Election Updates

From today's Washington Post:
Petition Fails to Derail Md. Early Voting

An op ed in the The North County Times (Riverside, CA) claims that voters choose to vote absentee--or use a paper ballot at the polling place--because voters don't trust electronic machines.

The writer's view of public opinion is correct, although frustrating for many scholars and election officials. Survey results that I recently saw from Alvarez and Hall show that the public remains distrustful of electronic voting machines, even though all the evidence indicates that paper ballots are far more fraught with error.

Convincing the public of the reliability of electronic voting machines remains one of the main challenges for the election reform community. If public opinion is lost on this issue, expect to see an even faster movement toward absentee and other "reliable" paper systems.

Ben Adida, a graduate student in Ron Rivest's group at MIT, is defending his PhD thesis today at MIT. Ben's thesis is titled "Verifying Secret-Ballot Elections with Cryptograph", and here is the announcement. The abstract for Ben's talk is here:

In the US, the secret ballot is 115 years old: the first 23 Presidents were elected using public polling. Introduced to stem
voter coercion, the secret ballot carries, to this day, a significant audit-ability and transparency cost: how can voters be given direct assurance of their vote without enabling coercion? Cryptography often solves problems with conflicting requirements: in this case, cryptography can fully reconcile ballot secrecy and election auditability.

This talk presents an overview of cryptographic voting techniques developed over the last 20 years and introduces two new ideas:
(1) Scratch & Vote, a practical, paper-based, cryptographic voting system that is particularly useful in illustrating the capabilities of cryptographic voting, and

(2) Public Mixing, a new theoretical definition and construction that achieves anonymization (of votes, for example) through public computation.

This work asks "if crypto voting is so good, why aren't we using it yet?" and offers some tentative answers.

Ben's thesis defense will eventually be available on the web, as will his thesis. Of course, when these links are available, I'll post them here and then provide additional details regarding his great work.

I've heard Ben give talks based on his thesis research, and have read some of his research. It's important work, and it is also highly significant to see that there are now graduate students who are doing serious research on voting technology issues, something that we really need more of! So far graduate students associated with the VTP have produced a range of research papers, and prior to Ben's PhD, there have been four excellent MA theses on voting technology at MIT.

The renewal of the Voting Rights Act has been stopped in its tracks in the House because of concerns about bilingual ballots. As the Washington Post notes:

But nearly 80 House Republicans signed a letter by Rep. Steve King (R-Iowa) objecting to the Voting Rights Act's provisions that require state and local governments to print ballots in foreign languages -- or provide interpreters -- in precincts showing a need for such services. The requirement is a costly unfunded mandate for many counties and municipalities, the letter said, adding: "The multilingual ballot mandate encourages the linguistic division of our nation and contradicts the 'Melting Pot' ideal that has made us the most successful multi-ethnic nation on earth."

There are two issues here of interest. First, there is a big difference between being bilingual--speaking two languages--and being biliterate--being able to read two languages. This problem is most pronounced for voters from many Asian nations that have symbolic writing and not a Latin alphabet. These individuals may be able to speak and read basic English but still want the comfort of a translation.

Second, having just watched language minority voters cast ballots in Los Angeles in the California primary, I question whether it is beneficial to do away with interpreters. Many of those who I have seen who need interpreters are older voters, many of whom fled Communist nations in Asia to come to America and live the American Dream. Do we really want to take away interpreters from these citizens?

I was at the Salt Lake County (UT) election offices on Friday and they told me the most interesting story. An elderly man came in to vote in early voting in the primary election. Because of the party nomination process used in Utah, there are only primaries in non-partisan races and in cases where incumbents face strong challengers who fared well in the party caucus meeting. The man lived in a precinct where there were no non-partisan races on the ballot and there were no Democratic party primary races. (I too, live in one of these precincts).

The man looked crestfallen and told the poll workers, "I have terminal cancer; I will be dead before the general election. I just wanted to vote one last time."

The election workers were quick to come up with a solution. They let him cast a provisional ballot in the Republican primary. His vote won't count--he isn't eligible--but he left the polling place very happy.

While I just posted commentary about how voters in Utah liked electronic voting, this story from the San Francisco Chronicle details that election officials may not like the machines (or at least the transition has been rocky).

A columnist in the Deseretnews News gives the thumbs up to early voting using DREs.

The radio show "Open Source," which broadcasts on some NPR stations (but is not an NPR produced show), ran a show on Tuesday called "Steal This Election." The show had Dan Tokaji, Congressman John Conyers, Mark Miller, and Bill Todd. The show page has links to all sorts of reading and materials that may be of interest to election watchers, especially those who have followed the Robert Kennedy story in Rolling Stone and the related replies. (The Kennedy article and critiques of it are all on the Open Source website.