This document debunks many of the common misconceptions about electronic voting machines.
Myth: A voter verifiable paper trail is a printer attached to a touch-screen machine.
Fact: A voter verifiable paper trail (VVPT, otherwise known as a "voter-verified paper ballot," or VVPB) is nothing more or less than a permanent paper record of the vote that the voter can check for accuracy (by some trustworthy method, such as visual inspection) before the vote is cast. The record must be deposited in a secure ballot box for use in a manual recount or audit.
This definition is quite broad, and encompasses "plain old paper ballots" that are manually marked and counted, central and precinct-based optical scan ballots that are hand-marked but read by computers, and printers on a touch-screen or other computerized input device.
At this time, the Verified Voting Foundation recommends precinct-based optical scan technology. It is widely used and proven in practice and studies have shown these systems to have lower residual vote rates (votes for too few or too many candidates) than e-voting machines.
Historically, the biggest problem with optical scan has been accessibility for people with certain disabilities, such as blindness, that make it difficult to vote a paper ballot. However, this situation has changed. New equipment is now coming on the market that provides an accessible computer interface to optical scan ballots (e.g., the AutoMark system being sold by ES&S). Some places also use low-tech tactile ballots for accessibility. For more information on voting machine accessibility, see our statement on disability access at http://www.verifiedvoting.org/article.asp?id=1875
Myth: Receipts will enable voters to prove how they voted to someone outside the polling place, enabling vote influencing or selling schemes.
Fact: This concern is based on a misunderstanding. Voter-verified paper ballots must be deposited in a secure ballot box in the polling place, even though some people call VVPBs "receipts." There is no more risk of vote selling with optical scan ballots or ballots printed on a touch-screen machine than with other kinds of ballots. There is much less risk of vote selling than with absentee ballots.
Myth: E-voting machines cannot be hacked because they are not connected to the Internet.
Fact: Computer systems can be hacked in many ways without using the Internet. Making systems secure against outsiders, such as voters and poll workers, is very hard, and, as multiple studies have shown, the current e-voting systems fail miserably. However, making them secure against INSIDERS, possibly even the programmers themselves, is close to impossible.
The way we make systems honest is to enable truly independent audits. Each voter should be able to check that his or her vote is recorded correctly, and it should be possible to count the paper ballots manually to double-check any machine counts.
Some e-voting machines are believed to have wireless connectivity that might enable Internet access with or without the knowledge of poll workers and election officials.
Myth: H.R. 2239 advocates "a return to flawed systems."
Fact: This is not the case. The Voter Confidence and Increased Accessibility Act (H.R. 2239) simply requires voter-verified paper ballots, as described above.
Paper systems are not "flawed." A 2001 Caltech/MIT study and several subsequent studies have found that paper ballot systems have lower error rates than all other voting methods, including direct recording electronic (DRE) machines.
Myth: Voter-verified paper trails "would force voters with disabilities to go back to using ballots that provide neither privacy nor independence, thereby subverting a hallmark of the HAVA legislation."
Fact: So far as we know, no one is proposing to suspend or delay the HAVA requirement that there be at least one accessible voting system in
each polling place by 2006.
Paper ballot systems can be made accessible in several ways: There is a touch-screen interface for optical scan ballots (described above); touch-screens that print paper ballots can also have equipment to read those ballots back to voters using an audio interface; "ballot on demand" systems that print blank optical scan ballots as needed in the polling places can also have accessible interfaces that allow voters to make their ions on the computer, then print out a ballot that is marked appropriately; and there are even low-tech ballot "tactile ballots" that have been used in Rhode Island and several countries to make optical scan systems accessible without computers.
Myth: The Technical Guidelines Development Committee of the Election Assistance Commission (EAC), not Congress, can "ensure the reliability of the computer technologies being employed in voting systems."
Fact: This is impossible, especially as applied to the next general election. The EAC Commissioners themselves were not appointed until December 2003, and to this day there is no Technical Guidelines Development Committee. Even if there were, the EAC is not required to produce guidelines until nine months after its members have been appointed.
Even if it released guidelines tomorrow, the guidelines are voluntary, as determined by HAVA’s authors, and could be changed with the stroke
of a pen. Ironically, those who favor waiting for technical guidelines to be issued for a paper audit system are advocating the use of technology that has been shown to be error-prone and un-auditable.
Myth: "There has never been a documented case anywhere in the country where an electronic voting machine has produced an inaccurate tally of the votes."
Fact: This statement is misleading at best. There are many cases where e-voting machines appear to have RECORDED votes inaccurately, including the 2002 election in Wake County, North Carolina where 436 votes were lost because of a software bug.
The use of the word "tally" is perhaps a semantic trick, meaning that incorrectly recorded votes were then totaled correctly. If so, it misses the point that the vote totals fail to represent how the voters voted.
Myth: E-voting machines are not computers, so they are not subject to problems of computer security.
Fact: Florida Secretary of State Glenda Hood actually said this. It is a totally incorrect statement. The most widely used models machines have the same microprocessors that are used in PCs. By any reasonable definition, they are computers that execute computer programs, so they are subject to the same hardware and software bugs, and the same security issues, as all other computers. Just because they don't normally have a keyboard or mouse attached does not mean that they are not computers.
Myth: Printers are unreliable; VVPB printers will jam and/or run out of ink or paper, causing long lines at the polls.
Fact: High-reliability printers are available and deployed in many applications, e.g., ATMs, self-service boarding pass printers for airlines, etc., and printers print thousands of receipts everyday with little difficulty. While such VVPB printers might occasionally require attention by a poll worker, the same is true for e-voting machines.
Of course, optical scan machines don't require printers.
Myth: It just costs too much money to print paper ballots. E-voting will save money.
Fact: The purchase price of e-voting machines is three times as much as precinct-based optical scan. It will take at least 15 years of ballot printing costs to make up the difference. Also, the operational costs of e-voting machines are often underestimated. Some jurisdictions have found that they needed more poll workers to conduct an election with e-voting machines (e.g., San Diego needed twice as many!). There are increased costs for equipment maintenance and storage. Testing is more expensive, and so on. | 
