96.6% Of Memory Cards OK, Half Of Registrars Follow Last Minute Procedures

At the request of the Secretary of the State, Dr. Alex Shvartsman and his team at the UConn VoTeR Center tested memory cards which were shipped to UConn by registrars across the state. <full report>

From the summary:

The total of 522 cards were received and tested by the VoTeR Center, out of which 378 cards were received before the election. Out of the total number of cards, 18 cards, or 3.5% were found to contain “junk” data, that is, they were unreadable, which is easily detected by the tabulators as such, and could not have been used in the election. The rest of the cards, or 96.6%, were found to have been properly programmed for election. These cards contained valid ballot data and the executable code on these cards was the expected code, with no extraneous data or code on the cards.

About half of the cards were found to have been tested and set for election-the intended state of the cards following the prescribed testing procedure. Most of the remaining cards were tested by the towns but not set for election; while this is not a problem, this suggested that the relevant towns/districts either misunderstood the instructions or did not follow the instructions. One card was found in the state set for election but with non-zero counters, indicating that the district tested he card in election mode and did not reset the card. This is a potentially problematic, but detectable situation, since proper procedures require that the “zero counter” report is produced at the start of the election

Full Disclosure: I played a role in suggesting this program to the Secretary of the State’s office at a meeting in October. Professor Michael Fisher of Yale and TrueVoteCT greatly improved on my suggestion.

I never thought or intended that this could be accomplished by the November 2007 election, my thoughts were directed more toward November 2008. So, I applaud the Secretary of the State’s Office for moving this forward and Dr. Shvartsman and his team for implementing it so quickly. I also have great sympathy for the registrars because the program was added to their many new responsibilities shortly before the election. I do not criticize the 14% who did not or could not ship cards to UConn; those that shipped them after the election; those that did not follow the procedures exactly. Lets work the kinks out of the process and improve it over time.

Commentary:

That 3.5% of the cards were unreadable is totally unacceptable in as simple a piece of computer hardware as a memory card!

The registrars can be excused for not testing the cards prior to sending them in. Diebold cannot be excused for selling a product ($250 per card) with such a failure rate. This is not an anomaly, it fits with what was uncovered in Florida. <read: Dori Smith’s excellent report on CT and FL>

LHS cannot be excused for allegedly programming and shipping memory cards that cannot even be read.

While I have tolerance for registrars not understanding procedures added at the last minute, it is incorrect when the Secretary of the State is not concrned and does not find the 3.5% failure rate significant.

From a November 25 New York Times article previously covered by CTVotersCount <read>

Mrs. Bysiewicz said the pre-election audit of the cards found that none of the cards had been tampered with, but that a handful were blank, meaning that they had not been programmed with the proper ballot information.

One more thing, since half of the registrars did test the cards before sending to UConn, perhaps another 3.5% were detected and replaced prior to the UConn test. So maybe 7% had “Junk Data” and at most 93% were OK.

Would we accept 93% of our sewer and drain pipes working? Would we accept 93% of trucks being safe to descend Avon Mountain? 93% of parolees not committing serious crimes shortly after their release? 93% of air traffic control computers working on a particular day?

Update: Just in from the Greenwich Time, excellent article by Neil Vigdor <read>

Secretary of State says safeguards for computer voting are strictest in nation

“We have many checks in place to ensure the integrity and security of the election process,” Secretary of the State Susan Bysiewicz said in an interview.

Bysiewicz said she had no reservations about using the fax-like machines known as the Accu-Vote system,…

Under Connecticut law, election results from 10 percent of all voting precincts in the state must be audited, a mandate that was put in place when the switch to the new machines was made.

“I believe it’s the strictest in the country, which calls for manual hand-counting of paper ballots to ensure that the machines are counting the votes accurately,” Bysiewicz said…

In addition to those measures, Bysiewicz said the state requires local registrars to randomly select a memory card from one machine prior to an election to send to computer science researchers at the University of Connecticut for testing to make sure it is not corrupt. The cards also get sent to UConn after an election for more testing, she said.

As we know from the UConn report the cards were not handled exactly as the Secretary of the State seems to believe and we have less faith in those “strictest in the country” audits: <read>