Washblog

King County Elections has frequently said that the brand new Diebold scanners/tabulators will be subject to a stricter standard of testing, including line for line review of source code.

Electronic voting equipment must be certified at the federal level in order to be used in this state. To become certified, the equipment is "tested" by independent testing labs that are "accredited" for this purpose. There are three labs: SysTest, Wyle and Ciber. The latter lost its accredidation about a year ago based on its history of inadequate quality assurance and inability to document that critical tests were performed.

The ITL's are said to conduct tests based on the Election Assistance Commission's voluntary standards.  One of these voluntary standards is that a voting system cannot rely on "interpreted code" to operate, and the Diebold machines rely on interpreted code to function. It is this interpreted code, which is located on the memory card, that allowed the now famous Harri Hursti hacks.

The labs, also referred to as ITA's, used to be accredited by the National Association of State Election Directors (an obvious conflict of interest). Now the labs are accredited by the Elections Assistance Commission, a 4 person committee whose members are appointed by the President of the United States (another conflict of interest).

New standards have been established since the EAC took over accredidation. The standards are said to be stricter. I have not had a chance to review the new standards. However I was told by a rep of Voters Unite that he does not believe that the new standards require line for line review of the source code (which is not sufficient anyway, but may be discussed in a future post). Plus, the testing process is still secret.  The testing is done pursuant to a confidentiality agreement.  If any problems are discovered in testing, no one will hear about them.

Furthermore, although the new uncertified Diebold tabulators are currently undergoing certification testing, the same Voters Unite rep indicated that these machines are being tested to the 2002 standards, not the 2005 standards, as indicated by KCE. I will try and look into this.

 

by raincity calling on Fri Jul 27, 2007 at 04:25:02 PM PST

---

Hi DWE. Thank you for your work on behalf of election integrity.

I've been remiss in not blogging about some recent events. This comment is a two-fer: the 46 LD's resolution and our Q&A/debate with Bill Huennekens.

---

This is the resolution the 46 LD passed two weeks ago. It's pretty good. Thanks to Chad Lupkes for bringing the county's resolution, which was used to amend the 46 LD's resolution into something much better.

Resolution of the 46th District Democrats Supporting All Vote-By-Mail in
King County and Calling for Thorough Evaluation and Public Testing of
Elections Equipment for Use in 2008 Election

WHEREAS, the 2008 election will be one of the most important elections in the history of
our country and our state; and

WHEREAS, it is a priority of the Democratic Party to transition our electoral process in King
County to all vote by mail to maximize voter participation and strengthen our democracy;
and

WHEREAS, 38 of 39 counties in Washington state have already made the transition or are in
the process of making the transition to an all vote by mail system to maximize voter
participation and strengthen our democracy; and

WHEREAS, King County is considering purchasing new election equipment that will allow
King County to transition to all vote by mail for the 2008 election; and

WHEREAS, it is a requirement of Federal Law that any ballot tabulation election equipment
used by King County be fully certified by the Federal Elections Commission prior to use in
an actual election; and

WHEREAS, in accordance with Federal Law, the election equipment being considered for
purchase by King County will be fully tested and certified prior to use in an election; and

WHEREAS, if the election equipment being considered for purchase by King County does
not receive the required certification, then it will be returned to the manufacturer at no cost to
King County taxpayers; and

WHEREAS, the King County Elections Office has many public oversight mechanisms in
place and has received national recognition as a leader in election management;

NOW THEREFORE BE IT RESOLVED that the 46th District Democrats urges the King
County Executive and the King County Council to thoroughly and completely evaluate all
options for equipment that will enable All Vote-By-Mail in King County in 2008--including
the use of existing equipment, new equipment or a combination therein; and

BE IT FURTHER RESOLVED that the 46th District Democrats urges the King County
Executive and the King County Council to require King County Elections to conduct a local
public demonstration of equipment, new or used, that would be utilized in the 2008 elections,
and that the local public demonstration and testing of the equipment be completed before
King County commits to purchasing new equipment.

Adopted by the 46th District Democrats on Thursday, July 19, 2007.

(link to original)

---

So here's the thing. The intent of the above resolution is pretty clear: Try before you buy. Common sense.

During our Q&A session with Bill Huennekens last week, he responded that "Diebold will not get a cent before this equipment is certified and tested."

I rebutted by saying "You and Diebold are trying to push through a contract with the county right now." Huennekens acknowledged that was true.

Ah ha. Huennekens is making a distinction between having a contract and buying the equipment. A nominal distinction if there ever was one. And enough to placate the viewing audience. (For the most part, whenever we tell people what's happening right here in King County, they get pretty grumpy pretty quick.)

I'd prefer our county was a bit more skeptical. The intent of our resolution is pretty clear. I used the analogy of how buyers always test drive their new cars.

Let Diebold show that the DRS et al works as advertised. Then, and only then, should the county consider contracts, purchase orders, statements of work, or whatever else one wants to call it.

Thanks for listening.

by zappini on Sat Jul 28, 2007 at 08:16:46 AM PST

---

As I understand it, the CA Secretary of State (D) required a report from an independent expert panel on security of the voting machines they are using (3 different systems). Here's a DKos article on the panel's findings. The article's author cites this as a key finding:

The short time allocated to this study has several implications.  The key one is that the results presented in this study should be seen as a "lower bound"; all team members felt that they lacked sufficient time to conduct a thorough examination, and consequently may have missed other serious vulnerabilities.  In particular, Abbot's team reporter that it believed it was close to finding several other problems, but stopped in order to prepare and deliver the required reports on time.

[emphasis theirs]

In the meantime, I'm told that the King County Council will receive a briefing on the new election equipment from its expert panel on Monday morning--just before it's scheduled to consider the purchase of the equipment. Will they have time to digest the report and make thoughtful decisions when they consider the purchase?

by DWE on Sat Jul 28, 2007 at 09:57:44 AM PST

---

Ron Sims is wrong. Since this has all happened on his watch... how can he be so wrong?

Diebold didn't create the Accuvote, they bought the company. To the best of my recollection, Diebold bought their way into the business here in King County by buying the company which originally produced the Accuvote and sold it to King County Elections.

Problems? We got problems! Notwithstanding that the problems which occurred which led to the formation of the CEOC weren't specifically hardware-related, the 2007 primary and general elections were the only times that I am aware of where the Accuvote tapes agreed with the pollbooks at the end of the night, or when they didn't the discrepancies were explicable. In previous elections there have been pollbooks where there the only thing to write on the line where you're supposed to explain the discrepancy was "I don't know!"

This is the problem with going to all-vote-by-mail, beyond any technical or logistical considerations: it is going to take this process and put it out of sight and out of mind, and make it that much harder to refute when someone like Ron Sims makes bold and wrong assertions which, at this juncture, given the gravity of the situation he should have thoroughly checked. If he hasn't checked them by now he's incompetent; if he did, then he's a liar.

by m3047 on Sun Jul 29, 2007 at 01:03:58 PM PST

---

I recall reading somewhere recently that a significant number of the "5 million ballots" in London weren't actually vote-by-mail, that that represents the total number of ballots. Furthermore, that the ballots were vastly different, far less races.

I also recall something about GEMS not being able to handle the number of projected ballot styles or something like that, and that was why we had to go to the new equipment; that there was an upgrade option but that that was not being pursued because doing so would undermine the case for going to the new equipment.

by m3047 on Sun Jul 29, 2007 at 01:23:12 PM PST

---