Common Sense | CTVotersCount.org

Common Sense: Justified Confidence

“I think the biggest issue facing us is trust in the elections,” said Denise Merrill, Connecticut’s secretary of the state. – As Feds struggle, states create their own anti-election propaganda programs

Trust and confidence are important – Justified trust and justified confidence. – Luther Weeks, Facebook comment

As we have said before Connecticut is above average in election integrity and security for statewide elections, less so for local elections. Above average, is not saying much. Many states, including Connecticut, have a long way to go to achieve justified confidence. PR alone will not protect us from outsiders and insiders. Will not protect us form loss of confidence in democracy.

Note: This is the fourteenth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut.

“I think the biggest issue facing us is trust in the elections,” said Denise Merrill, Connecticut’s secretary of the state. – CNN: As Feds struggle, states create their own anti-election propaganda programs

Trust and confidence are important – Justified trust and justified confidence. – Luther Weeks, Facebook comment

Every time I hear election officials talk about ‘confidence’ I tend to hear what CNN heard, a desire for the public to be assured at all costs that elections can be trusted. Whenever I get the chance, either publicly or one-on-one, I point out to officials that we agree, up to a point; I want the public to have confidence, justified confidence. I always remember the first time my wife and I passed out a flyer about election integrity concerns at a public hearing on our flawed post-election audits. An apparently respected registrar said to us:

You shouldn’t be passing that out, you will scare the public. If there is ever a problem, we will fix it in the backroom, just like we did with the lever machines – Respected Registrar who trains Moderators

One caveat, I don’t know that elections officials always intend to imply that weak, PR type of confidence. Sometimes I may be too sensitive, but no matter the intent, the public often hears what CNN heard.

Case in point, the election complaint settlement in Hartford, covered in yesterday’s Hartford Courant: Hartford Registrar of Voters Penalized:

The registrar, Giselle Feliciano, and Martin Allen Jones, a ballot moderator, violated state statute during the Hartford Democratic Town Committee primary in March 2018, according to a June 19 order from the State Elections Enforcement Commission.

Feliciano and Jones agreed to a civil penalty of $750, which will be reduced to $500 if Feliciano repeats a certification class on absentee voting.The settlement was approved by the city’s corporation counsel, and will be paid by the city, according to city communications director Vas Srivastava.It stems from a complaint filed last year by Anne Goshdigian, a challenge slate candidate in the town committee primary, and Thomas Swarr, whose wife Donna Swarr was running on the same challenge slate…

The pair alleged they were trying to watch election officials count absentee ballots at City Hall about 1:30 p.m. March 6, 2018, when Jones said Goshdigian could not be in the room and all others needed to leave as well, including Jones himself.When Goshdigian and Swarr asked for written proof they had to leave the room Feliciano said Goshdigian could not be within a polling place during voting hours. City Clerk John Bazzano was consulted, and he said the room was not a polling place, according to the complaint.

“Feliciano then ‘changed tact,’ and asserted that it was within her authority to remove anyone disrupting the process,” the complaint said. The registrar called the police and had Goshdigian and Swarr removed.

The situation was partially resolved after Donna Swarr spoke with a state’s attorney, who confirmed to her and Feliciano that members of the public could observe absentee ballot counting.

But Feliciano still did not allow Goshdigian to observe the count, according to the complaint. And when Tom Swarr returned to City Hall that evening to observe the final absentee ballot count, he was again denied access .

Swarr said he’s frustrated that the commission’s decision doesn’t address the second incident.

According to the agreement the commission reached with Feliciano and Jones, it “does not believe that there was any untoward intent insofar as the handling of the absentee ballots themselves.” Feliciano admitted she did not know that the law required her to open the count to all members of the public.

“I don’t know how you could say she didn’t understand what the law was when I came back in the evening and was denied again,” Swarr said. And, the commission said it “does find a troubling lack of contrition and/or remorse in the Respondents’ statements in response to the allegations here. Their failure to acknowledge a material error of law in their answers to this matter weighs in the commission’s decision here…

The registrar’s procedural manual on counting absentee ballots also covers, on its second page, who may observe a count.Of what he saw, Swarr said the absentee ballet counting process seemed perfectly secure. However, the lack of transparency gave a false impression that the system was corrupt, he said.

“It’s silly to exclude people and basically create distrust of the system when letting people observe would add confidence in it,” Swarr said. “The lack of particpation is, to me, the greatest threat to Hartford elections. You should be encouraging people to come in and view and build confidence.

The one good thing is that their was a penalty for this violation, yet everything else should leave us all with less than justified confidence in the system:

Like many complaints, this one took a stretched-thin SEEC months (15) after the election to resolve.
Although there was a penalty it was not paid by the Registrar or Moderator involved, like the legal defense it was paid by the City – what actual penalty is there in that?
The Registrar can save the City almost nothing after tuition and mileage, if she spends a half a day taking an applicable class. Yet the Moderator is not required to be recertified. In the law, the Moderator is just as much, if not more, responsible for following the law here.
Sadly, if the Moderator were re-certified he would not be taught about absentee counting – that is not covered as we recently pointed out to the General Assembly, to no avail, as they cut the moderator certification requirements in half.
And perhaps, worst of all, there was no real remedy to the problem. In one of the most partisan election situations, with insiders in charge of an election won by insiders, there was no transparency, no public verification. A complete lack of credibility and a formula for cheating available for the future: Steal some votes, if someone tries to watch and complains, have your city pay a small fine and WIN!

Common Sense: Limits on Testing From Turing to Self Driving Cars

At first this may not seem like Common Sense. We have the famous Turing Halting Problem which has some very important consequences for voting which may not, at first, make common sense:

Note: This is then thirteenth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <previous>

At first this may not seem like Common Sense. We have the famous Turing Halting Problem which has some very important consequences for voting which may not, at first, make common sense:

We cannot use testing to be sure that the software in a voting machine will provide accurate election results.
And any hardware circuits are also part of the machine and come under the limits of the halting problem

It is worse, beyond the halting problem:

We really have no way of knowing if the software that actually ran on a machine when the results were created and printed was actually the approved, tested software.
We really have no way of determining if the results were somehow changed by some some means external to the software.
We have no way of really determining that the components of the hardware were what were tested were actually those running the machine.
There could also be permanent or intermittent hardware errors.
The hardware errors could include logic circuits, wires, or sensors.

At this point you may be complaining that this is crazy or at least not common sense.

Consider the idea of self-driving cars. How comfortable are you with them today? Do you think testing is sufficient? Maybe. Yet, they could be subject to intermittent errors and hacking – similar to today’s vehicles that rely almost entirely on software to translate the driver’s commands into action. See: <60 Minutes Shows Threats to Autos and Voting Machines are Real>

Common Sense: The Skeptics Guide to Election Integrity and Fraud

Two events in the last week or so prompt this post. First, last Saturday I was at the Reason Rally at the Lincoln Memorial. One speaker said “Be skeptical of everything”. A later speaker assured us, among other things, that two things I believe to be true were actually conspiracy theories.

Second, a recent series of posts by Richard Charmin, essentially claiming that in many states the primary was stolen.

So, where do I come out? I stand with Carl Sagan who said, “Extraordinary claims require extraordinary evidence.” and the speaker at the Reason Rally who said to “be skeptical of everything”. Here we have competing extraordinary claims:

By Richard Charmin: That, in a large number of states the election results were manipulated in favor of a single candidate.
Implicitly by complacence: “Move on, nothing to see here, exit polls are always wrong in the U.S. Don’t be concerned that every time someone brings this up, they are always wrong in favor of one candidate or party”

Note: This is then twelfth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <previous> [just an interesting coincidence the last Common Sense post was exactly one year ago!] <next>

Second, a recent series of posts by Richard Charmin, essentially claiming that in many states the primary was stolen, based on, among other things, a pretty consistent difference between raw exit polls and the results, almost always favoring one candidate. Looking at the details, we see that Connecticut is one of those states.

So, where do I come out? I stand with Carl Sagan who said, “Extraordinary claims require extraordinary evidence.” and the speaker at the Reason Rally who said to be skeptical of everything. Here we have competing extraordinary claims:

By Richard Charmin: That, in a large number of states the election results were manipulated in favor of a single candidate.
Implicitly by complacence: “Move on, nothing to see here, exit polls are always wrong in the U.S. Don’t be concerned that every time someone brings this up, they are always wrong in favor of one candidate or party”

It is clear that both of these are extraordinary claims. We are disappointed in the lack of others providing factual evidence and solid arguments refuting or confirming either of these extra ordinary claims.

Democracy requires solid answers. Voters and Candidates deserve solid answers. What is required is Evidence Based Elections, elections that provide strong evidence that the outcome reflects the votes of the voters.

At this point we do not have that evidence in Connecticut. One approach would be a strong post-election audit showing that votes were counted accurately, that ballot counts matched the voters checked in, and that polling place and central count absentee counts were accurately accumulated.

In Connecticut, there are gaps in the post-election audit, transparency lacking in the totaling process, and challenges in verifying all the data. We are at work on developing answers which might provide reasonably convincing evidence for Connecticut.

Update: Skeptics Guide Part 2: Absence of Evidence is Not Evidence of Absence

Common Sense: Laws must be Sufficient, Enforceable, and Enforced

There is an analogy with laws, including election laws. Laws must be Sufficient, Enforceable, and Enforced. Missing one of the three, all value is lost.

Note: This is then eleventh post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <next> <previous>

In one of his books, Gerry Weinberg pointed out that employee evaluations should be multiplicative not additive, that is, the various dimensions of performance and capabilities should be multiplied rather than added to determine the overall value of an employee. e.g. If my writing and verbal communication are poor, no matter how much technical knowledge I have, what I can contribute is very limited, yet with just average skills in every other area matched with high technical knowledge one can accomplish a lot. Similarly with great interpersonal skills, yet poor technical judgement, I can be less than valuable!

There is an analogy with laws, including election laws. Laws must be Sufficient, Enforceable, and Enforced e.g.

Sufficient – Laws have to be sufficient to prevent that which they are designed to prevent, or to protect what they have been designed to protect. e.g. In Connecticut we have weak ballot security laws: They do not protect all ballots until they are needed for post-election audits; some of the security requirements are ambiguous, open to multiple interpretations; and are based on unwarranted trust in weak seals and entirely lacking in seal protocols.
Enforceable – There has to be a reasonable means of enforcing the law. Once again, we point to ballot security in Connecticut where it is generally believed (ambiguous law) that two individuals from opposing parties must be involved in any access to ballots, yet most ballots are locked in cabinets in rooms with a single lock, with both registrars and often others having access to a key, along with a log of access maintained by an honor system.
Enforced – In reality the must actually be enforced. In recent years we have seen many examples, from banking fraud, leaks of classified information by high-level officials, and campaign finance laws.

We were reminded of this limitations today with an article, one among several recently, on the Federal Elections Commission: More Soft Money Hard Law <read> The FEC is stymied by partisan gridlock.

We all have seen the lack of strong enforcement against the fraudulent activities of big banks, their management, and employees. Or, perhaps less known, existing trade agreements with environmental and labor protections which are ignored, rendering the provisions that sound powerful, generally meaningless.

So, whenever we ask for sufficient election laws, we remind that more is needed. They must also be enforceable and enforced. Missing one of these three components, all value is lost.

Common Sense: [How] Do you know if your vote counted?

The Citizen Audit has just opened up our signup for the audits for the primary, which start fifteen days after the primary. The primary is August 12th, so the audits will begin Aug 27th.

Q: So, why bother signing to spend a day observing the audits?
A: To understand and the question ” [How] Do you know if your vote counted?”

Note: This is then tenth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <next> <previous>

The Citizen Audit has just opened up our signup for the audits for the primary, which start fifteen days after the primary. The primary is August 12th, so the audits will begin Aug 27th.

Q: So, why bother signing to spend a day observing the audits?
A: To understand and the question ” [How] Do you know if your vote counted?”

The Citizen Audit’s new flag promotes the theme Vote, Audit, Observe. It summarizes three of the keys to approaching “knowing” if our votes are all counted and accounted for correctly. They highlight the importance of voting, auditing the vote, and independent observation of the audits. Given the requirement of the Secret Vote to prevent successful selling, buying, or intimidating votes it is impossible for any one person to determine if their individual vote was counted accurately, yet possible collectively for independent observers to determine if everyone’s vote was counted accurately – or at least to provide some assessment of how confident we are in the reported counts and winners.

From the Citizen Audit:

Voting is the foundation of democracy. Some say it is a per-Constitutional right, to choose our government, established by the Declaration of Independence. Today voting integrity is as important as ever, and perhaps we are even more aware and concerned with voting vulnerabilities.

Vote: Citizenship is of little value if we do not vote. Not voting assures that your vote won’t count. If you value democracy, vote! If you want change, vote! If you prefer stability, vote!

Audit Connecticut has paper ballots which provide evidence to verify election results and provide confidence. Paper ballots are of little value without sufficient recounts and audits. Connecticut is one of about half of the states with post-election audits.

Observe Checking ballots by officials is insufficient. Public confidence requires public verification – transparency and direct public observation and analysis. That’s where you can participate further in authentic Democracy!

Common Sense: The good, bad, and ugly secret ballot

We often take for granted the idea of the secret ballot. One alternative to the secret vote is the public vote. Sometimes we would prefer a public vote, sometimes it is necessary,

Note: This is then ninth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <next> <previous>

We often take for granted the idea of the secret ballot. We see elections in other countries with people putting a folded ballot into a ballot box. We go into a voting booth and want our privacy. Some want to conceal their vote from relatives, friends, employers, fellow union members, or church members. Yet there are alternatives and sometimes we want them, sometimes they are necessary.

Researching the history of the secret ballot, in The Hidden History of the Secret Ballot, a couple of years ago, I was surprised to learn:

It was implemented in the U.S. generally after the Civil War
It was implemented for partisan reasons – to suppress the black vote, under the assumption it would hamper the illiterate from voting.
There is also the tradition of the New England Town Meeting with voters standing or raising their hands to vote.

This year we have fought hard to protect the secret vote in Connecticut, yet not for partisan reasons. It has several benefits:

It keeps votes from being bought, sold, or intimidated – prior to the Civil War votes were regularly bought and voters under peer and other pressures to vote in particular ways.
It helps the losers accept the election result as the actual will of the people.

One alternative to the secret vote is the public vote. Sometimes we would prefer a public vote, sometimes it is necessary:

Our representatives in the Legislature, Congress, and most public bodies vote in public. Not so long ago, committees in Congress voted in secret. Members could tell both sides of the public and lobbyists that they agreed with them and it was ‘others’ who voted the wrong way, blocked, fixed, or promoted legislation. Without a public vote we would have no idea how our elected officials actually voted. No way to hold them individually to account.
Where votes are proportional, they need to be recorded and verified by ownership. Examples include stockholder votes or condo associations where some votes are based on ownership proportions.
A public vote is much easier to verify. After every close election we hear charges by the losing side or their supporters than a particular vote was incorrectly counted – by a candidate or party in Connecticut – or a government official unhappy with the declared winner in another country. Sometimes those charges may be true. In every case it reduces the trust in the process and in democracy.
On the other hand, a public vote can also lead to charges of payments or intimidation deciding the election.

The Connecticut Constitution gives us the right to the secret vote. In considering the Constitutional Amendment this year, we note that it represents a third alternative: A semi-secret vote, if anything a worse alternative than either a secret vote or a public vote with the disadvantages of each. It would allow a special, very vulnerable class of voters, deployed military, to waive their right that their vote be secret – a right we claim is not theirs, but every voters’ right that everyone’s vote be secret. It would have none of the advantages of a public vote, since nobody knows if the soldiers votes are known, only perhaps, a few insiders may know. There would be no public verification, even for the soldier.

One final example. Last year we had a petition circulated in our condo complex to ask the Board to rescind a vote to remove part of a border fence. It became very hot issue. Over two-thirds of the residents signed the petition. The contingent in favor of removing the fence charged that many people were intimidated in signing the petition, others mislead, and others did not know what they were signing. There probably were some cases where each of those concerns were justified. We have had similar changes in every public vote. There are two issues: 1) What would be the actual free choice of the majority? 2) How can the losers accept the result? The answer is to actually conduct a secret vote – yet it must be in a way that everyone trusts the process.

Common Sense: Public Transparency and Verifiability

In our last post in this series, Why Should Audits Be Independent, we ended with “When it comes to elections, are independent audits sufficient? Not really. We need public transparency and verifiability as well.” In this post, we will address transparency and verifiability.

Note: This is the eighth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <next> <previous>

According to Wikipedia:

Transparency, as used in science, engineering, business, the humanities and in a social context more generally, implies openness, communication, and accountability. Transparency is operating in such a way that it is easy for others to see what actions are performed.

According to Miriam Webster:

Verifiable: Capable of being verified: Verified: to establish the truth, accuracy, or reality of <verify the claim>

What we mean by Publicly Transparent and Verifiable is that members of the public can actually determine the election or audit outcome because the process is transparent in a way that allows the public to verify the result.

This goes beyond a call for independent audits. Independent audits require that we trust the auditors. Certainly an independent audit is more trustworthy than a non-independent audit, which in turn is seemingly better than no audit at all. Perhaps not. No audit at all, clearly provides little confidence the process beyond “trust me, and trust all officials in their integrity, in their competence, and in securing the process”.

Yet, a non-independent audit can provide a false sense of confidence. We can never be completely sure, even of an independent audit – the auditors may be biased, incompetent, sloppy, or compromised. Far fetched? We are not so sure, given the history of transgressions in all walks of live. Elections are especially vulnerable because of the central role they play in democracy and the requirement of the secret ballot, making end to end verification impossible.

But if audits, recounts, and elections are publicly transparent and verifyable, there is an alternative! Members of the public (and candidates) can judge the process and determine the results themselves. When it comes to audits and recounts, for hand counting, the public should be provided an opportunity to clearly see and verify,:

each ballot and vote on the ballot as it is being classified
that votes are correctly classified
that votes and ballots are counted correctly by counting teams in batches
that the totals are the valid sums of numbers determined by counting teams in batches
that consolidated reports are based on the numbers determined in audited or recounted districts

This is really the minimum for hand count audits and recounts. For audits and recounts (and elections) to be fully trusted, then the election day process must also be publicly transparent and verifiable and the chain-of-custody needs to also be publicly transparent and verifiable. Let us leave the voting process and the chain-of-custody for another day.

Common Sense: Why should audits be Independent?

Why do we need independent audits for elections just as we have for other business and government functions?

Note: This is the seventh post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <previous> <next>

In the last few weeks, since the revelations of Edward Snowdon, we have seen the limits of Congressional “Oversight”, when a small number of legislators have access to the “facts” of the implementation of the Patriot Act, FISA, and NSA. But these facts seem to be “we assure you we are not doing anything wrong”, “our warrants are reviewed by a Court (with the facts of the warrants being secret, and the Court apparently rubber stamping every proposed warrant)” etc.

This situation is not much different from audits that lack independence. One of the requirements of the Principles and Best Practices for Post-Election Audits is Independent Audits.

The authority and regulation of post-election audits should be independent of officials who conduct the elections. The actual work of post-election audits may be best performed by the officials who conduct the elections and their designees.

a. The independence of authority and regulation may be satisfied from resources inside or outside state government.

b. The actual work of post-election audits—i.e. the handling and counting of ballots and reporting the results—may be best performed by the officials who conduct the elections.

We do not find that Connecticut’s Post-Election Audits meet that criteria, since the authority and regulation of our audits are under the control of the Secretary of the State who is also the Chief Elections Official and responsible for selection and approval of election equipment in the State. What we have seen from Official Post-Elecction Audit reports is the dismissal of all differences between audit counts and machine counts as “human error”, largely without investigation. An independent audit should be expected to provide a thorough investigation of questionable results, rather than providing excuses.

We do not audit our own taxes and send the results in to the IRS. They audit our taxes and expect us to be able to justify random items. The IRS does not attribute differences in their information or calculations to their own counting errors! Similarly businesses often have internal independent auditors and external independent auditors to protect the interests of the shareholders. It does not always work but it often does. Given human nature, we could hardly expect non-independent auditors to approach being equally effective.

Other areas of government conduct independent audits often uncovering serious problems, like recent internal audits in the City of Hartford: <read>

The audit, by Chief Auditor H. Patrick Campbell for the city’s Internal Audit Commission, focused on the revenue management unit of the city finance department, which is responsible for the oversight of general fund revenue due the city, about $550 million this year. The auditors found a number of problems, including:

• Poor control of lease agreements, rental properties and other revenue-producing arrangements. For example, one city parking lease had not been renegotiated since 2000, for reasons that are unclear, the auditors report.

Lease issues are troubling because they were identified in an earlier audit, which supposedly spurred corrective action. The position of asset manager was created to oversee and manage leases, and a lease, licensing and contract database was to be created.

In this review, the auditors discovered that the asset manager position was eliminated and the database was never completed. Eliminating positions that bring in revenue is eating the seed corn.

• 151 bounced checks totaling more than $392,000. More than half of these checks had not been followed up and resolved. Nor was it clear whether late or insufficient-fund fees were being charged. Many fees and charges are handled by individual departments; the auditors recommend they be centralized. You can’t have people bouncing checks at city hall and getting away with it.

• Missing documentation. The department is supposed to keep track of actual vs. budgeted revenues, to determine what might be causing any discrepancies. The finance department did reviews, apparently, but “documentation to support the reviews and follow-up performed is not maintained on file,” the auditors wrote.

Another example comes form the Connecticut Office of State Ethics audit of Statements of Financial Interests filed by state legislators and other officials.<read> It serves as an example of a positive report, yet finds exceptions, and areas for improvement.

Would we really expect a report as critical as the one in Hartford if it was done by the managers and staff of the department making these significant errors? Would we expect legislators and employees to correctly evaluate and report on the accuracy of their own disclosure statements? Would we trust the results of a positive report written by the individuals involved? Would we expect them to find the suggestions for improvement?

That is why we need independent audits for elections just as we have for other business and government functions.

But when it comes to elections, are independent audits sufficient? Not really. We need public transparency and verifiability as well. The subject of a future post.

Common Sense: Tension between Convenience, Confidence, and Cost

Many of the issues we discuss here and debate in the Legislature revolve around tradeoffs between Convenience, Confidence, and Costs. At a basic level we find three fundamental values/goals behind every initiative and debate: These tradeoffs and competing goals are the context within which we all constantly evaluate new laws and proposals.

Note: This is the sixth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <previous> <next>

Elections like many complex activities are subject to many demands and conflicting priorities. Reading Edward B. Foley’s excellent and fascinating paper, The 1792 Election Dispute and Its Continuing Relevance, one of its main themes:

Another lesson to be learned from the Clinton-Jay dispute of 1792 concerns the deep-rooted nature of the jurisprudential debate between strict and lenient enforcement of election statutes. As a review of the 1792 dispute reveals, this basic jurisprudential debate has been with us from the very beginning. The 1792 dispute also demonstrates that this jurisprudential debate involves competing interpretations of our nation’s most elementary commitment to the existence of democratic elections. Proponents of both strict and lenient enforcement appeal to the fundamental value of a free and fair vote among citizens. Yet each side of this jurisprudential debate appeals to this fundamental value in a different way. As a nation, we are essentially stuck in the same place regarding this debate as we were in 1792.

Many of the issues we discuss here and debate in the Legislature revolve around similar tradeoffs between strict and lenient enforcement – tradeoffs in laws set out before close election results, charges, counter charges, and arguments actually occur. At a basic level we find three fundamental values/goals behind every initiative and debate:

Convenience – Access for voters, access for candidates, and efficiency for officials
Confidence – Integrity and confidence in the process, voter qualifications, and accuracy of the results
Cost – What it takes to register voters and run elections

Convenience suggests capabilities such as: Election day registration; early voting, online registration; universal registration; online voting; many well staffed and equipped polling places; systems for those with disabilities; school or general holidays on election day; voter friendly ballot design; easy to use web information; low bars for third party candidates; public financing; short hours and increased staffing for officials; technology to save officials work; easy to setup technology; etc.

Confidence suggests requirements such as: Voter verified paper ballots; adequate supplies of pre-printed ballots; strong ballot security; strong equipment testing and security; fully transparent operations; careful, extensive registration checks; careful, effective voter checkin requirements; strong recount and post-election audits; stronger, more uniform, or faster enforcement of election laws;

Costs are usually required to increase confidence or convenience. Sometimes an investment in new equipment and methods can actually save money in the long run. In other cases waste can be eliminated or a more efficient method found. In other cases a well intended initiative can be accomplished in a wasteful, ineffective, even detrimental way.

In general we can tradeoff one of the three goals for one or two of the others, yet it does not always work that way:

We can save money and add to convenience at the expense of confidence when we loosen registration checks. Similarly we can add confidence, with costs and inconvenience by stronger voter ID requirements.
We can increase confidence and costs, along with more work by officials when we increase the standards for protecting ballots or camera surveillance of storage and official work areas.
We can invest in online registration, which adds to convenience, and confidence, while it reduces costs. (As we used to call it in IT, a “sweet spot” application)
We could invest in paperless DRE (touch screens) which increase work for officials, can result in long lines for voters, high risks to confidence, and huge costs over optical scanners – a lose, lose, lose, lose proposition.
Better procedures and regulations can provide a huge payoff, only if they are accompanied by effective training and compliance.
Also tradeoffs may not be uniform: Costs or additional work can be greater for small towns, or for towns with many small polling places. Voter ID laws can disproportionately greatly inconvenience and cost some voters, while hardly making a difference to others.
Finally, some impacts are really, really difficult to determine. In recent years, the Legislature changed to require special elections for U.S. Senate vacancies – each election would cost several million dollars. How many will we have? And when? Some could be very critical and valuable to democracy, others inconsequential. What is the value of ballot security we can all trust vs. questionable security almost impossible to prevent and demonstrate fraud, should it occur?

These tradeoffs and competing goals are the context within which we all constantly evaluate new laws and proposals.

Where Common Sense fails: Do insider attacks require a sophisticated conspiracy?

In this post, we address where Common Sense fails. Where what seems obvious to individuals and election officials is often counter to the facts or science. Those that are unfamiliar with technology and a specific area of science often overestimate how difficult or easy specific things are to accomplish.

Note: This is the fifth post in an occasional series on Common Sense Election Integrity, summarizing, updating, and expanding on many previous posts covering election integrity, focused on Connecticut. <previous> <next>

We frequently hear versions of the following comments, often from election officials:

“It would take a very sophisticated operation to steal an election. Computer experts with access to the election system.”

“Our staff is trusted and they don’t have that level of expertise.”

“You are a conspiracy theorist, you just don’t trust election officials, and the security of our voting machines”

To some of these charges I plead guilty and with others items beg to disagree:

I do believe in the existence and possibilities of fraud by conspiracy, yet in the case of election integrity argue that compromising an election does not require the existence of a conspiracy of the sort implied by the current definition of conspiracy theory. In fact, individuals have been convicted or exposed for small to moderate size conspiracies.
I do trust most election officials. The problem is that many election officials express and request blind trust of all election officials. This despite regular instances of errors by officials, and occasional successful prosecution of various election officials for criminal violations. Unless election officials are cut from a different class than other citizens and public officials, some of the time, some of them will make errors, and others will comitt fraud, sometimes without prosecution, and sometimes undetected.
It does not require a sophisticated operation to steal an election. Fraud would not necessarily require computer experts with access to the election system.

In this post, we address where Common Sense fails. Where what seems obvious to individuals and election officials is often counter to the facts or science. Here we have to be careful trusting our own initial views and those of honest officials, we need to be open to the idea that we may not individually have all the answers -willing to listen to, if not completely trust, scientists and the facts. (We are not just talking about elections here, but many other areas which are critical to democracy and life.)

Those that are unfamiliar with technology and a specific area of science often overestimate how difficult or easy specific things are to accomplish. As we often confuse conspiracy and conspiracy theory, we often confuse the meanings of theory, between the common meaning of theory and a scientific theory. They are as different as a Pat Robertson theory of earthquakes and the germ theory of disease.

For instance, people often think technologists can do anything such as solve the nuclear waste problem, cure all cancer, make smoking safe, produce clean coal, or provide safe internet voting. These are all hard problems that have, so far, eluded teams of the best scientists. I frequently recall a friend in middle school, in the late 1950’s, who had no concerns with smoking, saying “By the time I get lung cancer in 30 or 40 years, science will have a cure”.

Once even “scientists” believed with the right recipe sea water could be turned into gold. In the dark ages of the 1950’s it was believed it would be possible to predict the weather and the economy, if only we had enough data and the right programs. Since then, with the advent of Chaos Theory, we have learned both are impossible, yet that fact has provided us the opportunity to deal with the economy and weather more rationally and realistically. Since the 30’s or 40’s we have also known that it is impossible to prove that any computer software/hardware system is accurate and safe – there is no recipe possible. (And thus it is also impossible to build a computer or communications system that is provably safe. In practice, we can see from failed attempts of government and industry that the best systems are, in fact, regularly compromised, providing practical as well as theoretical reasons to avoid trusting any computer/communications system.)

On the other side, many things are much easier than the public and many elections officials believe. Smart individuals and small groups continue to create computer viruses and hack into the best systems of the most sophisticated government agencies and industries. On the easy side, the U.S. Government believes, apparently with good reason, that a single Army Private could access and steal a huge number of confidential documents from many Federal agencies. (That he was a low level insider with lots of access, just emphasizes how vulnerable systems are to a single insider and that it would take steps in addition to a safe computer system, even if that were possible, to protect us from an insider.)

How often have we each gone to an expert with what we viewed as a tough problem, only to have it solved quickly and inexpensively? For example: Recently, my condominium unit needed a new main shut-0ff valve. The maintenance staff and I believed it would be a big job requiring service interruption to dozens in my neighborhood requiring a shut-off of a valve in the street. Enlisting the help of a general plumbing contractor, the contractor simply froze my pipe while installing a new valve.

When it comes to election machine hacking, online voting, and conventional stealing of votes it is relatively easy in many jurisdictions to compromise the vote, especially when it only requires a single insider. Some attacks take extensive technical knowledge which many hackers possess and could help or intimidate a single insider to execute or could simply get a job in election administration. Other attacks take very little technical expertise. When officials misjudge how easy it is for attacks to be accomplished, when officials don’t understand technology, it makes it all the easier for a single trusted insider.

One company, LHS, programs all the election memory cards for Connecticut and other states. LHS’s President said that we are safe from hacked cards because he has no employees with software expertise (including himself). There are several fallacies in this:
— How would he know if a particular employee has technical expertise?
— It is not all that hard to miss-program memory cards.
— A single employee could gain outside technical help or be intimidated to do what an outsider demands.

Similarly, many election officials would claim we are safe because they do not have computer experts on their staff. Once again, how would they know how much it would take and what a person does not know?

As for outsider attacks, one example: To our knowledge, in only one instance, a Internet voting system was subjected to a open, public security test. It was compromised extensively and quickly. Even if it had not been compromised so easily or was subjected to a more extensive test it would hardly be proven safe, hardly be safe from attack by insiders.

In our view, the best we can do realistically is voter created paper ballots, counted in public by machine, a printout of results in public, followed by a secure ballot chain of custody, followed by effective independent post-election audits, and where necessary complete recounts. All transparent.

Finally, we need to emphasize the requirement for a “secure ballot chain-of-custody” or at least a reasonably secure system making it difficult for single insiders to compromise ballots. For those with blind trust in security seals we provide presentations by an expert <view> and examples of quick seal compromise by that same expert and an amateur <read>