Is my suggestion from '05 still crazy? Postcard Affidavits to check results?

[mod mom]

Here's my concern...I attended an Election Observer training put on by the Democratic Party in my state this past weekend. One of the things listed to immediately place a call into the Party boiler room is incidences of "VOTE FLIPPING". Now I recall being called a conspiracy theorist after '04 for even mentioning the occurrence. There was a State Senator from Ohio who was told not to go public with his experience. Now we have a party admitting this occurs but to wait until it is noticed to notify so that machines can be taken off line on election day. What if the voters miss it? What if they don't check but assume they have chosen the candidates they prefer?

So isn't this admitting that the machine tabulation might be fixed (which we know from multiple studies) but we are going to HOPE that we get a fair election?

I suggested back in '05 that when results vary from expected, the Party organize a postcard affidavit program where voters can sign that they chose X candidate. If the Dems can get enough signatures, it seems to me that they can prove that the results of DREs are in fact tainted. Of course, this would have to be organized by the campaign to effectively get enough involvement. I was shot down in '05 told this was impossible, but knowing what is at stake, isn't this one way we can challenge an outcome?

In my county we already have machines impounded and treated as a crime scene:

Sunday, March 16, 2008 3:22 AM
BY BARBARA CARMEN
THE COLUMBUS DISPATCH

When Jennifer Brunner cast her vote last fall, she is certain she saw something so odd on her touch-screen voting machine that it prompted a state criminal investigation into the Franklin County Board of Elections.

At least 15 of the county's electronic machines are under double-lock at an Alum Creek warehouse. It is being treated as a crime scene.

County elections officials asked the Ohio Bureau of Criminal Identification and Investigation to seize the machines during the investigation by Attorney General Marc Dann and forensics consultants.

Brunner said consultants from SysTest Labs in Colorado, however, were skeptical. When she described the gray box with the faint words "candidate withdrawn," the investigators told her, "That's exactly what you'd see if someone masked a name."

A SysTest report notes that voters in other precincts -- in Victorian Village, Clintonville and Hilliard -- also reported seeing "candidate withdrawn" on their machines.

SysTest investigators also found that the board had not performed a routine test of the computer software on each machine, instead testing just one machine in each precinct.

-SNIP

SysTest investigators also found that the board had not performed a routine test of the computer software on each machine, instead testing just one machine in each precinct.

-SNIP

Investigators also discovered that a board programmer turned off "audit logs" in the voting machines in April 2007, hindering investigators from reconstructing software changes. White found that the vendor had instructed a board employee on how to disable audits to speed programming.


http://www.columbusdispatch.com/live/content/local_news/stories/2008/03/16/BOEPROBE.ART_ART_03-16-08_B1_9F9LIV3.html?sid=101

THE MACHINES CAN BE TAMPERED WITH, SHOULDN'T WE HAVE SOME BACK UP PLANS IN WHICH TO CHALLENGE THE OUTCOMES?

[OnTheOtherHand]

1. education on the paper records may be the priority right now

Brunner should be announcing a post-election audit directive (and Franklin was part of the pilot audits), so if the paper records are (even approximately) correct, there can be the wherewithal to correct any wrong results. There are lots of "if"s down that road, but I think it ought to be easier to teach voters to check the paper records -- and to squawk if there is a discrepancy -- than to get them to send in postcards.

Also, numbers seem to favor the paper record approach. If 5% of votes were flipped, tens of thousands of voters in Franklin County would have the opportunity to detect and report the problem -- whereas with postcards, the problem can't be authoritatively demonstrated unless almost everyone submits a postcard.

I'm not flat-out against any "parallel election" approach, as long as it doesn't distract from other work that has to get done.

[mod mom]

2. There is only enough paper ballots for 25% of a precinct based on '04 turnout:

"On November 4, 2008, we will continue this best practice by asking every Ohio DRE precinct to
provide back-up paper ballots equal to 25% of the precinct turnout in the 2004 general election.
The cost to counties for offering the back-up paper ballots will be reimbursed by using federal
funding.

Poll workers will have the option of asking voters if they want a paper ballot, but will not be
required to do so. The Secretary of State's office will, however, provide four informational
posters which must be prominently posted in each polling location, alerting voters to the paper
ballot option."

-snip

http://www.sos.state.oh.us/SOS/Upload/elections/directives/2008/Dir2008-59.pdf

and as I mentioned in a previous post, although 30% of Ohio voters are expected to utilize early voting, they expect an historic 80% turnout, coupled with average voting times (over 8 minutes in '08 versus under 5 minutes needed per voter in '04 due to ballot initiatives) they expect long lines. Although pollworkers are not required to offer paper ballots to alleviate the lines, 4 posters will be displayed at each precinct to inform voters of the option.

BTW I clarified that the paper ballots will not be inputted by poll workers, but instead will be transported along with other materials down to the county BOE where they will be inputed to an Op-scan on election evening. The votes on paper ballots, will not therefore, be recorded on the precinct level slip that will be posted outside each precinct.

[OnTheOtherHand]

3. I'm talking about records, not ballots

iVotronic VVPATs in Franklin County, I think? I'm not saying I like them, I'm saying they will probably be audited, and Franklin County in fact audited them after the primary -- so voters need to be told to check them.

[mod mom]

4. iVotronic is correct and absolutely they need to be checked. (BTW this is the same machine

type that resulted in the 18,000 undervote during the Christine Jennings race in Florida.)

[OnTheOtherHand]

5. I guess so, except for the VVPAT

which is not my idea of good voting technology by a damn sight, but does offer some benefits if voters take advantage of it (and if the printers mostly work!). At least the Franklin County audit seemed to go a lot better than the infamous Cuyahoga County audit where basically nothing matched anything else.

Oh, and while I'm being semi-cheerful, from what I've seen, Jane Platten at Cuyahoga County BoE seems like a big improvement.