Teresa Hommel

December 22, 2009


Commissioner Frederic M. Umane, President

Commissioner Julie Dent, Secretary

Commissioner Jose Miguel Araujo

Commissioner Juan Carlos “J.C.” Polanco

Commissioner James J. Sampel

Commissioner Nancy Mottola-Schacher

Commissioner Naomi C. Silie

Commissioner J.P. Sipp

Commissioner Gregory C. Soumas

Commissioner Judith D. Stupp


Dear Commissioners:


In evaluating the two state-certified scanners, I urge the Board to examine and compare the types of errors in ballot programming that are allowed to be made by the Election Management System software of each vendor. This software will be used to program every precinct-count scanner that the city purchases.


It is common for software to help people prevent errors, and we should expect well-designed software to prevent people from making known types of errors. For example, when you are creating a document, a small popup box may appear with the question "Do you want to save your document?" When you send email, the software will not allow you to send without specifying who the recipient is. Software often asks you “Are you sure?” before performing functions that could cause the loss or corruption of data.


Apparently Erie County's ES&S Election Management System allowed a serious error to be made in programming their scanners for the November 3 general election.


Transcript of New York State Senate Standing Committee on Elections, November 12, 2009,

Page 13-14, Testimony of Douglas Kellner, Co-Chair, New York State Board of Elections



there is one other scanning issue that had come up with the 9th Legislative District [of the County Legislature] in the Town of Cheektowaga in Erie County where the machine had not been set up properly.


... the ballot -- the election management system was improperly programmed so that the scanning results did not accurately report the results on the ballot.  ... even though there were two different candidates' names, the persons who had programmed the machine had marked those candidates as the same candidate so that the votes were counted as if you were voting twice for the same candidate.


I have not been able to clarify the exact nature of the ballot programming error. However, we should expect our Election Management System ("EMS") software to prevent obvious, predictable errors that can occur when election staff are entering candidate information for preparing ballots and tallying votes (this is called “Election Configuration” in the State Board’s Regulations, Part 6209).


Last week I asked both vendors to bring their EMS software to the public demonstrations on a laptop and to show Howard Stanislevic and me how the software would prevent errors such as tallying votes for the wrong candidate.[1] Only Dominion did so, and showed how errors in attributing votes to the wrong candidate was prevented by the software.


ES&S said that they would be unable to get a copy of their EMS software that was certified by New York State, but they would demonstrate a different version of their EMS software. At the demonstration Monday, Dec. 21, in Manhattan, however, they had a laptop but said their EMS could not be loaded and so they could not show it to us.


In evaluating the two scanner systems and their EMS software, the ability of the EMS to prevent predictable human errors or malicious election configurations is an important factor. I hope that the Board's evaluation team can obtain the information that we members of the public were unable to obtain. At this time, we cannot determine whether the State Board of Elections has performed similar testing or evaluation, but the problem in Erie County suggests that such testing was not part of the certification program.


Respectfully yours,


Teresa Hommel

Chair, Task Force on Election Integrity, Community Church of New York



Marcus Cederqvist, Executive Director

George Gonzalez, Deputy Executive Director

Pamela Perkins, Administrative Manager

Lucille Grimaldi, Electronic Voting Systems

Steven H. Richman, General Counsel

Valerie Vazquez, Public Affairs and Communications



[1] Stanislevic has advised members of the University of Connecticut’s Voting Technology Research Center, who have demonstrated a number of sophisticated attacks on that state’s optical scan voting system that could be used to bias election results accidentally  or maliciously. U. Conn’s computer scientists were able to: manipulate the count so that no votes for a particular candidate are counted, thereby “entirely neutralizing one candidate”; swap votes for two candidates; report the results incorrectly by shifting some votes from one candidate to another, etc. See: http://voter.engr.uconn.edu/voter/Reports.html