the programmers aren’t qualified to program a VCR.
In the eyes of some, the fact that the software is poorly written does not explain the density of the known software problems. Nor does it explain to them why some of the features seem more like “backdoors.” They believe that it is a GOP conspiracy to fill seats with Republican candidates. Surely, most reported incidents of voting boot “failures” have benefitted Republican candidates.
One simple fix would be the addition of printers that make hard-copy backups of the results. Currently, electronic voting booths do not have printers installed by default. That means that it could show the users that his/her vote counted for their candidate, but secretly increase the other candidate’s tally. Without a hard-copy as proof, there would be no way to verify who voted for whom.
With no laws to force precincts to use printer-equipped voting booths, the prospect of including them is bleak. They represent a 10 percent markup on the cost of the booths, and some districts would be hard-pressed to come up with that money. Because the law does not require it, there is no incentive to include them.
While the situation seems bleak for voters this year, the technology promises to improve before the next election cycle. Due to the level of scrutiny that Diebold has been under, they stepped up their efforts to produce secure software, and the latest versions of the their voting booth are reportedly quite secure.
Despite the advances, electronic voting will be slower in catching on, due to the untrustworthy nature of technology. Some systems are just not meant to be upgraded into a digital incarnation. The security issues will never be able to be satisfactorily resolved, and the uncertainty of just how the computer in the booth works makes it intrinsically untrustworthy.
It is not unfeasible for a cracker, or a politically-motivated company insider to write a patch for the booth that does something other than what the voter expects. Plus, pulling a lever or punching holes on a card feels more official than clicking on icons on a touch screen. It will feel less like picking our future leaders and more like setting the options on your music player. Computers will never be able to recreate the experience of voting. In the meantime, demand to use a paper ballot instead, and for Pete’s sake, make sure you take care of your hanging chads!