Continuing Concerns with Hawaii’s Election System
The Office of Elections (OoE) has been embroiled in some controversy in recent weeks and a number of people have asked my opinion on the various issues, so I am attempting to put down my thoughts, however fragmentary and half-developed, so that others can pick them over, perhaps finding an idea, fact or insight worthy of further development. Or, worthy of rejection after consideration.
[For those who may not know, I have been an official election observer for the Democratic Party for the last two election cycles and will be one again this year. I was one of the founders and core activists for Safe Vote Hawaii, an election integrity group best known for initiating and getting passed Hawaii's "Voter-verified Paper Audit Trail" bill, which requires that any electronic voting machine used in Hawaii provide a VVPAT, which can be used to verify the accuracy of the electronic record of the votes cast. I have also attended several Election Commission meetings over the years and have monitored, as best I could through the official secrecy, the last three procurements of voting systems for Hawaii. I have come to be the "point man" for the state Party on election system matters.]
I do not pretend to have a solid grasp on all the complexities involved. In fact, a BIG complaint I have is the unnecessary secrecy surrounding many of the operations of the Office of Elections which requires tactics akin to industrial espionage or “intelligence gathering” to find out information which SHOULD be readily available for public consideration and comment. We should not have to be relying upon “leaks” to know what the hell is going on with our elections. Nor should people have to rely upon the suppositions and poorly-sourced reflections of a very small handful of “election integrity” volunteers like myself. Confidence in our elections system is essential for many practical reasons, not just fuzzy “feel good” ones.
First off, it is imperative that voters, candidates and parties have complete confidence that the selection of legislators was fairly determined. Rumors of vote fraud are extremely corrosive. Some folks want that confidence to be based upon trust. I want it based upon hard evidence that the elections were fairly conducted. I want a voting systems that could be run by the most ruthless and corrupt thugs, but which has enough transparency and safeguards so the crooks would be unable to steal an election.
Please note: I am NOT saying the OoE staff, or the employees of the voting machine vendors (in this case, Hart-InterCivic) are crooks. I want a system with enough safeguards where they COULD be crooks and it would have no impact due to the safeguards we put into place and the technologies used. Because we are not at such a place, I think it is understandable that the more suspicious voters might easily succumb to suspicions of the integrity of the people actually running the elections. Let’s prove the suspicious people wrong.
Among the recent controversies:
ISSUE # 1: The withdrawal of Kirk Caldwell from the House race, the disqualification of both Kirk and Chrystn Eades and the selection of Isaac Choy as a replacement.
I think Chief Elections Officer (CEO) Kevin Cronin was prudent to decide that Kirk’s withdrawal was not official IN THE EYES OF THE OoE until they received it in writing. I believe they were also correct to initially accept Eades’ papers, but then rule they had not been completed prior to the filing deadline.
This triggered the 3 day deadline for the Party to come up with a replacement candidate to “fill the vacancy.” The Party did not receive notice of this ruling until mid-afternoon on the Friday, leaving them just over 24 hours to assemble an emergency meeting of the 24th District Council, solicit potential candidates and have the Council reach a decision.
I am inclined to disagree with the City Clerk’s decision that she must be bound by the OoE’s decision on when Kirk formally withdrew. Since her office had had direct conversations with Kirk at the time of his filing for the Council race, she could have reached her own determination that Kirk had withdrawn from the House race prior to filing. The OoE may need it in writing for them to have confidence, but the Clerk’s needs were not the same. Most people are unaware the Clerk did NOT have the power to rule Kirk’s filing as invalid. Under HRS, she had the power to go to court to challenge his filing, but it would have been up to the Court to make that determination. By this point, Kirk, in order to distance himself from any taint, was playing the gentleman and had agreed to abide by the Clerk’s decision, though it was NOT legally binding on him.
The law does NOT specify that a candidate’s withdrawal must be in writing. The written requirement is only explicit in the cases where a candidate withdraws for health reasons The OoE does, however, have a standing policy, reflected in its information sheets, which does require a written withdrawal. The statutes should be amended the next legislative session to clarify the process–probably require a written withdrawal, perhaps on a form made available at the various filing offices.
Because of the chaos provoked by the last minute filing of Ann Kobayashi for the Mayor’s race, opening up her seat and the way so few people were aware in such a short time of the openings created by the musical chair chain reaction, several proposals have surfaced to prevent these problems in the future. Among the best, set an earlier deadline for candidates to draw filing papers, perhaps two or three days prior to the filing deadline. Extend the deadline when an incumbent does not file for re-election–perhaps for a week. In the Manoa case, Kirk’s withdrawal from the House race created a mad scramble for people to file. There appears to have been some cooperation between Kirk and Mayor Hanneman, yet Hanneman’s aide, Eades, was still unable to get her papers completed in time for the deadline. Potential candidates should not have to rely upon access to advanced inside information in order to have a chance to file under such circumstances. There are a number of well-qualified potential candidates who briefly considered running when they heard of the potential opening, but recognized the decision, and filing, required more time than they had available.
Extending the filing deadline for a week would allow all residents in the district a fair chance to file and would be an improvement over granting the political parties a chance to name a replacement. I say this even though I believe the Party acted honorably and fairly in their selection of Choy. I just wish the process could have been more open to all those interested.
ISSUE #2 : The decision of Chief Elections Officer Kevin Cronin to order the ballots be printed without allowing the political parties an opportunity to review them first AS REQUIRED BY LAW.
This was a screw up and reveals a fundamental misunderstanding by Cronin, and the Office of Elections generally, about their role and the role of the public, candidates and parties in the conduct of the elections. The various stakeholders MUST be consulted and agree to the fairness at each step of the way. Cronin makes a BIG mistake when he fails to get a “buy-in” from the parties on these matters.
The GOP complaint that the ballots might have to be reprinted if the challenges to Choy and Caldwell prevailed were groundless. The deadline for potential challenges is long after any prudent deadline for sending the ballots to the printer, so the printing cannot wait until all challenges have been resolved. There is a workable remedy available when a candidate has been removed from the ballot: a notice is posted at each polling place and inserted into the absentee ballot envelopes. Even if a voter fails to learn that a candidate’s names has been stricken, what is the harm? A few wasted votes that would otherwise not be cast?
The Democrats have a concern with the ballot design, beyond the mere correct spelling of names. This year, for the first time, voters will be asked to state a party preference on the primary ballot. Among the parties listed is the “Independent Party.” The Dems have two concerns. First, a lot of Hawaii voters consider themselves to be “independent”, even if they generally vote for one of the two major parties. We fear there is SOME risk that voters may misunderstand the instructions, select “Independent” but proceed to vote exclusively in one of the major parties’ primaries. Such a mistake may void all the voter’s partisan votes. It is also unclear what would happen should a voter neglect, or refuse, to state their party preference in this way. Would all the votes cast in the primary be ignored?
This ballot design feature was clearly ported over from the Hart software from their eScan eSlate DRE voting machine. The question served a valid purpose in that context. After indicating an intention to vote, say, Democratic, the voter would only be presented with Democratic ballots, preventing the possibility of a crossover mistake. But when applied to paper ballots, the same safeguards no longer exist. Each voter will still have access to the paper versions of all parties, so the crossover effect is not protected against and is still possible.
Ideally, the software on the voting machine would be intelligent enough to only rely upon the expressed party preference in those occasions when a voter casts votes in the primary of more than one party. I suspect it is NOT set up that way, however.
Cronin’s decision to print the ballots without first explaining the ballot design to, and winning acceptance from, the political parties was a mistake, as well as a violation of state law.
ISSUE #3: The award of the voting system contract to one vendor, Hart Inter-Civic for $52 million, when another qualified vendor, ES&S, was offering a similar service for $18 million.
This is shocking and needs to be better explained to everyone concerned. If the reasoning is valid, perhaps it can be accepted. But on its face, it is a continuation of a biased, sweetheart relationship with Hart that I have observed for the 3 procurement cycles I have been watching the Office of Elections. The hearings officer from the State Procurement Office blasted the award as proceeding from “bad faith.”
This matter has been discussed at some length and I do not have the time right now to go into depth, but one point I would like to emphasize is the “bad faith” of the OoE in chosing Hart. When Safe Vote Hawaii was lobbying for passage of the “paper trail” law at the legislature, we were opposed by the Office of Elections. Their reasons varied, but a recurring argument they made, and an argument they have continued to make, is that they opposed the paper trail because it would not have been “equally accessible” to blind voters. Let me make it clear what they were saying. If the small number of blind voters who choose to vote in the polling places were not able to read the paper printout and then compare it against the onscreen electronic record, no voters should be able to do so. The OoE had long demonstrated their inability to understand the function of the VVPAT, so it is not totally surprising that they would raise this objection. After all, it makes it SOUND as if they are compassionate defenders of the rights of blind voters. But the argument is weak for many reasons, the MAIN reason is that the benefits of the VVPAT do not go to the individual voter. Not all voters have to check the paper printout in order to have it serve its function. In fact, it only requires a small percentage of voters check the paper to ensure a high degree of confidence that the electronic vote is accurate. The people who do check, provide a statistical sample of voters, which need only be a small percentage to be effective at catching errors. So ALL voters, blind or sighted, benefit when a small number of voters check the ballots.
OK, but it sounds insensitive to to adopt a technology that is not “equally accessible” to blind voters. So Safe Vote Hawaii dedicated ourselves to try to find a verification technology which WOULD be accessible to verification by blind voters. We came across the AutoMark, electronic ballot-marking system, which would provide blind voters (and those with other disabilities) the same accessibility features (headsets with audio menus, etc) as provided by the “touchscreen” DRE voting machines, but it would only MARK the ballot for the voter, not tally the results. The blind voter, or any user of the AutoMark, would use the same cardstock ballot as those using the optical scan voting machines. It would be marked on the AutoMark, but fed into the same optical scan machine as all other ballots. This provides the additional benefit of ensuring anonymity for the relatively few ballots cast on the handicapped accessible voting machines. More ino on the AutoMark can be found on the ES&S website HERE.
If a previously marked ballot is inserted into the AutoMark, it will tell the voter it has already been marked and offer to read the votes back to the voter through the headset! Here is a device which would allow a blind voter to “independently verify” the accuracy of the VVPAT! Here was a machine which overcame the core objection (excuse) used by the OoE for opposing the original VVPAT requirement. You would think they would be delighted! But when the OoE conducted their “objective” quantitative evaluation of the competing voting system proposals, they awarded no extra points for this ability of the AutoMark when offered by ES&S for use in our elections! Nor did they take off any points from the Hart DRE machines for NOT providing equal accessibility to the VVPAT for blind voters!
What gives? Was the equal accessibility really a concern for the Office of Elections when they opposed the VVPAT statute? If so, why did they drop that concern when offered a machine which was finally capable of satisfying their objection? Why was it no longer a concern when their favored vendor, Hart, was not able to satisfy their demand?
For this election, the SPO will allow the OoE to use the Hart machines, both DRE and optical scan. But the contract will end on Dec 31st of this year and the procurement cycle will have to be restarted. Hopefully, the process will be less biased this time, but that will require a great deal of scrutiny from the media, the parties and the general public. In some locals, once the Request for Proposals (RFP) has been drafted by election officials, it is made available for public inspection and comment. This allows concerned citizens to raise their issues/warnings in advance. Had this practice been in effect in Hawaii, the delays in the 2006 cycle could have been avoided. Hawaii should adopt this practice. If Cronin would institute such a change, it might help rebuild public confidence.
In some locales, the election boards invite the vendors to bring in their voting machines to a public meeting whereby citizens are able to get hands-on experience “playing” with the machines and offering feedback to the purchasing body. Hawaii, in contrast, conducts its procurement process in great secrecy. The excuse, another fraud, in my opinion, is that it prevents the various companies from learning about the details of the competition’s proposals. I call this a fraud because these vendors are extremely familiar with each other’s products and prices, because they are bidding against each other literally hundreds of times each election cycle. Hawaii is one of the few states where voting machine purchases are made by the state. In most places, it is the local county election boards who pick the voting systems, meaning that the companies are truly bidding against each other hundreds of times in a multi-billion dollar industry. They know with a high degree of certainty what software and hardware are being proposed, and they probably know the price point and profit margins as well. As is so often the case, the only ones kept in the dark are the members of the public.
The Office of Elections would go far towards restoring public confidence if they would conduct the procurement with much more transparency. We should work to change the procurement process to prevent abuses like what just happened. Sunshine is usually the most effective remedy for corruption or for the conditions which can easily give rise to suspicions of corruption. We saw similar problems with Ted Liu’s award of a contract to a high bidder with a DBEDT contract. That controversy elicited a senate inquiry.
ISSUE #4: Transmission of election results inter-island via the Internet
A lawsuit was filed on Maui to block this practice, which appears to be another violation of Hawaii law. The concern is that election results might be transmitted to an intermediate stop where the results might be “flipped” and re-transmitted. While this would require an “inside job,” I believe it is technically possible. But it is also relatively easy to guard against with simple safeguards. Prior to transmission, the results from Maui would be reviewed by the election observers, hard copies made and signed off on. If the results arriving on Oahu would be different, it would be rather trivial to catch any such hack.
The “flipping of votes” is a reasonable concern, but I think the suspicion that it would likely happen during the transmission is rather far-fetched. More likely, it would be done on the computer prior to transmission. IT security experts have demonstrated the ease with which vote tallies on central computers can be manipulated without leaving a trace.
ISSUE #5: Inadequate audit standards for the paper records.
When Safe Vote Hawaii helped pass the Voter Verified Paper Audit Trail law, we thought we had imposed a strict audit standard. The law required that 100% of the paper records in 10% of the precincts be audited to ensure the accuracy of their electronic counterparts. Unfortunately, we screwed up. We did not know that the Office of Elections only audits ONE RACE per ballot to check the accuracy of the vote! It is generally a statewide race. This is a ridiculously low standard! If there is a software glitch, or more ominously, a deliberate malicious hack, there is a high degree of likelihood it would be missed by such an audit. A more realistic audit would include many more races, but with a smaller number of ballots reviewed. (I would guess 2% might be sufficient, but the state could bring in statisticians to set the standard.)
IMPORTANTLY, none of the votes cast on the electronic voting machines during the extended period of walk-in absentee voting are audited! Software bugs usually express themselves under unforeseen conditions. The conditions of machines being used for several weeks, instead of on a single day might trigger a bug which would otherwise not occur. And if we fear for malicious hacks (intentional vote fraud) the machines being scattered across the state for so long would provide more opportunity for unguarded access. Or the software could be hacked so that the hack would only express itself during walk-in absentee voting and not during “normal” voting. This is somewhat akin to routinely leaving a backdoor to your house unlocked and not worrying if everyone in the neighborhood knows they can break in while you are out of the house.
CONCLUSION
Hawaii’s voters deserve accurate and fair elections which we can trust. To reach that point, the accuracy and integrity of the system used must be verifiable by the ordinary citizens who volunteer to help out. If the proverbial “little ole lady in tennis shoes” cannot vouch for the accuracy and security of the vote, the problem lies with the system being used to count the votes, not with the citizen volunteers. The current system is needlessly complex, with significant processes incapable of being monitored by the people charged with being the “eyes and ears of the public,” the official observers.
After the “hanging chad” debacle in Florida in 2000, it was perhaps understandable that there would be a call for a more reliable voting technology. Florida became an excuse for spending billions of dollars on over-priced and hurriedly developed voting machines, which have moved away from the transparency necessary for a democratic voting system. Public opinion in Hawaii, as across the U.S., has turned against complex voting systems and towards more trustworthy technology.
My own feeling is that traditional optical scan machines offer the best balance of verifiability, security, speed, price and efficiency. DRE “touch screen” voting systems, even with an attached printer, are a needlessly complex and expensive means for recording and tallying votes. They are not trusted and should not be trusted. The level of expertise and verification to ensure that code does not contain bugs or malicious hacks is beyond the capability of most election observers and even the staff of the Elections Office. Beyond that, the election officials in Hawaii are legally prevented from examining the code used in the voting machines, even if they did have the technical ability.
Open source software for DREs would be an improvement, as the pool of potential observers/inspectors reviewing the code would greatly expand, but I don’t believe voters should have to rely upon the sayso of elite experts. I think part of the glory of democratic elections is that they are conducted, and monitored, by our neighbors and fellow citizens.
While much of my involvement with these issues has been as a Democrat, not everyone associated with Safe Vote Hawaii has been a Democrat. I have discussed my concerns with Republican election observers as well, and while these writings/rantings/ramblings are my responsibility alone, I believe many election observers, regardless of their party affiliation, are in agreement with many of the concerns I have raised here.
People interested in more information may want to review the findings of the Ohio Secretary of State on voting systems here.
The California Secretary of State also has useful information. The CA SOS de-certified Hart’s voting machines at the end of 2007, only re-certifying them under strict conditions. Hawaii residents, or journalists, might well ask the Office of Elections if they are concerned about the vulnerabilities which California found in the Hart system and whether they will be taking active steps to institute similar safeguards in Hawaii. The CA reports can be found here. The specific comments on Hart can be found here.
For more information on Hawaii’s elections, go to the Safe Vote Hawaii website here.
POSTSCRIPT
Some voices have called for Chief Elections Officer Kevin Cronin to quit or to be fired. I am not ready to join that chorus. Cronin was hired in large part because he was not a part of the Office of Elections and, it was hoped, would provide a fresh start for the Office, which has suffered problems for a long time. The day before he took office, the outgoing acting CEO announced the award of the voting system contract to Hart-InterCivic. Cronin was in a difficult spot. He had little previous knowledge of the workings of elections and was dependent upon the advice of his staff on the contract. It was his staff who had pushed through the contract, though the original sweetheart bias towards Hart had started with Dwayne Yoshina.
Cronin is NOT good with the press. He should be more forthright in talking with them, but that may not be in his nature. Cronin should actively reach out to forge relationships with the official observers, the League of Women Voters, the political parties, etc.. He should seek advice from IT professionals from other state agencies as well as seek advice from related professional associations to set up the appropriate industry standards and ” best practices” for the use of IT and the auditing of the elections. If he bothered to reach out his hand, I suspect he would find friendly people reaching back.
The Office of Elections was deliberately detached from the Lt. Governor’s Office to ensure its independence. Unfortunately, the flip side of that is a lack of accountability. The bodies which can provide some oversight need to step up to the plate. Most especially, the Elections Commission has the legal authority (perhaps the “obligation”?) to hold public hearings on election related matters and to advise the Chief Elections Office. Instead, I have witnessed the CEOs, both Yoshina and Cronin advise them, never the reverse. I appreciate their fear of “micro-managing” the office, but they are the representatives of the public and have managed to exclude the public from direct knowledge of election activities by going into “executive session” at the slightest excuse, have truncated public testimony by restricting it to a “public comment period” at the start of the meeting, rather than in response top important items as they come up on the agenda. In my opinion, they are not doing a good job of looking out for the public’s interest in all this.
I recognize the commissioners are intelligent and well-intended people, striving to avoid the partisan bickering which has sometimes impaired the Election Commission’s functioning in the past. But they need to take a leading role in soliciting public concerns and getting the information, or establishing to procedures, to satisfy these concerns. Going into “executive session” in order to discuss every controversy with their lawyer, outside of the scrutiny of the public, is not only a violation of the state Sunshine Law, but it is a failure to rise to the occasion when circumstances require it.
And, NO, I am not calling for them to step down, either. If anything, I am calling upon them to “step up”!

![[PDA - Heathcare NOT Warfare - Sign the Petition.]](http://pdamerica.org/images/ads/HealthNotWar_final.jpg)
Thanks for spending the time to write this up!
Issue #1: I agree with your analysis
Issue #2: Ballot design is such a crucial issue. The infamous “butterly” ballots used in Florida for the 2000 election, come to mind. A good review by a few extra pairs of eyes can help eliminate simple problems with a ballot.
In this case, the addition of a NEW question regarding which party a voter intends to vote for could easily be interpreted by a voter to imply which party they belong to. As you point out, most voters in Hawaii do not officially belong to a particular party so consider themselves “independent.” These same people are often hesitant to state any affiliation with a a party, even if they predominantly vote for one of the big two. That point was clearly made in the public response to the idea of a “closed primary” in Hawaii.
As for what happens if a voter neglects to choose a party in that new question… At my training to be a Precinct Chair the other night, I asked the Hart representative that very question. What happens is the ballot is spit out with an error message saying that the “ballot was incorrectly marked.” At that point, the precinct official in the area can ask the voter to cover their ballot and be available to help. If the precinct official determines what the problem with the ballot is (the error message is not very specific), then they may inform the voter that they may either fill in the box to complete their ballot or submit the ballot anyway. In my experience serving as a precinct official in past elections, several voters are inclined to simply submit their ballot anyway rather than go back to their booth and re-analyze their votes. *If the ballot is submitted without stating which party they intend to vote, then all of the partisan votes on that ballot will not be counted.*
To clarify terminology, the eScan machine is NOT the DRE machine… that one is referred to as eSlate. The eScan machine is the one that scans and counts the paper ballots. In my mind the only advantage to the additional question is that if a person indicates that they intend to vote in the Republican races (this is the example used by the Hart rep), then proceeds to vote in the Republican part of the ballot and additionally votes in the Independent races, then only the Republican votes will be counted (after the voter is given the opportunity to spoil their ballot and start over).
Issue #3: I too was shocked when Hart was awarded the contract not only to provide the DRE machines as they have in the past, but also the optical scanning machines that ES&S had previously provided. As a precinct official in the past few election cycles, I, along with all other repeat precinct officials, had gotten used to the ES&S machines.
As I mentioned before, I went to my training to become a Precinct Chair the other night. I noticed, a couple of differences between the ES&S machines I have used before and the new Hart machines. First, the ES&S machines required a key (typically held by the Precinct Chair for the day) to power the machine on and off and to create the required beginning of the day (zero report) and end of the day (total counts) report printouts. The Hart machine can be turned on and off with a little toggle switch easily accessible by anyone. The Hart rep assured me that if the machine were inadvertently turned off, all you have to do is flip the switch again and it would be fine. The beginning and end of the day reports require a password to be entered twice on the Hart machine, but no key is required. Perhaps that is not a big deal, but having read stories about how someone with knowledge can hack into electronic voting machines to wreak havoc, it makes me a little uncomfortable that the computer can be accessed without a key. The other thing I found interesting was that the Hart rep indicated that “emergency ballot box” would be inaccessible during normal operations. This is the box where ballots would be stored temporarily on the occasion that the scanner went through a glitch that may render it unable to process ballots for a while. The box is still inside the main optical scan machine, but the ballots are simply dropped through a slot in the side rather than run through the scanner. The plate that blocks access to that box can only be removed by opening the whole thing, which would expose all of the ballots that had already been processed through the scanner. That just seemed sloppy at best, to me.
Just one other note of clarification… The Hart DRE (eSlate) voting systems used here in Hawaii are NOT touch screens. The voter must use buttons and a scroll wheel mounted below the screen to make their choices. That being said, I agree with your sentiments expressed about them.
Comment by rachel — August 20, 2008 @ 2:10 pm
Rachel is right. I made a mistake when I referred to the eScan. I shoulda said “eSlate.” Also, I spoke of the DRE as a “touchscreen,” though I knew it is technically not a “touchscreen.” I should have defined the acronym, DRE, to let readers understand it means a “direct recording electronic” voting machine. DREs are sometimes called “touchscreen voting machines,” but some designs, like the Hart eSlate, do NOT use a touchscreen. It was sloppy of me to use the term that way.
I am concerned by the statement:
*If the ballot is submitted without stating which party they intend to vote, then all of the partisan votes on that ballot will not be counted.*
I was told by a reporter that Cronin had explicitly told him the votes would be counted even if the “Select a Party” question were left blank. Which struck me as an unusually intelligent response for the Office. Now that you are saying they explicitly said the votes would not count if the question was not answered, I am concerned again.
Comment by Bart — August 20, 2008 @ 6:47 pm
Bart,
I am concerned about votes not counting if that box is not completed too. As I said, I explicitly asked the Hart representative that question. Voters WILL get an error message that will allow them to “fix” their ballot, but I am concerned about how many will just push it back through the scanner anyway.
Comment by rachel — August 21, 2008 @ 9:37 am
I asked Lance Collins about your suggestion and he said that there is no reading of the cards permitted at the precincts. The real problem is that there are no rules. Cronin has refused to promulgate Chapter 91 rules for anything so everything just ends up being arbitrarily “ruled” by guess who? That is really what the Maui case is about but the press thinks that readers are too dumb to “get” that (or maybe they’re too dumb to get it) so they say it is about the insecure phone lines which is only part of the case.
Also although that Candidate Fact Sheet contains wrong information on the law it says right on the front it should not be counted on as a substitute for the real law. The problem has been that some policies that conflict with the law and some were overturned and some were enforced o matter their conflict. I’m glad you did point out a few of the laws especially Decosta’s lack of adherence as to going to court..
Cronin is to blame for all of this and should be fired,. He is incompetent and capricious in his decision making. And you didn’t mention his race discrimination suit from Wisconsin. I came across a lawsuit from 1987 where Cronin, working as an elections attorney for the state, was involved in a case where he and the Town of Blackwell were accused in a civil and voting rights case of denying a group of blacks the right to vote. He only got out of the suit due to qualified immunity after two lower courts ruled his actions so egregious that there was no immunity,. He had to go to the 7th US Appeals court to get his immunity. See http://parxnewsdaily.blogspot.com/2008/08/diggin-up-yard.html for details
Comment by Andy Parx — August 22, 2008 @ 10:33 am
Andy,
When Cronin’s name first came up as a possible CEO, I found accounts of Cronin’s involvement in the lawsuit you cite. I have just gone to your website and have re-read the sections you have quoted. I do not minimize civil rights violations in voting cases–it has a long history and must be opposed.
But I read the lawsuit differently than you do. Not knowing the actual facts and dynamics in the small town where this took place, I am relying upon fragments within the rulings. The suit arose when a number of people at a Job Corp facility attempted to vote in the nearby towns elections. Apparently some residents took exception and challenged whether the would-be voters were actually legal residents of the town or if they were transients only temporarily residing there, but legal residents of another locale. According to the filing quoted on your site:
Cronin advised a member of the Town Election Board that any person who indicated to election officials that he did not consider himself a resident of Wisconsin was ineligible to vote and should not receive a ballot.
I am unsure why you believe such advice is inherently “racist” or should have opened Cronin up to a lawsuit.
This issue of temporary residents affecting local elections comes up all the time across the country. Small towns with significant college student populations are concerned, as are places like Florida which has a large “snowbird” population. (In the recent Democratic caucuses here in Hawaii, I heard credible reports that non-residents voted.)
In a section of the Appeals Court ruling you quote, is this passage:
In Wisconsin, when a voter is challenged as to residence, he is to be placed under oath and asked whether he intends to make his home in the ward in which he is seeking to vote. If notwithstanding the answer given, the vote is still challenged, the voter may take an oath that, among other things, he is a resident of the ward. He may then submit a ballot under challenge which is retained separately from unchallenged ballots. The fact that Cronin’s advice did not include all of the implementing details under the Wisconsin statutes is plainly insufficient to permit a finding that Cronin violated clearly established Wisconsin election laws.
So contrary to your assertion that Cronin “only got out of the suit due to qualified immunity,” it appears from the quote that they did not believe he violated the law and so dismissed the suit on a key point, not on a technicality.
I am unsure what you are speaking to when you say: “there is no reading of the cards permitted at the precincts.” If you are speaking to my suggestion that the results at the County-level counting centers should be tabulated, printed out and witnessed prior to transmission, this does not require “reading the cards at the precinct.” I work at the state counting center, not at a county one, but I believe the cards are all read at the county center before transmission. I am just saying an extra safeguard should be instituted where the results are confirmed and verified prior to transmission. The Maui observers should insist on it, in my opinion. Perhaps Lance could have proposed to the court such a safeguard as a stopgap measure if the judge were to decide it is too late to change the current system for this upcoming election.
And finally, I do not think firing Cronin is necessary at this time, so close to the elections. I think you misjudge the amount of work it takes to put on an election if you think the resulting disorder would not create any problems. If you think Cronin’s deputies are capable of running the elections without Cronin, I think you do not understand how much his deputies, and the state AG’s office, have been involved in making those mistakes about which you, and I, are complaining.
Let’s get through this election as best we can, then decide whether Cronin’s performance has been such that he should be jettisoned. If Cronin is smart, he will start becoming a bit more receptive to those calling for changes in the Office’s operations. If he fails to listen, he can then be terminated. He has time to turn this around. Will he take advantage of the time? Dunno.
Comment by bartman — August 22, 2008 @ 5:26 pm
As a technologist and someone with a great deal of experience dealing with computers and networks at all levels, my concerns fall primarily into categories relating to the use of technology in voting.
It is said that familiarity breeds contempt, and that is certainly the case for me. Modern computers are far from the simple, easily verified calculators they used to be. For that matter, I have heard from math teachers that even modern calculators are computers subject to the kinds of programming errors, bugs, and incorrect answers that are endemic in traditional computing.
Thus I believe we should treat computers in voting as fundamentally untrustworthy, and focus on how they can be made the most reliable, and how we can safeguard ourselves not just from the potential for deliberate misconduct but also from the very common, very real threat of poor design and programming which is so common in voting machines.
This can be greatly improved by having real audits of real votes, as in Bart’s point #5 above. Voting has very strange requirements that no other computing system requires - full anonymity while maintaining full transparency and full auditing capabilities. This is only possible if the SOURCE of the audit can and is reconciled with the RESULT of the audit.
You do this by making sure that the original vote is correctly recorded, and by checking the original vote against the count with a secondary mechanism not related to the primary counting mechanism.
How do we make sure the original vote is correctly recorded if we put a computer in between the voter and the vote? The short answer is that despite many attempts and claims to do so, this has not been solved. Pens marking Paper assure that the vote recorded is the one the voter intended. For those in need of assisted marking, systems like the AutoMark are a reasonable compromise but clearly require additional safeguards, which the AutoMark to its credit does provide (the read-back reinsertion).
How do we check the original vote against the count? The time-honored solution is to have humans do a percentage count, and compare that to the totals. Sadly as Bart notes, while we have the best audit percentage, we only audit a single race - and because it’s a statewide race, we will never know if any of the other counts can be trusted.
By complicating our voting with machines where none are needed, and by performing inadequate audits, we leave ourselves open to the greatest risks associated with the use of technology in voting.
Voting must be Transparent, Anonymous, and Auditable. Any piece of a voting system (hardware, software, or process) which does not meet this standard, *for any reason*, is not valid and must be improved or replaced.
- jason
Comment by Jason Forester — August 23, 2008 @ 2:09 pm
Thank you all for this very informative discussion of election issues facing Hawaii. I’m startled by some of it, but have only one specific comment at this time. Rachel, you said: “If the precinct official determines what the problem with the ballot is (the error message is not very specific), then they may inform the voter that they may either fill in the box to complete their ballot or submit the ballot anyway.” HAVA says that another required option is to offer the voter a replacement ballot. (See Section 301(a)(1)(A)(ii)-http://www.fec.gov/hava/law_ext.txt) Isn’t this mandate being followed?
~Ellen
Comment by EllenTheisen — September 3, 2008 @ 3:42 am
Ellen,
Yes, precinct officials are being trained to offer the option to spoil the ballot and offer them a new ballot to complete. Sorry I didn’t mention that. It still concerns me though since I have noticed in past elections that people tend to be in a hurry and often don’t want to take the time to start all over again.
Comment by rachel — September 3, 2008 @ 10:14 am
I voted at Honolulu Hale today, walk-in absentee, aka early voting. It was quick and easy, though I still felt unprepared about the BOE election.
After voting, I spotted Glenn Takahashi of the City Clerk’s office and asked him why 3 of the 5 Oahu absentee polling places only allow for electronic voting and do not provide optical scan ballots for those who prefer voting on paper. He started to say the polling places could only provide them with spaces too small for both systems, but it became clear as he talked that the office is trying to convert people over to all electronic voting.
I do recognize the inconvenience of having to provide absentee voters with all the different ballot types– Oahu must have 40-50 different ballots. We have 34 House Districts, plus some House districts are further split with different council districts and Senate districts.in addition, ballots must be available in several recognized languages. So having the right number of ballots in all the varied types requires some storage space and guesswork. All these ballot types can easily be stored in the memory of a voting machine. Fair enough.
Except the electronic votes cast during absentee walk-in are not audited. In exchange for this incredible convenience, a gaping hole is opened in the security and verifiabilty of the election system.
Early on in the debate, advocates for paperless electronic voting system frequently argued that we all trust the ATM machines for making cash transactions, so we should also trust electronic voting systems. I am sitting on the beach, writing this on my iPhone, so I won’t go into the whole argument. But to borrow that analogy, what safeguards would we require if the election officials, politicians and the private companies who provide the voting systems were to being allowed to shift around and withdraw public funds from the state treasury? Would we simply trust them, or would we institute auditing safeguards to ensure they were not embezzling money without telling us? Why should we require more scrutiny over our money than over our votes? Why are slot machines and video poker machines in Vegas subject to much more testing than these voting machines? Would any bank use an ATM machine if they were prohibited from reviewing the software running it?
Comment by bartman — September 18, 2008 @ 5:00 pm