Today's Offers:

Arguments > Their View

Electronic voting machines are trustworthy
2/14/2008 9:30 AM  

SZ200_BIreland.jpg

By BETTY IRELAND

CHARLESTON -- In recent months, several news reports have raised questions about the trustworthiness of touch-screen voting devices manufactured by Election Systems & Software (ES&S). The scrutiny arises from reports issued by a handful of states barring the further use of ES&S devices in their elections.

Since ES&S voting machines are used in 34 West Virginia counties, I feel compelled to address this issue and clarify several points in regard to their accuracy and reliability. As Secretary of State, I am personally devoted to the premise of fair, accurate and trustworthy elections in our state. After carefully and thoroughly reviewing the reports issued by these other states, I remain confident that our current voting technology will provide voters in West Virginia with dependable and accurate voting results.

The fact is, so-called "problems" that have been widely reported about ES&S voting machines, including optical scan and precinct counting machines, have never been evidenced in actual balloting by voters participating in any elections. Put another way, to my knowledge these machines have never been hacked on Election Day, and no votes have been tampered with at anytime in any election. Rather, all of the issues reported in the press have come as a result of "tests" conducted in a few states, tests which varied widely in their method and controls.

The conditions under which Ohio and Colorado conducted their tests simply do not mirror the actual conditions present in an election precinct, and they do not reflect the measures in place before, during and after an election to ensure votes are cast and counted properly.

For example, in Colorado, state elections officials claimed that a magnet held close to the device caused a failure in operation. But when county officials in Mesa County, Colorado, tried to duplicate those tests, they encountered no problems at all.

A recent news story in The (Grand Junction) Daily Sentinel reported on the Mesa County tests as follows: "Thursday's mock election conducted by the Mesa County Clerk's Elections Division, with its decertified ES&S electronic vote machines, went off without a hitch. Try as they might, elections officials could not reproduce the problems that led Colorado Secretary of State Mike Coffman to decertify the county's voting machines recently." (The Daily Sentinel, Jan. 10, 2008)

West Virginia has one of the most stringent testing procedures in the nation. State law requires that the machines be subjected to a pre-test, a public test, an Election Day test, and a test prior to canvass. Security measures required by statute and those developed by my office will protect against the concerns, however remote, identified in these reports.

Elections are always subject to corruption and error. Long before the advent of electronic voting devices, parts of our state were historically plagued with election corruption. In fact, the easiest ballots to corrupt are paper ballots.

When individuals take it upon themselves to manipulate or corrupt an election, they will, sadly, often find ways to do so. But I believe the electronic systems we have in place now in West Virginia, which include touch-screens as well as optical scan and precinct count machines, make such efforts more difficult than almost any other method of voting.

What would be truly irresponsible would be for anyone to advocate decertification of the ES&S machines based on news reports, or testing that was conducted under improperly controlled settings, or complaints that are politically motivated. (Notably, many states, including South Carolina and neighboring states Virginia, Kentucky and Pennsylvania, have used or plan to use ES&S devices in their '08 primaries.)

Responsible decision making on this issue requires a consideration only of sound, properly controlled testing procedures, and actual election conditions and results. Under those criteria, West Virginia voters can be confident in the performance, integrity and accuracy of ES&S touch screen voting machines.

Ireland is West Virginia's Secretary of State.

EMAIL A FRIEND | PRINT | DIGG THIS | POST TO DEL.ICIO.US

Comments on this article

  • Electronic Voting Machines

    The major problem with ES&S touch screen voting machines (and with similar DREs) is that there is no way to know for sure that the election "went off without a hitch." Since there is no paper record (that the voter verifies) of how the voter voted, there is no way to know if the voting machine is reporting the correct totals. The computer scientists who have looked at the building blocks of these machines have found serious flaws that can cause votes to be incorrectly recorded, not due to hacking, but due to poor design and flaws in the machines themselves. Claims that the machines always work correctly are unprovable. Without a guarantee that the vote recorded is the vote cast, our democrary is in danger. Paper ballots and always hand-counting a random sample of the ballots is necessary any time a machine is trusted to do the counting.

    by Eleanor Hare

    url:

  • Electronic Voting Machines Trustworthy?

    Ms. Ireland’s article on electronic voting glosses over serious concerns raised by the use of un-auditable and un-verifiable electronic voting systems by suggesting that “news reports have raised questions” about their trustworthiness. By choosing to ignore the overwhelming evidence gathered in a number of authoritative studies by professional Organizations (Association of Computing Machinery), government agencies (GAO) and educational intuition including Brennan Center at NYU, and John Hopkins University and others involved in voter integrity efforts she is misstating the problem and misleading the people of West Virginia. I fully understand that part of her job is to maintain voter confidence in whatever technology is used for conducting elections, but that confidence should be well placed and justified by ensuring that there are methods for the voter to verify that their vote is counted (and cam be recounted if needed) as cast and that the process is transparent and auditable. We depend on our Public Officials such as Ms. Ireland to get it right. This time not only does she not get it right, but attempts to convince the public that these are simply misleading news reports and that there is no real underlying problem. Voting is the basis of our democracy and thus every effort should be made to ensure that elections are conducted fairly and accurately. GAO in [http://www.gao.gov/new.items/d05956.pdf] pointed out numerous flaws with software based electronic voting systems, including the following: 1) Cast ballots, ballot definition files, and audit logs can be modified. 2) Supervisor functions are protected with weak or easily guessed passwords. 3) Vendors installed uncertified voting system software before elections. 4) Voting systems experienced operational failures during elections. NIST, (National Institute Standards and Technology) called for systems that are “Software Independent” [http://vote.nist.gov/DraftWhitePaperOnSIinVVSG2007-20061120.pdf], meaning that the record of how a voter cast his or her vote should be independent of any software in the voting machine. In order to meet this requirement, NIST recommends the use of voter marked paper ballots or ballot marking devices. Ms. Ireland’s statement that “so-called problems” have never been evidenced in actual balloting by voters may be her only agreed on statement. Aside from those instances when voters have observed votes being assigned to the wrong candidate, these systems by their electronic nature do not leave evidence of vote changes. Unless voters were able to examine the electronic data stream as it is transmitted to the central server (another software controlled electronic computer with its own problems), and compare it to their intended vote there never will be any evidence that the machine performed either correctly or incorrectly. It is understood that voting systems cannot be changed in time for the 08 Presidential Elections, but it is irresponsible for officials to continue to use these machines beyond that point. I also agree that decisions should be based on “sound, properly controlled testing”, and that the testing should be done by experts in computer security (WVU would be a good start) and not by in-house personnel or by the vendor funded Testing Authorities.

    by J. Johnson

    url:

SEND US YOUR COMMENTS ON THIS ARTICLE:

* - Required fields

Subject: *
Message: *
Contact Name: *
Contact URL:
Contact Email: *
This Is CAPTCHA Image
Write the characters in the image above: