 Update: View Hearing at <CT-N>
There were about seventy five people at the West Hartford Government Administration and Elections Committee hearing. It was standing room only, with just a handful need to stand. Unlike the other hearings it was taped by CT-N.
The vast majority of attendees were registrars, moderators, and other election officials. It is hard to believe that this was not an orchestrated turn-out. In fact, before the hearing began, one disappointed registrar left – he told us he had been told it was some type of training, he had not expected a hearing – he had changed his plans to come, cancelled a meeting, and was leaving to go to another obligation.
Training also came up several times in the hearing. The registrars want more training for themselves and other election officials. The registrars and members of the public suggested more training for voters. One problem though was the problems with finding election officials willing to work for $8-$10 per hour for the fifteen plus hour election day and that many are recruited too late for any training.
The first speaker was Alyssa Fukes of Hartford (I will do my best at names here, and welcome any corrections). Alyssa moved from Georgia to Connecticut sometime after the 2004 election. She criticized Connecticut for using optical scan – she much preferred the touch screen voting machines in Georgia and felt they were much more modern and voter friendly than our optical scanners. Earlier, on the way in she had confronted us for suggesting that optical scanners were a good choice.
Next up was Eleanor Brazel, Registrar of Voters, West Hartford. She stated that as long as the count of voters on the optical scanner tape matched the number of voters checked in there was nothing to worry about. (Ignoring or unaware that many electronic voting errors or frauds, such as the Hursti Hack, involve switching votes between candidates or adding undervotes which do not require changes in the total votes) . She argued against audits if the number of voters checked-off equaled the number on the scanner tape. She also mentioned that the costs using optical scan tripled over lever machines. Other election officials repeated the same claim later.
Note: We would be surprised if costs have tripled. A possible area for future research would be to check the budget and expenditures of several towns. Here we have a technology that was paid for by Federal Funds, audits paid for by Federal Funds, LHS employees paid for by Federal Funds replacing machine mechanics paid locally. We do have the cost of $0.35 per voter for paper ballots. We do have the IVS which can be costly. Yet we don’t have the old lever machines to lug around, and while we do have to do pre-election testing, each of the lever machines used to require setup/programming in each town.
We do have one example on file: Greenwich. Costs doubled (not tripled) for the primary from $45,000 to $90,000 but the Greenwich Registrars attributed the cost increase, not to scanners, but to democracy, because there were two parties having presidential primaries not one. <read>
Ms. Brazel also mentioned the weight of the optical scanners and the burden transporting them added to other duties of aging election officials. I’m no expert, but it just seems that the old lever machines certainly looked, to my untrained eye, to be quite a bit heavier than the new scanners. Then there are options, even though we have questioned the expense of moving companies, it is a ongoing alternative that seems to have been employed for years. <read>.
On another topic she characterized the idea of election day registration as “insane”.
Over the course of the hearing several registrars echoed her complaints about the audits, the tripling of costs, the IVS, and the lugging of heavy machines and paper ballots.
Next up was Alan Hoffman, Town-wide Moderator, for West Hartford. (I’m not sure but it seems that town-wide moderators perform their duties centrally with the registrars on election day.) He said that he was not a fan of the optical scanners. He said there were new wrinkles each time. He made some suggestions – that the machines make noise when votes are scanned to give voters comfort like opening the curtain on the lever machines. That the Secretary of the State should test them in the real world, perhaps with school children and get their feedback.
Several other officials and citizens echoed his suggestion of a noise, such as a “ca ching” etc. Testing in the real world is a good idea – I might not suggest school children since they are all pretty good at filling in ovals for tests, they might be better at it than the average citizen and less enthusiastic about taking another test. Perhaps an interesting suggestion, but what would it cost and is there a certified voting machine available than makes a “ca ching” or some other fitting sound?
William York, West Hartford election official. Liked the “ca ching” idea, suggested it would solve the problem of the number going up too fast on the LCD and the voters questioning if it advanced.
Michelle Jacklin, Glastonbury Voter. Michelle criticized the under purchase of ballots and the many citizens that had to vote on paper without the assurance of both a paper record and electronic counting.
Denise Weeks, CTVotersCount and software testing expert was next. (We expect to have her testimony to publish here soon, so we will be brief). She testified to the need for hand counting of ballots, continuing the audits, the inadequacies of the November audits as conducted, the lack of follow-up to discrepancies found in the audits, memory card problems uncovered by UConn, the risks of programming errors, and the risks of fraud.
She emphasized that memory cards are programed anew for each election; that errors could occur in any election without regard to previous accuracy; that many ways to compromise elections had been discovered and verified by groups such as the Secretary of the State of California, the Brennan Center, and the University of Connecticut; that among those identified were many clever ways of circumventing pre-election testing.
After her testimony the committee had several questions about solutions. She outlined the advantages of in state programming and independent pre-election testing of all memory cards; the advantages of centralized auditing to answer cost and drudgery concerns of election officials while providing the means to support more efficient and useful audits, such as quicker “Hot Audits”.
Next was a registrar for Southington. (I do not have his name). Before the hearing began, he told us in the hallway, that if there were problems with the machines they should be fixed secretly so that the public would not be worried.
He said that all went well with the 1st optical scan election in Southington in 2006. But, 2007 was another story with the public worried with the concerns “planted in their minds”.
He stated that all errors in the recounts in Southington were the result of hand counting errors. They hand counted and there was a discrepancy between the machine and the hand count. They counted with another machine and it equaled the first machine. He therefor concluded that the problem was with the hand count. Under questioning from Rep Caruso, he added that all the counters agreed that they must have counted incorrectly.
He did say that if there was a better system for hand counting he would be willing to give it a try. He said that counting votes was like count $5.00 bills – they stick together. (I wonder how banks solve the problem – they have been counting bills accurately for years.)
Next was Gail Stempien, Assistant Registrar Simsbury. She testified that the Voter Registration System was a bad example of programming in Connecticut; the voters like the optical scanner; and that the IVS was the worst part of the process. The Voter Registration System is actually an example of outsourcing, the fallacy of which, I have covered elsewhere <read>.
Next was Machael Aron, West Hartford Voter. Michael supported the idea of the noise when a vote is scanned. He emphasized the chain of custody aspects – counting and tracking blank, voted, and spoiled ballots. He suggested a machine that would count and dispense votes one at a time. Such a machine could be balanced against the checkin list and the count on the scanner. I’m not sure this particular approach would provide much added security, but certainly there is enhanced security needed in accounting for all ballots.
George Lewis, West Hartford Election Official, testified that he was surprised at how many voters were pleased with the optical scan voting machine. He pointed to problems in privacy in the machine handling spoiled ballots and the opportunity for election workers to see how a person voted on the spoiled ballot. Shocked to learn (I presume from earlier testimony) of such a chance of fraud.
Steve Kovac, Moderator West Hartford, testified to the long hours, low pay, and lack of training for poll workers. Part of the problem is the difficulty finding workers and the need for appointing them close to the election because of that difficulty.
Judy Aron, West Hartford Voter, gave a hearty thank you to the registrars for all their hard work. She expressed concern with the integrity of absentee ballots, especially those not completed by nursing home patients that were no longer able to vote at all. She also brought up the issues of wet ballots, ballots, and the margins of error we have seen. She expressed concern with the tripling of costs to municipalities.
Gary McBride, Election Official West Hartford discussed privacy concerns and that there were virtually no problems in the optical scan elections.
Walter Galsco(?), Election Official West Hartford, stated that the machines worked very well and that people need to get used to them. He said we were safe because all poll workers take an oath. (I am glad that they do, but am reminded that Richard Nixon, Mayor Ganem, and John Roland all took oaths.). I also note that it was West Hartford that reported minus three questionable ballots in one race in the audit.
An Election Official from Montville, recalled the 2006 optical scan election where they were one of the twenty-five towns. She said there were no problems. (I recalled the questions in Montville surrounding the integrity of the recount in that election, due to unauthorized actions by LHS, uncovered by Dori Smith. <read>)
Mike DeRosa, Voter and Candidate. Mike recalled an original promise by a former GAE Chair of a random audit of all races in 20% of districts, later replaced by three races in 10% of districts. He said that the “proof of the pudding is the proof”. He also pointed to an error in programming of a cross endorsed candidate which resulted in a candidate in his party (not involved in the cross endorsement) being listed as “unknown party”. He joked that his Green party was unknown enough already.
Martin Valuscka, Election Official for 20 years, loves the new machines, but not the IVS. He echoed the lack of training. He said the optical scanner suddenly started rejecting ballots (perhaps a cause similar to the misscut ballots mentioned by a registrar in the Norwich hearing). They almost ran out of ballots, which would have kept voters waiting while they restocked from the town. He also said that he now sees the potential for fraud.
Darlene Burrell, Registrar of Voters Suffield. Said she was a skeptic until the recount after the municipal election when all the discrepancies were explained. Believes the state has very good procedures, if they are followed – as they are in Suffield. They have locks galore. She questioned if the risks presented by the “computer person” had been proven. (Yes, they have if one believes the UConn VoTeR Center, the researchers hired by the Secretary of the State of California from Berkely and Princeton etc., the Brennan Center for Justice, the Carter-Baker Commission, and a huge stream of computer scientists in this country and Europe. Otherwise No.) (If we don’t trust UConn research then perhaps we should be talking to the committee that handles their budget.)
Jan Murtha, Retired Registrar South Windsor, currently voting system Trainer. She explained that the lever machines did not work reliably, that there were many problems and the registrars did not tell us – we did not know about this. (Recall the registrar mentioned earlier who told me problems should be fixed without worrying the public.). She said costs can be less with the optical scanner, a town (hers?) went from 7 to 5 districts, no longer pays a mechanic at $1000 per election, but that some cost for registrars went up because responsibilities and funds were transferred from the Town Clerk’s office. She said the Secretary of the State’s Office has done a “terrific” job of training. She was in favor of the double check of the machine, and recommended a recount by machine and the audit by hand.
David Wawer, Registrar Enfield, said he was a new registrar and only has experience with optical scan as an election official. The cost of ballots should be paid by the state, his budget for November was $10,000. He said a recount like East Haven’s would cost $5,000 (Note: that is half the cost of the paper). He suggested the SOTS do more to educate voters – said problem was worse in February with more new voters and less public education.
Carol Mulready, West Hartford Voter, expressed that she found optical scan voting easy, yet could see how others could have problems with polling place setup.
Joel Huntington, West Hartford (Election Official?) expressed a unique but interesting observation. He had seen three voters use the IVS system and that they each liked it.
(That is it for now. Once again apologies for people that I missed, points that I missed and misspellings or errors – I will gladly correct any brought to my attention).
