Maryland: Diebold E-Poll Book Modifications Come Into Question
By John Gideon, VotersUnite.org
October 15, 2006

Diebold's new ExpressPoll 2000/4000 electronic poll book failed miserably in the September Maryland primary election. The poll book is separate from the Diebold TS voting machines used in the state. It contains all of the information on registered voters that was contained in the old paper poll books and it is used to check in voters. The failure of the e-poll books in Maryland, as well as the manner in which the problems have been “fixed” raises serious concern about whether they can legally be used in November’s elections.

According to the Associated Press a software patch produced by Diebold fixed one problem that plagued the equipment in Maryland's primary. Another problem, reported by the Baltimore Sun (article in archives) required a software patch that was produced by a Diebold sub-contractor, Advantech Co., Ltd., who was responsible for the e-poll books "losing synch".

The Sun goes on to say:

"The poll books contain a trove of information about each of Maryland's more than 3 million registered voters, and were designed to replace the cumbersome alphabetized binders filled with the same data. When a voter signs in at a precinct, the system marks him or her as having voted.

"When the machines stop talking to each other, different poll books at a given precinct might not agree on how many people have been checked in to vote.

"Underwood [Ross Underwood, the director of the ExpressPoll Division of Diebold] said Diebold was aware that Georgia, which uses an earlier model of the e-poll book, had similar problems during its primary in July, but had thought the problem was in the earlier model's hardware.

"Since Maryland reported its problems, Advantech concluded that was not the case.

"A solution to the third significant flaw likely will involve affixing a piece of polyester film [mylar] to the machine's hardware, so that a computer card that is programmed with a ballot for each voter makes sufficient contact with the e-poll book so that it can be configured."

According to the Associated Press article, Maryland Project Manager for Diebold, Thomas Feehan contradicted Underwood and said:

"The problem did not develop in Georgia, the only other state to use the machines statewide, because it did not use that field, Feehan said. When software was changed to bring up that field in Maryland, the company did not discover the flaw because it did not test enough names on individual machines, he said. "

"I would say it was an oversight," Feehan said."

The e-poll book is a federally qualified part of the overall voting system used in Maryland. Maryland requires all voting systems to be tested by an Independent Test Authority and qualified by the National Association of State Election Directors (NASED). [MD HB 1457 – 2001, requires that voting systems be tested by ITA to FEC standards – 1990 at that point – and approved by NASED).

The process of federally certifying voting systems is being taken over from NASED by the Election Assistance Commission (EAC), as mandated by the Help America Vote Act (HAVA).At a public hearing in Santa Fe in July, the EAC announced that all new voting systems or modifications to voting systems that began the testing and certification process prior to 10 July 2006 are still qualified by NASED. All system modifications submitted after 10 July 2006 are the responsibility of the EAC. No new voting systems can be federally certified if they began the process after 10 July and that will be the case until after the election in November.

The failure and resulting modification of these pieces of election equipment requires many answers to many questions. Were there problems with the e-poll books in Georgia and if there were why weren't repairs made to the machines in Maryland before their primary? What about the rest of the states that have counties using Diebold E-Poll Books? Will repairs to their equipment be left until after they too fail? Do those jurisdictions even know there is a problem and has Diebold stepped-up to make repairs?

For Maryland's part in this; why did the state not question Diebold about certification of the repairs to the e-poll books? If they did question Diebold then why did they knowingly violate state law? Is it normal business procedures for the state to just allow Diebold to do anything they want with the state's voting equipment?

And about those repairs; two software patches and a mylar addition to the hardware. Why were these modifications made without any ITA testing and federal certification? Even though there are no direct Voting Systems Standards for the e-poll books they are a certified portion of a complete voting system and making modifications to them modifies that system. Why, when a phone call was made to Sandy Steinbach, the chairwoman of the NASED Voting Systems Technical Panel did she not know anything about any problem? Why was I sent to Ms. Steinbach by the EAC when it is clearly a modification devised and made well after 10 July 2006? This mess belongs to the EAC and it is their responsibility to clean it up.