Friday, August 26, 2005

What's the Buzz?

Do you ever do a series of sums on a calculator, but because you're dissatisfied or unsure of the results, redo your arithmetic with pencil and paper? Or, alternatively, do you ever work out some figure using pencil and paper and then, to doublecheck your results, use a calculator to confirm your arithmetic was correct?

If you answer yes to the first question, you are very rare indeed. If you answer yes to the second question, you are typical -- indeed, one might say, normal.

My posting about electronic voting machines from Monday, which was both a response to Charlottesville City Councilor Kevin Lynch and a partial report from a legislative hearing in Richmond that same day, has generated a number of responses around the blogosphere, mostly from people who, it would seem, answer "yes" to that first question I posed, and who disagree with much of what I had to say.

Commonwealth Conservative
, One Man's Trash, and Bearing Drift each linked to my original post. George Loper was kind enough to reprint it in full. He also published responses from Rich Collins, who has been in the news himself this week, and Justin Moore, a computer science graduate student at Duke University who also testified before Delegate Hugo's subcommittee on Monday. Other responses that I have seen came from the Jaded JD and from Waldo Jaquith (with a comment from Janis Jaquith)

I provided a partial response to Rich Collins and to Justin Moore, to which Mr. Moore also responded. And Jim Heilman, former General Registrar of Albemarle County who now hops around the world helping to organize elections in emerging democracies (the last I heard he was in Mindanao in the Philippines, after stints in Bosnia, Afghanistan, and Ukraine, among other exotic locales), offered some words of support for my position:

The to-and-fro over the safety and security of voting machines has been interesting, even though the arguers and arguments are a bit long-winded. If I may throw in a bit more pithy two cents of my own…..

Rick Sincere is right on in his defense of our voting systems for two major reasons that are continually overlooked and ignored by some of the voting machine naysayers:

(1) You can't hack in to something that has no means of electronic communication. You can't hack in to a typewriter unless you have the typewriter in your lap. As Sincere says, electoral boards with brains would never adopt any of the systems whereby results are directly communicated from machine to some central location. We get results election night by telephone from an election official, and that's the way it should stay.

(2) To "fix" a system, you've got to bribe a whole bunch of people along the machine "chain" and count on all of them keeping their mouths shut. In our day and age, it's hard to count on anyone keeping his/her mouth shut - even Deep Throat.

I hope Sincere keeps calling it like it is.

For the record, here is what I sent George Loper under the subject heading "A Partial Reply to Justin Moore":
While I appreciate Justin Moore’s detailed reply to the piece I wrote about Charlottesville’s electronic voting machines (the Hart eSlate system), which you kindly reprinted, and while I hope this debate continues to take place on a civil and fact-based level, I am afraid that the article he cites as refutation to mine is lacking.

Mr. Moore writes:
Mr. Sincere testified that an attacker would have to bribe nearly a dozen people -- if not more -- to hack into a voting system. This shows a lack of imagination as to how hackers actually get into computer systems. Last summer I created a write-up of how one might break into a voting system, with only modest means.
Upon reading his article, which he calls “President Nader, or How I Learned to Stop Worrying and Love DREs,” it became clear that his plan of “modest means” rests upon an assumption that there are no security procedures associated with elections that could prevent his imagination’s attacker from achieving his goals. It betrays a surprising lack of familiarity with how elections are organized and the hierarchy of responsibility that gives different election officials, from the State Board of Elections through the local Electoral Board to the Chief Election Officers at the precinct level to all of the other officers of election, specific duties designed to ensure the integrity and security of the electoral process. This hierarchy of diversified responsibility entails a system of checks and balances that places a series of nearly insurmountable obstacles in the way of Mr. Moore's phantasmagorical scenario.

Take this sentence from his 2004 essay, for instance:
Less obvious (and more difficult to trace) is an outsider who has access to a DRE machine and can reverse-engineer the data storage format.
An “outsider,” by definition, cannot get access to a DRE – certainly not in Charlottesville. The only people who have access to the DREs are election officials. Except on Election Day itself, the DREs are kept in a locked room inside the inner office of the General Registrar. Even City cleaning staff are not permitted to enter that room or the room immediately outside its door.

Moreover, Mr. Moore’s entire argument rests on the assumption that the electronic voting machines are connected to a central tallying location via a modem or some other telecommunications link.

George, let me assure your readers: No voting equipment used by the City of Charlottesville is ever connected to any telecommunications device, either by modem, over a telephone line, via a wireless modem or radio signal, or through a link to the Internet. All voting equipment in Charlottesville is stand-alone and secure.

On election night, results are called in by Chief Election Officers to the Electoral Board by telephone and delivered orally. The tallying devices (what we call “JBCs”) are sealed several days before the election and carried by hand, with seals intact, by Electoral Board members or the Board’s officially designated representative, to the Registrar’s office, where the seals are broken in the presence of the Electoral Board and other witnesses. (The unique identification number on each seal is matched to the number recorded when the seal was made and only if the numbers match can the seal be broken.) Then the Mobile Ballot Boxes (MBBs) are removed and inserted into a stand-alone computer that is used for only one purpose, to tally votes.

When those votes are tallied, the unofficial results of the election are reported to the press and public on the night of the election. The next day, the Electoral Board convenes for a canvass of the vote. Like every other formal meeting of the Electoral Board, this canvass is open to the public for observation. (Seldom, however, does any member of the public, or even the news media, come to observe the canvass.)

At the time of the canvass, the Board reviews the results from the night before, double- and triple-checks the arithmetic, counts any provisional (paper) ballots that may have been cast and adds them in to the totals, and prepares an abstract of the vote that is returned to the State Board of Elections in Richmond. All records of the election – including any paperwork produced at polling places on Election Day, all poll books recording who voted that day, all paper ballots (used and unused), and all Mobile Ballot Boxes – are turned over to the Clerk of the Circuit Court for storage for a time specified by the Virginia Code. All those records are available for review during the period they are stored by the Circuit Court.

Mr. Moore’s musings about the ease of hacking into an electronic voting system may apply to other systems used in other states or by other localities in Virginia. Since he was present at the August 22 hearing in Richmond, he knows that several speakers warned against the use of modems or other telecommunications devices for transmitting vote totals. There is general agreement within the elections community that using telecommunications tools for this purpose is, frankly, a dumb idea and it should be rejected out of hand. Mr. Moore’s elevation of this avenue of attack as something deserving of concern is a red herring.

Let me remind your readers, Mr. Moore included, that in my testimony on both July 19 and August 22, I stated that while we have concerns about how a VVPT or VVPAT – a paper receipt showing how an individual votes – might be implemented, because of technical, cost, and other problems, we have no objections to the concept in principle, and the Charlottesville Electoral Board is on record as inviting the State Board of Elections to use our city as a test site, if and when the VVPT technology available for the Hart eSlate system is authorized for testing and certification in Virginia. We have no wish to be obstructionist but we do want to work out the kinks before we decide to rely on what is an essentially untested system that could cause more problems than it solves.

Permit me to comment briefly on Rich Collins’ response to my article. I am grateful for his thoughts and agree that the situation in Maryland is worth looking at – as a salient example of what not to do. My concern about Maryland is that it uses the same kind of voting equipment everywhere in the state. There is no diversity among counties and cities. This is a dangerous situation, because it makes the job of Mr. Moore’s hackers that much easier.

The more various the types of voting equipment there are, the harder it is for even the most determined conspirator to disrupt an election or alter its results. Virginia uses more than 20 different types of voting equipment, from paper ballots to touch-screen machines to optical scan machines to old-fashioned lever machines (which will soon be retired) to the Hart eSlate. Even in a state Senate or House of Delegates district, unless that district is wholly contained within a single city or county, voters are likely to encounter two, three, or four different voting systems. This is a vital measure to assure the security and integrity of our elections.

One final note of reassurance for your Charlottesville readers: In discussing electronic voting machines, or DREs, such as those that concern Mr. Moore and others, the Wall Street Journal's John Fund, who is highly critical of election security procedures and greatly concerned about election fraud, notes in his 2004 book, Stealing Elections: How Voter Fraud Threatens Our Democracy, that “in the twenty-plus years that these machines have been used, in many counties all across the country, there has never been a verified case of tampering.”

We are about to use the Hart eSlate system, on November 8, for the eighth time in Charlottesville. In the seven elections since May 2002, not only has there never been "a verified case of tampering," there has never even been an allegation of tampering. The record speaks for itself.
Next week I will be attending the annual convention of the Voter Registrars Association of Virginia (VRAV) in Roanoke. I suspect that the questions raised in these discussions will be a major topic of conversation among the elections professionals attending that confab.


Anonymous said...

Justin Moore here. I posted a response to Mr. Sincere's response, also available at Mr. Loper's blog. My response is available here.


Joe Berenguer said...

Congratulations Friend for your excellent blog on Gateway Computer Store!Keep up the good work!
If you have a moment, please visit my site:
Gateway Computer Store
I send you my warm regards and wish you continued success.
Have a nice day! :-)