The New York Times
March 14, 2005
For Voting Machines We Can Trust (4 Letters)
To the Editor:
Re "Virtues of Optical-Scan Voting" (editorial, March 9):
You are right to suggest that the heavily lobbied touch-screen voting machines now under consideration by the New York Legislature should not be used at all.
Optical-scan voting is far cheaper, faster in the polling place and produces the best kind of paper trail, a ballot marked by the voter.
While optical-scan voting is superior, there must be safeguards. Initial scanning should be done on the precinct level to minimize error before results are sent to the local board of elections. Results should then be tabulated county by county and transmitted to the final tabulating location.
There should not be a central, politically motivated state bureaucracy in charge of the process. Vote counting should be from the bottom up.
Paper ballots, after they have been scanned, should be guarded like gold in Fort Knox. A security system must be in place that will prevent the lifeblood of a democracy from being lost, strayed or stolen.
Dobbs Ferry, N.Y., March 9, 2005
To the Editor:
A recent Caltech-M.I.T. study clearly shows that touch-screens are the most accurate and efficient method of voting. The study recognizes Georgia, which uses touch-screens across the state, as making the greatest improvement in voting accuracy throughout the country.
Regarding the cost advantages of optical-scan machines, you do not mention the long-term costs related to printing ballots that are inevitably passed on to taxpayers. These costs, particularly in large cities that require many ballots in several languages, are one of the primary reasons most election officials prefer touch-screens to optical scanners.
Additionally, optical-scan machines are not "far cheaper than touch-screens." Per unit, the cost of optical scanners is about $1,000 more than a typical touch-screen machine.
If the bills in the New York Legislature are more focused on touch-screen voting as opposed to optical-scan technology, it's because forward-thinking legislators are acutely aware of the advantages of touch-screen voting.
Thomas W. Swidarski
President, Diebold Election Systems
McKinney, Tex., March 10, 2005
To the Editor:
You urge New York legislators to favor optical scanners because they are "the best voting technology now available." They aren't.
Despite the fact that the voter personally marks the ballot and has the chance to verify his or her choices, no machine has ever been built that can read a ballot the way a human eye does, and there is no assurance that the machine will count the ballot the way it was marked by the voter.
Even if a manual recount is performed flawlessly (an impossibility considering the charged atmosphere under which such recounts occur), the mark made by a voter may not be counted because the states have developed different and obscure criteria for what constitutes a valid optical vote.
The fundamental problem is that a ballot offers only a finite number of candidate choices, but an optical-scan ballot can be marked by a voter in an infinite number of ways.
There is no consistent method of determining voter intent from an optical ballot, so some voters will necessarily be disenfranchised through their use.
Electronic machines do not suffer from this defect. They offer a finite number of yes-no choices, so there is no possibility of mistaking voter intent.
Michael I. Shamos
Pittsburgh, March 9, 2005
The writer, a professor at Carnegie Mellon University, is a consultant to the secretary of the Commonwealth of Pennsylvania on electronic voting.
To the Editor:
We use optical-scan ballots throughout Minnesota. We don't have long waits at the polls, and recounts are easier than with other systems.
Optical-scan ballots are certainly better than electronic voting machines.
Brooklyn Center, Minn.
March 9, 2005
Copyright 2005 The New York Times Company
FAIR USE NOTICE
This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available in our efforts to advance understanding of political, democracy, scientific, and social justice issues. We believe this constitutes a 'fair use' of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml. If you wish to use copyrighted material from this site for purposes of your own that go beyond 'fair use', you must obtain permission from the copyright owner.
Errors in the letter from Thomas W. Swidarski, President of Diebold Election Systems:
Error A. "Additionally, optical-scan machines are not 'far cheaper than touch-screens.' Per unit, the cost of optical scanners is about $1,000 more than a typical touch-screen machine."
In truth, New York State will save over $50 million dollars in up-front costs by purchasing paper-ballot-optical-scan systems (PBOS) rather than touch-screens (DREs). Mr. Swidarski failed to mention that we would need to purchase fewer PBOS systems than DREs.
One or more DREs would be needed to replace each lever machine. Meanwhile, one optical scanner can replace several lever machines. Small polling places can use one PBOS for several election districts. Only polling places with a very large number of election districts may require two or more PBOS systems.
Mr. Swidarski claims that a DRE is less expensive than an optical scanner, but this is true only for ATM-style DREs that are made by Diebold. Due to New York's full-face ballot requirement, we cannot use ATM-style DREs. The full-face DREs we will purchase are MUCH MORE EXPENSIVE per unit: a scanner is around $5500; a full-face DRE is around $8000!
Error B. "Regarding the cost advantages of optical-scan machines, you do not mention the long-term costs related to printing ballots that are inevitably passed on to taxpayers".
In truth, the long term costs of DREs are far higher, even considering the cost of ballots for the PBOS systems. Not only do DREs require substantially higher storage and transportation costs, but they have a lifespan of about 5 years. Optical scan systems have a lifespan of 12-15 years. Counties will need to buy completely new equipment in 5 years with DRES! Also, paper ballots must still be printed with DRE systems for absentee, affidavit, and emergency ballots. A cost comparison of two Florida Counties using DRES and PBOS respectively showed that the county using DREs spent over a million dollars more per year (http://www.votersunite.org/info/costcomparison.asp).
Error C. "A recent Caltech-M.I.T. study clearly shows that touch-screens are the most accurate and efficient method of voting. The study recognizes Georgia, which uses touch-screens across the state, as making the greatest improvement in voting accuracy throughout the country."
In truth, no study has ever shown the accuracy of touch-screen systems, since there has NEVER been an independent audit of these systems. The CalTech-M.I.T. study used unverified and unverifiable numbers provided by the states, which simply made various assumptions about their equipment.
Moreover, when studies such as the one from Cal-Tech-MIT compare "undervotes" we need to know specifically what was counted for each system. It has been alleged that when Diebold reports "undervotes" they are reporting only races in which multiple candidates are elected to fill multiple offices, and the voter has selected fewer than the maximum number of candidates allowed. Diebold’s count of undervotes would not include a race for President, where only one choice is allowed, but no choice is recorded; Diebold would call this a "blank vote," not an "undervote."
Misleading statement A. "an optical-scan ballot can be marked by a voter in an infinite number of ways."
In truth, almost all wrongly-marked paper ballots will be detected as invalid ballots by the optical scanner, and the voter would be allowed to correct the ballot BEFORE it is cast. Scanners detect over-votes, under-votes, and stray marks.
Misleading statement B. "no machine has ever been built that can read a ballot the way a human eye does, and there is no assurance that the machine will count the ballot the way it was marked by the voter."
In truth, machines work differently than the human eye, but machines can read much faster and more accurately. Optical scanners used by our post office read scribbled addresses on envelops, and optical scanners used by our banking system read billions of checks daily (with an error rate that is close to zero).
A machine might count the votes on a ballot incorrectly because the machine might be programmed wrong. This problem is not related to how the machine reads. This problem is related to human errors in programming, and has to be detected by routine logic and accuracy testing before the election, and manual recounts of ballots after the election.
Misleading statement C. "the mark made by a voter may not be counted because the states have developed different and obscure criteria for what constitutes a valid optical vote." "There is no consistent method of determining voter intent from an optical ballot"
In truth, there is much corruption in the laws, regulations, and practices of various states, and the solution is to reform the laws, regulations, and practices so that the obvious intent of the voter is counted. Problems can be hidden but not corrected by using electronic voting equipment that cannot or will not be audited so that errors cannot be detected.
Misleading statement D. "Electronic machines ... offer a finite number of yes-no choices, so there is no possibility of mistaking voter intent."
In truth, electronic machines are dependent on their programming, which can make any mistake imaginable. Electronic machines offer infinite possibilities for undetected loss and change of the voter's intent via hacking, corrupt insiders, and innocent errors in programming.