The New York City Board of Elections approved on Tuesday a new process of reporting election night vote totals that it said would provide the public with faster and more accurate results.
Under the current procedure, at the end of the night, poll workers tally the votes from each ballot scanner by hand on Canvass sheets. Police officers take copies of the Canvass sheets, they do not have the software to read the optical scanner's memory sticks, back to their precincts, where officers enter the totals into the Police Department’s computer system. The police transmit the results to The Associated Press, which then shares them with other news media outlets.
Under the new procedure, the police officers, instead of collecting the Canvass sheets, will collect the memory sticks from each ballot scanner and take those back to the police precincts. There they will hand the memory sticks over to Board of Elections staff members, who will download the results into laptop computers and then transmit the data to the Police Department — which will, as before, share them with The Associated Press.
The commissioners said the new procedure, which they had claimed was not permitted under state election law, would avoid the inaccuracies that plagued the initial vote count in last month’s Congressional primary and prompted criticism of the board’s practices. (The City Council has scheduled a hearing on August 8th to determine what went wrong in the initial count.)
The meeting on Tuesday, in a cramped conference room, was not without contention. At one point, a co-chairman of the State Board of Elections, Douglas A. Kellner, suggested that the city’s process of producing official results on election night which requires poll workers to cut and collate many pages of printouts from the scanners was “unnecessarily complicated.” One copy of the printouts are posted by each scanner for the different poll monitors to record the totals for their candidates.
That prompted an outburst from one of the city election commissioners, Juan Carlos Polanco, from the Bronx. “I object to that,” Mr. Polanco said, adding that the board had “interpreted the law in a manner that the counsel and commissioners thought was appropriate.”
Mr. Kellner interrupted him. “Commissioner, with all due respect, the state board has also interpreted the law,” he said, and had reached a different conclusion.
Mr. Kellner added that the city’s procedure was not illegal, simply inordinately cumbersome, “so there’s nothing we can do other than recommend that you consider a simpler procedure.”
I have issues with this approach just to satisfy the media's need to report, as quickly as possible, initial counts. Why not take the sticks to the Board of Elections offices were they have the equipment to read the memory sticks. With this new process, the voters' will have to pay around $300,000 to buy new laptops, ballot count reading software, and the encription network software to transmit the memory sticks information. Security experts will not be happy with this option.
There is time before the next primary, September 13, to improve this process.
This is what I wrote to Council member, Gale Brewer, who is conducting the City Council meeting:
1. Poll Monitoring - Any new system needs to allow the monitors to do their jobs. So local printing of the tapes will still be needed so we can record our candidates numbers directly from the scanning units.
2. Chain of Custody - You still need to print a copy of the tapes from the memory sticks, before they are sent to be read by the next process, now NYPD. Security experts will want some type of verification that the memory stick was not tampered with before it is used in that next step.
3. Canvass Sheet - I agree that by using the memory stick you do not have to fill this in but use the polls printed tapes if there is a question. I have seen some of the canvas sheets used in the Rangel/Espaillat race. No poll supervisor I have meet would have put their signature on the data entered on those sheets.
No matter what new system you use, without the desire to do a great job, human error will prevail.
There are other problems with the Rangel/Espaillat race vote counting.
More than 500 votes in the controversial Democratic primary contest were never counted for any of the candidates. A review of official precinct-by-precinct results for the 13th Congressional District shows that electronic vote scanning machines the Board of Elections has used for the past two years failed to record any voter choice on 436 ballots. Those nullified ballots represent 1% of all votes cast in the race — a significant figure, given that Rangel won by only a 2% margin. The BOE already knew about a possible heat problem with units used in that district. The Board of Elections discarded another 78 write-in votes as “unattributable” to any candidate the review found. It defies logic that 514 people went to the polls in this hotly-contested race and voted for no one.
The biggest number of both “unrecorded” votes (104) and “unattributable” write-ins (20) came in the 72nd Assembly District in Washington Heights/Inwood, where insurgent candidate Espaillat had the most support. The nullified ballots by themselves would not have changed the final results, but they add to a string of troubling questions about how the Board of Elections ran this primary.
In several precincts, the counts were off by hundreds of votes. That error rate was far higher than the 9% rate the Board claims is normal.
Meanwhile, the civil rights group Latino Justice/PRLDEF has asked for a federal investigation of possible voting rights violations against Hispanics. The group sent a second letter this week to the Justice Department detailing accounts from more than 60 Spanish-speaking voters who claim they were not provided federally-required language assistance while trying to vote at the polls.
NYC Wins When Everyone Can Vote!Michael H. Drucker
Technorati Tag in Del.icio.us 
