BRAD BLOG Asks Dean Logan and Leon County, FL's Ion Sancho About Experimental Election Night Program for Small Number of Precincts
Q&A's on That and a Few Other Points of Note, Documented on Video During Recent Appearance in Los Angeles by Both Election Officials...
By Brad Friedman on 7/30/2008, 5:31pm PT  

-- Brad Friedman

I had the opportunity to ask an important question of both the legendary Leon County (Tallahassee), FL, Supervisor of Elections, Ion Sancho, and Los Angeles County's no-longer-interim Registrar of Voters, Dean Logan, last week here in Los Angeles.

Readers of The BRAD BLOG will remember Logan from his poor showing, as acting Registrar, in the handling of L.A.'s Super Tuesday "Double Bubble" ballot debacle, in which at least 12,000 mostly-perfectly-countable votes from the Democratic Primary went completely uncounted, as well as his handling of the failure I encountered myself, in the more recent state primary in June, when four of my own votes were printed incorrectly by the county's ES&S electronic voting system.

Sancho was in town for press avails, in advance of the opening of Dorothy Fadiman's new documentary Stealing America: Vote-by-Vote (opening for week-long runs in NYC on Aug 1st, and in L.A. on Aug 15th - DISCLOSURE: I appear briefly in the film). Logan, to his credit, was kind enough to show up, with his deputy Peter McNamara, for a screening of the film held by PDA last Thursday.

After the screening, Sancho, a compelling presence in Fadiman's film, as he was in HBO's landmark 2006 documentary, Hacking Democracy, answered questions, and the activists in the crowd managed to shoehorn Logan onto the stage to speak to a number of local concerns as well.

I took the opportunity to ask a few questions of both men, as captured by Alan Breslauer, with answers, on the video posted at the end of this article.

Of most important note, I wanted to know if either of them would commit to a hand-count of ballots in a handful of precincts this November, on an experimental basis, on Election Night. The count, in such a "pilot program" as I described it, needn't be the official count, but could later be compared to the results as reported by both counties' optical-scanners, since both L.A. and Leon County use paper ballots for their non-disabled voters.

As I've advocated to a few other election officials in the past, hand-counting in this small, experimental way would help us all to begin to compile data about the effectiveness --- or even, lack thereof --- of transparent, polling-place based HCPB (Hand-Counted Paper Ballots) on Election Night, as many Election Integrity advocates have called for...

While HCPB is still done in a few, often smaller jurisdictions around the country --- perhaps most notably, and successfully, in some 40% of New Hampshire precincts --- the practice has been routinely dismissed by doubters who charge that hand-counting at the polling place is impractical in large jurisdictions (e.g., Los Angeles, the country's largest) or in places which have lengthy ballots with many candidates and ballot initiates (eg. California).

The most virulent of critics --- let's say, voting machine vendors, and many of their supporters --- argue that hand-counts are less accurate than machine counts. The folks who make that argument have little data to back up their assertion and, usually opportunistically, point to Florida 2000 as their most well-known and frightening example of the dangers of hand-counting. However, what happened in FL 2000 was most decidedly not the sort of HCPB that proponents are calling for.

Setting aside the duplicitous political wrangling that occurred in Florida in 2000, where Republicans fought successfully to not have ballots counted at all, by hand and otherwise, the counting (or lack thereof) that Americans witnessed in 2000 took place at the state's various county headquarters by a small group of counters from the local election boards.

HCPB, on the other hand, calls for fully transparent, polling place counting on the night of the Election, immediately after the close of polls, by citizen-volunteers of all parties, in full public view, with results posted immediately thereafter, before ballots are moved anywhere.

The disingenuous suggestion that somehow millions of ballots must be counted by a tiny group, as in FL 2000, making the process untenable, unreliable and impossibly long is without substance. The subset of ballots at any given precinct, HCPB-proponents argue, is small enough to allow for them to be counted manageably, quickly and accurately on Election Night, at each precinct, in full view of the citizen-stakeholders of our elections.

Hand-counting in a handful of precincts around the country this November, on an experimental test basis in such pilot programs, would allow us to begin to compile data as to the effectiveness and accuracy of HCPB. Doing so in a number of counties and in a number of states is, as I see it, a rather reasonable and practical step to take, if the necessity of citizen ownership, oversight, and transparency in elections is ever to be fully realized.

I have no idea what the results, good or bad, might be. That's why such experiments are necessary and, as I see it, quite reasonable. As a framer, and original signer of the Creekside Declaration, calling for "citizen ownership of transparent, participatory democracy," I'd heartily encourage such pilot programs all across the U.S. this year.

In response to my question for a commitment to such a project, Sancho noted that the state of Florida law disallows hand-counting of ballots, even in exceptionally close races, as the Orlando Sentinel shamefully detailed earlier this week.

California, however, has no such restrictions that I'm aware of. Furthermore, California has a provision in its Election Code (EC 19211) that allows for the "experimental use at an election in one or more precincts" of even uncertified voting systems, for official use in actual elections. "Its use at the election is as valid for all purposes as if it were lawfully adopted," the code goes on to say about voting devices used under this provision.

Given that, and given that I'm not even suggesting that hand-counts in this experiment be done for the official count, but rather, in parallel with the officially certified systems, so that we might compare and contrast results, I can see no reason not to carry out such an experiment and find out whatever we find out.

While Logan declined to commit to such a pilot program then and there, understandably, he responded that he's "not closed to that idea" and further pointed out that he's carried out such experiments in the past.

"I'm not gonna stand here and make a commitment to a specific pilot project, tonight," he said. "But what I will say is I have a history, both here and in my previous work in Washington, of doing pilot projects. So I'm not closed to that idea."

Glad to hear it.

Bowen Has Shown Interest in Hand-Counting

California's Election Code, as noted above, allows for experimental programs, such as hand-counting in parallel to the machine count as I'm suggesting, "provided provided that the use of the voting system or systems involved has been approved by the Secretary of State." (EC 19210)

As CA Secretary of State, Debra Bowen was present in New Hampshire last January to observe hand-counting (see Why Tuesday's two-minute video above right) during the state's embarrassingly flawed Primary. Flawed, at least, in the 60% of precincts which chose to use known-hackable, error-prone Diebold optical-scan systems instead of hand-counts, without bothering to audit the accuracy of even a single ballot before publicly announcing the "results." (You can see the exact same Diebold op-scanners used by NH being hacked before your eyes in Sancho's own office, in this clip from Hacking Democracy).

So it seems to me that Bowen would be open to allowing such pilot programs here, should any county election officials in the state request her permission.

For clarity, I'm calling for ballots to be counted as usual, by optical-scan and/or or touch-screen as per each county's practice, but I'm suggesting that before any ballots or memory cards are sent back to county headquarters for final canvassing, that they be delayed at the precinct to allow citizen representatives from all parties to carry out a precinct-based hand-count, the results of which will then immediately be released to all, and posted at the precinct. Afterwards, the county's existing tabulation process will proceed as usual and will still serve as the official count.

Seems reasonable to me. I hope to speak more with Logan, to follow up on this, and I hope other Election Integrity advocates (read: good citizens) will work with their local election officials as soon as possible this year to try and coordinate similar pilot projects around the country. I can see absolutely no harm in doing so.

In California, the Election Code, Article 5, "Manual Vote Count in the Precinct," sections 15270-15281, describes the official procedures for counting ballots in the state by hand at the precincts.

For other states, or even improvements on the current provisions in CA, the HCPB-advocates at the Election Defense Alliance have posted a downloadable Hand-count Handbook based on the very successful models as used in New Hampshire. The handbook offers excellent, step-by-step instructions for the process of carrying out secure, accurate, and transparent hand-counted elections.

Sancho Ends "Sleepovers"

My questions to Logan and Sancho, along with their answers, follow in the video below. It's almost exactly 10 minutes.

In his answers to my questions, Sancho also speaks about how he's changed Leon County's procedures for deployment of voting machines following the security risk revealed in voting machine "sleepovers" at poll workers' houses, prior to elections.

Such risks were revealed to be of grave national concern by The BRAD BLOG in great detail, in the wake of San Diego's 2006 Busby/Bilbray Special Election aberration. We have also been credited for coining the term "sleepovers," by the way.

"Once we ascertained that the vulnerabilities of the equipment were such that, in fact, at any unprotected moment you could insert a virus or a patch or alter the device," Sancho said, "we no longer allowed the devices to be out of an official chain of custody, at all."

Sancho noted, to applause from the assembled, that his machines in Tallahassee are now deployed to the polling place, "in steel cages," on the morning of Elections. Unfortunately, despite CA SoS Bowen's pre-election campaigning against the practice in 2006 --- she even released a campaign video highlighting the dangers, and later declared her belief that the practice was "illegal" --- she still allows the practice to continue in the state, albeit with new restrictions. Critics have charged, however, that her restrictions are not rigorous enough, and are poorly enforced.

After Sancho's comments, Logan takes his crack at my questions. Though he says, as seen in the video, that I asked "about ten," I was so appreciative that he showed up for the event at all, and agreed to take questions, that I bit my tongue to avoid cracking-wise in response that it was actually only three questions, which I'd have thought, as Registrar, he might have been able to count a bit more accurately.

So I decided to write the wisecrack here, instead of saying it there. Who says I can't reasonably compromise?


(Thanks to Election Integrity advocates Jody Holder, Tom Courbat, and Gregory Luke for cites to legal references as used in this article.)

Prices now slashed in The BRAD BLOG's 2008 Election Integrity Fund Drive! Please support our continuing coverage of your election system, as found nowhere else. Click here for a number of cool new collector's edition Premium products now available starting at just $5!