Open Letter to Congressman Ehlers (OPPOSE HR 811)

[kster]

15 April 2007

by David Griscom

The Holt Bill, HR 811

Dear Congressman Ehlers,

After long and careful consideration, I am asking you to OPPOSE The Voter Confidence and Increased Accessibility Act of 2007 (HR 811) because it is insufficiently protective of our democracy.

By way of introduction, I too have my Ph.D. in physics. I am a Fellow of the American Physical Society and also of the American Ceramic Society, where I listened to you speak at an annual meeting about a decade ago. I retired from the Naval Research Laboratory in Washington, DC, in 2001 after 33 years of civilian service and took up a succession of invited professorships of research at the Universities of Paris-6, Lyon-1, and St-Etienne in France and Tokyo Institute of Technology. Subsequently I was Adjunct Professor of Materials Science and Engineering at the University of Arizona before moving to Mexico.

While I see very many reasons to reject HR 811 (too many to have any hope of easy correction by amendments), I will emphasize only the following three here:

I. Touch screen voting machines (DREs) must be abolished because the only record of cast votes is electronic, making it impossible for "We the People" to know for sure how many votes were truly cast for each candidate. "Toilet-paper" roles of thermal paper are seldom verified by the voter, cannot be proven to correspond to the electronically recorded vote, are not durable, and will probably never be audited. Moreover, the blind and handicapped realize that there are far better systems than DREs available to enable them to vote http://www.huffingtonpost.com/brad-friedman/blind-disabled-voters-fi_b_43450.html.





http://www.opednews.com/articles/opedne_david_gr_070415_open_letter_to_congr.htm

[Wilms]

1. Can you really blame him?

II. Optical scanners must also be abolished. They are actually worse than DREs because they give a false impression of invulnerability to electronic election fraud by virtue of the ability to audit and/or recount the paper ballots inside them.

[Einsteinia]

7. Disagree. We CAN had 99% confidence

that our votes are tabulated as cast with optical scans IF we count, approximately 10% (use formula), ballots of each and every PRECINCT under citizen-controlled purview BEFORE the ballots are released into the anonymous dark tunnel on route to the central tabulaton processor. (In CA that would translate to roughly a mere 40 to 60 ballots in most precincts.)

In fact we could confer at least 99% confidence using optiscans and counting them at the precint in this fashion, but a full hcpb at the county level would confer a lesser percentage.

Why? Because once they are transported, out of view, even with chain of custody, even with surveillance tapes and armor trucks, you still could not have as much confidence as if you checked them before the go on this anonymous journey--much of which is more open to violation.

But that is NOT to say that we aren't aware that Optiscans can be hacked at least six ways. The point is it wouldn't matter. To check the outcome at the precinct would be ultimate in L&A testing, because it would detect "Easter Eggs." And when all the results of this sampling are complied together, it would tell us whether our election process was accurate or not. If not,the process would require a further first count until the margin of error between the machine count and the collective precinct counts would not overcome the election.

The issue though would still be that the paper scanned is not the BALLOT of record, and instead, in CA is still just an AVVPAT. But that could be worked out now that we have Bowen.

So, when we ask for hcpb, I think we should say the NH hcpb method (because that is an in-precinct and highly thought-out process).

[Wilms]

11. I'd be happy to see the OUTCOME confirmed to a 99% confidence level.

While I'm generally sympathetic to the idea of a precinct counting, I can't argue that in theory or in practice, one method to be superior to the other. Adjectives and metaphors aside, neither have you.

You're smart enough to imagine a methodology for ballot transport, or query Febble on UK procedures, and not resort to scientifically unsound terminology like "anonymous dark tunnel" and specious statements like the third paragraph of your post, over and over again.

That's not to say I'd oppose precinct based hand counts. But mind you, I'd want to see those counts subjected to audits. (I imagine that RINOS or even DINOS could create a lot of trouble in some areas of the country.)

As to fixed percentage audits, we've been there before. http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=470074&mesg_id=470115

I like good audits. Actually, one in particular. http://www.votetrustusa.org/pdfs/VTTF/EVEPAuditing.pdf
But not enough election reform advocates seem to have taken the time to consider them well.

Hence, I throw my hands up and ask "Can you really blame him".

[papau]

2. Optical scanners with mandatory random audits of greater than 10% will work -

but HR811 is not there yet.

Many mathematicians and election design experts (including myself but only as to being a math person familiar with statistics) have written to the committee on this - but no one has been allowed to testify - not even Kathy Dopp who is the most up on this topic of anyone in the US.

They have used a "safe" witness list and are listening to voting machine folks a lot more than the math side of the community that has actually followed the problem and which actually has a peer reviewed solution that actually costs less.

Seems either grassroots do not make enough campaign contributions, or the HR811 staff are lazy or arrogant.

[kster]

3. We have to take back our RIGHT to hand count ALL the paper ballots

I'm not a mathematician and I shouldn't have to depend on a mathematician to confirm for me that all the votes are being counted, its time to secure the vote count, by hand counting ALL the paper ballots, immediately following the election. :-)

[papau]

6. And that means audits need to be manadotory , random, and of more than 10% of the ballots n/t

n/t

[kster]

9. No need to audit an area, that a group sets up a 100% Hand Count

of all the Paper Ballots, you would only need to do mandatory audits in areas where there is not going to be a FULL Hand Count of the Ballots by the people.

[OnTheOtherHand]

4. that's not exactly how it works

If Holt controlled the witness lists, there would be a lot less "listening to voting machine folks," at least in that setting. Not that many people listen to hearing testimony anyway.

Besides that, I don't see any connection between the debate over audit sizes and what the voting machine folks want. I think the HR 811 audit size proposal is the sort of compromise that appeals to many folks because it seems so simple to understand -- which isn't the worst possible reasoning, although the record reflects that I'm pretty dissatisfied with it. But I've seen no sign that the voting machine folks give a fig how audit sizes are computed. They are concerned about their own bottom lines. Voter verification is a threat to the DRE business model because it drives up costs and eliminates the supposed logistical benefits of a paperless approach. Of course, manufacturers may vary somewhat in their preferences depending on how well positioned they consider themselves to respond to various changes in law.

[papau]

5. Interesting thought - but the witness lists were the election commission (that seems

controlled by DRE business) people plus folks either uninformed or technically incapable.

You do not get such poor witness panels without trying.

Holt does control the lists through his committee staff - or he ignores his lack of management ability for a reason. I spoke to staff and faxed, etc - and could not get anything changed or even get change discussed - although they are very polite and friendly.

The stacking of the record was no accident - and audit sizes was but one of about 20 changes being needed. But it looked like the bill was being set up to bargain away the mandatory part and the random part and leave us with some ignorable percentage. I worked in DC for many many years and all this smelled too familiar.

As to a business model, DRE is not sold on need or excellence for function or work flow improvement or cost savings - it is sold on mandate plus getting the disabled scared.

The op-ed open letter does not appear to have any holes - do you see any?

The Holt HR811 smells like its been corrupted by the corporations - and it is the only thing moving on election reform.

Can you offer a "this is not a corrupted process" explanation for either the EAC no actions on a host of issues, or the Holt committee staff setting up panels that exclude well credentialed strong critics of the system? Since we last posted to each other a lot more election theft news has come out of Ohio. I agree with your (and Liz's) red shift analysis - the corruption was not nationwide and had no pattern except to be always in the GOP's favor (a comment not based on statistics - only on subsequent revelations by those that were there who actually reported the problems - you have to look very hard to find any real reports on Dems doing odd things and then getting good election results), and the data is too sparse to conclude anything about a given machine.

Indeed it is the corruption I saw in 1960 with Nixon's Illinois voter suppression so as to keep "unqualified" (read non GOP) voters from voting and with random polling stations getting ahead of the curve and actually changing voting counts and losing ballots (reported in the media 6 months later when ballot boxes floated to the surface in the Fox and DesPlaines River spring floods outside Chicago - reported in the Waukegan NewsSun but only in the obscure section that reported the various towns' police logs during the flood - and with not one actual story for follow up.

It has not changed in 50 years - the corrupt GOP steals elections as they yell about keeping elections clean - and with the Holt bill it seems there will not be real change - just a calming of the crowd.

as to the U.S. Elections Assistance Commission (EAC) being in the pocket of the GOP/DRE company owners:

EXCLUSIVE: ES&S TOUCH-SCREEN VOTING SYSTEMS FOUND VULNERABLE TO 'SERIOUS' VIRAL VOTE-FLIPPING ATTACK; U.S. ELECTIONS ASSISTANCE COMMISSION REFUSES TO ISSUE WARNING
Scientific Report Finds 'Serious Security Vulnerability' Similar to 'Princeton Diebold Virus Hack' in Widely Used iVotronic System, Allowing a Single Person to Change Election Results Across Entire County Without Detection

Despite GAO Confirmed Mandate to Serve as Info 'Clearinghouse', Embattled EAC Says They Will Take No Action to Alert Elections Officials, Public

-- By Michael Richardson and Brad Friedman

While revelations surrounding the mysterious 18,000 "undervotes" in the November 2006 U.S. House election between Christine Jennings and Vern Buchanan in Florida's 13th Congressional district continue to inform the nation about the dangers of electronic voting machines, new information has recently come to light exposing a shocking lack of responsible oversight by those entrusted with overseeing the certification of electronic voting systems at the federal level.
An investigation into what may have gone wrong in that election has revealed a serious security vulnerability on all versions of the iVotronic touch-screen voting system widely used across the country. The iVotronic is a Direct Recording Electronic (DRE) touch-screen voting machine manufactured by Elections Systems & Software, Inc. (ES&S), the nation's largest distributor of such systems.
The vulnerability is said to allow for a single malicious user to introduce a virus into the system which "could potentially steal all the votes in that county, without being detected," according to a noted computer scientist and voting system expert who has reviewed the findings.
And yet, despite their federal mandate to serve as a "clearinghouse" to the nation for such information, a series of email exchanges between an Election Integrity advocate and officials at the U.S. Elections Assistance Commission (EAC), has revealed that the federal oversight body is refusing to notify states of the alarming security issue.
The recent email conversation shows that even in light of the EAC's review of the warning from the computer scientist who characterized the "security hole" as severe, needing to be "taken very seriously," and among the most serious ever discovered in a voting system, the EAC is unwilling to take action.
Recent reports by the Government Accountability Office (GAO) have taken the EAC to task for a failure to meet their legislated mandate for informing the public and elections officials about such matters. However, a review of the email communications to and from the EAC's Jeannie Layson, show that the federal body is steadfast in their refusal to take action to alert either elections officials or the public about the security risk recently discovered by a team of eight noted computer scientists.
The EAC's current Chairwoman, Executive Director, Director of Voting System Certification and other top officials at both the National Association of State Election Directors (NASED) and even the GAO were included in the series of email communications, The BRAD BLOG has learned.
The vulnerability was initially discovered by a panel of scientists convened by the State of Florida to study the possible causes for the FL-13 election debacle. The team's discovery revealed a design issue in the widely used iVotronic system could allow for a viral attack, by a single individual, which could then spread unnoticed throughout the electronic election infrastructure of an entire county.
A similar vulnerability was found in DRE touch-screen system made by Diebold last Summer by a team of computer scientists at Princeton University.
Attempts to seek information about EAC plans to notify other states and local jurisdictions who use the same vulnerable voting systems as the ones in FL-13 have been met with an astounding refusal, troubling denial, buck-passing, and a lack of accountability by the federal commission of Presidential-appointees. The agency has also come under fire in recent weeks for a number of questionably partisan decisions and other failures to perform as mandated by the Help America Vote Act (HAVA) of 2002.
Of late, the EAC has been forced to respond to a great deal of controversy, on a number of different operational matters and policies, as revealed by a series of articles on this site and in mainstream outlets such as the New York Times and USA Today. Several of those matters have drawn Congressional notice, questioning of EAC officials and letters of inquiry. Thus, this latest revelation is likely to add to the rising concern of Congress members as new federal legislation introduced by Rep. Rush Holt (D-NJ), and currently facing mark-up by a Congressional committee, would permanently fund the now-embattled EAC. Funding for the agency was originally mandated by HAVA only through 2005.
The new ES&S iVotronic vulnerability first emerged on February 23, 2007, when the Florida Dept. of State released a report detailing their findings from the investigation into what happened in Sarasota's still-contested Jennings/Buchanan race. That election was ultimately decided by just 369 votes. The state's official findings included a report conducted by an eight-member computer science and technology team under the auspices of Florida State University (FSU). The report sought, unsuccessfully, to determine the cause of the unexplained "undervotes" reported by the iVotronic touch-screen voting systems used in Sarasota's portion of the FL-13 race on Election Day and in early voting.
Although the reason thousands of votes turned up missing from those systems remained unknown, the study team did discover a serious security flaw in the iVotronic system that is used in Sarasota and many other jurisdictions across the country (and even the world, as France is set to use the same systems in their upcoming Presidential Election.)
Election integrity watchdog John Gideon, a frequent BRAD BLOG contributer and the Co-Director and Information Manager for VotersUnite.org, says that the security flaw pertains to "every ES&S iVotronic voting machine used in the US and overseas." A total of eight separate versions of the system --- without and without so-called "voter verified paper audit trail" (VVPAT)" printers --- all currently approved as qualified at the federal level, are affected, he explained...

COMPLETE EXCLUSIVE REPORT: http://www.bradblog.com/?p=4396
.

[OnTheOtherHand]

8. Holt doesn't chair the committee

If you want to speculate that Millender-McDonald is friendly to the manufacturers, you wouldn't be the first. I don't claim to know much about that.

I certainly agree that DREs haven't been sold on demonstrated excellence! (Well, the manufacturers claim to have demonstrated excellence.)

[papau]

10. No wonder I got nowhere in my calls - Holt doesn't chair - darn - well

Congresswoman Millender-McDonald says all the right things - even against date violence - but that is the only "issue" up on her site.

We may have a committee run by the staff - who have been "lobbied" - - :-(

[WillYourVoteBCounted]

12. doesn't Millender McDonald prefer paperless?

If EDA's plan worked out, more states could end up with paperless DREs.
If paper-less voting is not banned nationally,it will make a huge comeback, because many election officials will push for it.

The Truth About Hand Counted Paper Ballots:

Hand Counted Paper Ballots - What They Look Like Hand Counted Paper Ballots - an Informational Journal. There are many folks that wish Hand Counted Paper Ballots were a reality in the US....Many of these same folks have never seen a hand counted paper ballot, but they often will tell you about how Canada counts all of their ballots in about 3 hours....
http://blackboxvoting.com/s9/index.php?/archives/172-Hand-Counted-Paper-Ballots-What-They-Look-Like.html

Here is what hand counted paper ballots look like in Australia,
Canada, Germany, Iraq, Mexico and Russia (and a few others)
http://www.ncvoter.net/downloads/Hand_Counted_Paper_Ballots.pdf

HR 811 doesnt ban hand counted paper ballots. If folks want to do that, they
still can.

They need to find a better way to do it, hold more elections with less items on the ballot,
reduce the number of contests voted directly for....

If folks really want HCPB, they should at least persuade one jurisdiction to use that method.