A House of Voting Cards

elections, certification/testing, reform, news, problems, research, policy

The New York Times revealed last week that one of the Independent Testing Laboratories (ITAs) that qualify voting systems for conformance to the federal voting systems standards, Ciber Laboratories, Inc., had been suspended from approving new machines. Apparently, the Election Assistance Commission (EAC) suspended Ciber due to lax "quality-control procedures and [Ciber] could not document that it was conducting all the required tests." ("Citing Problems, U.S. Bars Lab From Testing Electronic Voting")

We'll have more to say in detail in the future about what this revelation means in terms of oversight of voting systems, and oversight of the EAC itself. However, a number of simple questions come to mind: Why was this development kept from the public? Why were machines that Ciber had erroneously approved allowed to be used without additional testing in last Fall's general elections? How many models of voting systems are we talking about here? How widely deployed are Ciber-tested voting systems in our elections environment?

Saving criticisms of secrecy for a different forum, the last few questions should be easily answered by figuring out which voting systems Ciber tested and looking at voting technology deployment statistics. However, since the test reports are not public, it is difficult to find information about who tested what when.

Under the old NASED-led regime?recall that there is a new federal voting system certification regime overseen by the EAC?an ITA would test a voting system and then submit a report to NASED. NASED's technical committee would review this report and issue a NASED Number if the voting system was found to be compliant with federal voting systems standards.

Why do I mention NASED numbers? Well, they look like this: N-1-16-22-22-001. That's not a simple serial number; those dashed fields hold information. I have seen a few different presentations where someone in NASED or the EAC explains what these numbers are, but I suppose I didn't take good enough notes, because I couldn't find them.

Today Kim Alexander of the California Voter Foundation sent email with exactly the notes I wanted ("What is in a NASED Number"). To make a long story short, the first number in a NASED number corresponds to the ITA that did the software testing (the software ITA is responsible for full-system testing). If a NASED number reads N-1- then it was tested by Ciber, if N-2- it was tested by Systest Laboratories.

With this key piece of information, we can use published lists of qualified voting systems to determine which models where qualified by Ciber. Here is the "NASED Qualified Voting Systems" list from March of 2006. Looking at this list, 41 out of 55 entries (75%) have N-1- at the beginning of their NASED number, indicating that Ciber was the software ITA. In fact, it is much more simple to list which systems were not qualified by Ciber.

Of course, what does this mean in terms of deployed systems? In detailed terms, it means any voting system that is not one of the following was qualified by Ciber:

  • AccuPoll,
  • Danaher (for their VSS 2002 qualified system),
  • Liberty Voting Systems/NEDAP, Populex,
  • One version of ES&S' voting system (N-2-02-22-22-004).

This population of systems that aren't cast in the shadow of the Ciber revelation, those qualified by Systest instead of Ciber, makes up a tiny part of deployed voting systems.

We can take it one step further. From the Election Data Services dataset in their "2006 Voting Equipment Survey", you can calculate what percentage of voting systems were certified by Ciber. What do the numbers say? They say that the voting systems used by 68.5% of registered voters (67.9% of precincts) in the 2006 election were qualified by Ciber. (Here is a spreadsheet of the EDS data that I used to calculate this in Open Document and Excel Formats: ODS, XLS.)

I suppose it would have been completely impractical to decertify all these systems. Even decertifying those systems in which the qualification testing Ciber performed was specifically lacking would likely be a significant double-digit percentage of voting systems used by registered voters (or precincts).

What is in a NASED Number?

elections, certification/testing

(From Kim Alexander of the California Voter Foundation)

Here's an explanation of the digits in a NASED Number, the number issued when a voting system would pass qualification of federal voting system standards. These notes were taken during a presentation by Brian Hancock of the EAC entitled, "Breaking down a NASED number".


Breaking down a NASED number

software ITA has responsibility for system testing -- this is either Ciber or Systest

Example number: N-1-02-12-22-001
N= Nased
1 - ciber or systest
02 - vendor ID (1 -16)

12 - software standards version

1 - 1990
2- 2002 standards
11 -- 1990 and 1990, not tested to 2002
12 - 1990 and 2002, changes tested to 2002 standards
22 -- fully qualified to 2002

22 -- hardware /firmware standards version

001 - system version number


elections, berkeley, friends, research, policy, legal, education, iSchool
macro close-up image of a champagne cork

Yesterday, I passed my phd qualifying exam. Yay!

Later today, I'll post some reflections here about the process and how invaluably useful it was for me. (Comments are open for a bit)

SMC Times: Elections Office Gets Tips from Experts

elections, reform, standards, news, privacy, friends, research, policy, legal

Rebekah Gordon of the San Mateo Times has written a piece about some of our auditing work:

Elections Office gets tips from experts

By Rebekah Gordon

SAN MATEO ? In an effort to make the counting and recounting of votes more open to the public, San Mateo County election officials have embraced those who some might call their critics.

Since the June primaries, some of the area's key voting researchers and advocates have been hunkered at the Elections Office on Tower Road, trying to understand how the county counts and verifies its votes, and how all of that is changing with the advent of electronic voting machines.


I should mention that the 10-sided dice work is from Arel Cordero, David Wagner and David Dill ("The Role of Dice in Election Audits?Extended Abstract"). Arel and Wagner are playing important roles in this auditing work.

This quote needs a few words added to it (in bold):

Hall knows these suggestions have to be tempered with practicality, which they can only get a grasp of by setting foot in the Elections Office.

"How can we get as close as possible to the ideal without completely devastating their election staff?" Hall said. "We talked about what is the ideal case and looked at some of the complications, trying to get as close as possible."

Why do Federal Election Laws only Apply to Federal Elections?

elections, reform, policy, legal

Federal election laws like HAVA, limit their mandates to apply to federal elections?congressional and presidential. It may seem silly to some that Congress can only pass election laws that apply to federal elections, unlike in other areas of law where federal law preempts state law. I frequently get asked about this so let me lay the constitutional and jurisprudential background out. In short: it's the Constitution, stupid!

The Constitution

The powers to regulate congressional elections between the federal and state legislatures is spelled out in Art. 1, Sec. IV, cl. 1 of the Constitution:

The times, places and manner of holding elections for Senators and Representatives, shall be prescribed in each state by the legislature thereof; but the Congress may at any time by law make or alter such regulations, except as to the places of choosing Senators.

And the States appoint electors to elect the President while the Congress can choose a day for them to select the President (Art. 2, Sec. 1):

Each state shall appoint, in such manner as the Legislature thereof may direct, a number of electors, equal to the whole number of Senators and Representatives to which the State may be entitled in the Congress: but no Senator or Representative, or person holding an office of trust or profit under the United States, shall be appointed an elector. [...] The Congress may determine the time of choosing the electors, and the day on which they shall give their votes; which day shall be the same throughout the United States.

Well, that's about as thick as mud, right?

How have the courts over the years interpreted this language? The Supreme Court has held that Congress has explicit authority to regulate congressional elections and a "broad implicit authority" to regulate presidential elections. In addition to Art. 1, Section IV, clause 1 there are a number of important Supreme Court decisions here; two for congressional election regulation (Smiley v. Holm, 285 U.S. 355 (1932); U.S. Term Limits, Inc. v. Thornton, 514 U.S. 799 (1995)) and one for presidential election regulation (Burroughs and Cannon v. United States, 290 U.S. 534 (1934)).

Congressional power over congressional elections

In Smiley, the Court was presented with the question of what does the text of Art. 1, Sec. IV, cl. 1 actually mean. It found that the U.S. Congress has full supervisory power over congressional elections (Smiley at 366-367):

[T]he second clause of article 1, 4, [provides] that 'the Congress may at any time by law make or alter such regulations,' with the single exception stated. The phrase 'such regulations' plainly refers to regulations of the same general character that the legislature of the State is authorized to prescribe with respect to congressional elections. In exercising this power, the Congress may supplement these state regulations or may substitute its own. It may impose additional penalties for the violation of the state laws or provide independent sanctions. It 'has a general supervisory power over the whole subject.'

In Thornton, the voters of Arkansas passed a constitutional amendment that altered the qualifications for AK representatives to the US. House. That was found to be unconstitutional as the Constitution prohibits states from adopting additional or conflicting qualifications.

Congressional power over presidential elections

As to power over regulating presidential elections, Burroughs recognized that Congress' power to regulate presidential elections was implicit compared with the powers over congressional regulations.

For presidential elections, the Constitution explicitly gives states the power to choose electors (Art. 2, Sec. I) and the manner that they're elected, and gives Congress the power for choosing the time and day of the election (with the day being the same throughout the U.S.). However, the Court said reading Congress' powers as being that narrow is "without warrant" and that there is a certain amount of regulation that Congress must be able to engage in over Presidential elections. They need to be able to protect the process and integrity of that election especially because states are be poorly positioned to do so or would necessarily do so inadequately.

In short, the Constitution implicitly gives this power (Burroughs at 545):

The importance of his election and the vital character of its relationship to and effect upon the welfare and safety of the whole people cannot be too strongly stated. To say that Congress is without power to pass appropriate legislation to safeguard such an election from the improper use of money to influence the result is to deny to the nation in a vital particular the power of self-protection. Congress, undoubtedly, possesses that power, as it possesses every other power essential to preserve the departments and institutions of the general government from impairment or destruction, whether threatened by force or by corruption.

and that how Congress chooses to regulate presidential elections is solely within their judgment (from 547-548):

The power of Congress to protect the election of President and Vice President from corruption being clear, the choice of means to that end presents a question primarily addressed to the judgment of Congress. If it can be seen that the means adopted are really calculated to attain the end, the degree of their necessity, the extent to which they conduce to the end, the closeness of the relationship between the means adopted, and the end to be attained, are matters for congressional determination alone.

It's the Constitution, stupid!

Contact / Help. (cc) 2021 by Joseph Hall. blog software / hosting.
Design & icons by N.Design Studio. Skin by Tender Feelings / Evo Factory.
And a few words about the structure of the eye . Everyone " retina ". Especially often we hear it buy clomid online in the phrase " retinal detachment ." So what is the retina ? This - the front edge of the brain, the most distant from the brain part of the visual analyzer. The retina receives light first , processes and transforms light energy into irritation - a signal that encodes all the information about what the eye sees . The retina is very complex and in their structure and function . Its structure resembles the structure of the cerebral cortex. The shell of the retina is very thin - about 0.14 mm.