Hart's E-slate system gave 2 diff. tallies for our votes last month:

[elehhhhna]

NOW WHAT? Here's the email from a friend of mine--any advice? Thank you.


"I was asked to help audit the votes from three eSlate voting machines after the November 2005 election in Fort Bend County, TX.

The computer printed two sets of documentation from each voting precinct that was audited.

One report, the CVR (Cast Votes Records) had a total of 220 voters in Precinct 1012 and 1 under vote. The other report, the JBC tally tape had a total of 221 voters in Precinct 1012 and 2 under votes.

The Elections Manager got out the Hart manual that confirmed that blank ballots do not show up on the CVR report. One person had supposedly left without voting.

The Elections Manager has said that Hart needs to correct the problem so that both reports have the same information. I am concerned about the program in the eSlate. How can it print two different tallies? However, I know nothing about computer programming.

The majority of the election machine examiners at the Secretary of State's office in Texas are not computer programmers. I am concerned about that also."

[Scurmudgeon]

1. I don't understand.

"How can it print two different tallies?"

I think the system is performing exactly as it was designed to do. If the machine didn't print two tallies how would you know that someone had left without voting? You wouldn't.

Also, I don't understand why either tape would need correcting. The JBC only shows that someone didn't vote properly. This is probably for internal use to determine whether the voting system is easy to use and clearly it is if only one person couldn't figure it out. Neither tape should be corrected. If both tapes were always corrected to be identical election officials might not know whether the machines are efficient for all users.

I have to admit that I don't know any of this for certain. From a pratical standpoint it seems logical.

[Bill Bored]

2. I'd start by looking at the actual vote tallies.

Whether the uncast ballot was counted as undervotes or not, the tallies of the votes cast should still be the same on both reports, shouldn't they? So were they?

Is there a voter-verified paper audit record that the voters checked as they cast their ballots? If not, you're just looking at results reports.

You should be able to print out the ballot images one by one if there is any question as to the actual vote totals, but that's the best you can do without a paper record verified by the voters at the time each vote was cast.

Let un know what you find out.