September 18, 2006
On voting machines and printers

Greg has gotten the discussion started on this Chron story regarding eSlate security. I'm going to pick one point of disagreement with Greg, and one point of agreement.

First, where we disagree:


Some critics of electronic voting believe the answer is printers that would produce "voter verifiable paper audit trails." If results are called into question, printer proponents say, a paper record exists of each vote and can be used for an official recount. Hart InterCivic's printer, which Harris County does not use, prints a voter's selections on a paper roll that the voter can review before leaving the booth. If a vote is wrong, it can be recast. Local Democrats say the county needs the printers, which would add up to $8 million to the county's $28 million investment in the current system.

[Harris County Clerk Beverly] Kaufman says Hart's printers have not been certified by the Texas secretary of state, and that printers are simply another moving part that can break and cause chaos.

A major national panel on which [Rice U. computer science professor Dan] Wallach served said this summer that printers should be used. But one computing professor from the Massachusetts Institute of Technology agrees with Kaufman. Ted Selker designed a study showing that volunteer subjects almost never caught errors he deliberately inserted in their printouts.


Greg is not a fan of incorporating printouts with the eSlates. I believe strongly that they are necessary as a backup and a sanity check. I'm not impressed by Professor Selker's study, at least not as reported here. I'll have to hunt it down to see if he's addressed this point, but it seems to me that if it were the case that printers were added to eSlates, voters would receive a certain amount of education about how to use them to verify that they cast the votes they thought they did. At the very least, I'd expect both local parties and a slew of candidates to include messages about checking receipts as part of their normal mailers. If you can convince me that no reasonable amount of such messaging will have a noticeable effect on the voters' ability to error-check, then I'll be forced to reconsider this position. For now, though, I don't think Prof. Selker has told me anything I wouldn't have expected.

Where Greg and I agree is that a lot of the rhetoric about voting machine security is counterproductive to progressive and Democratic goals. It is important to keep in mind that eSlates, with or without printers, represent a big step forward from punch cards, and any campaign to make them better needs to incorporate that theme. It's also still the case that ensuring the integrity of an election is at least as much about the people who run them and the processes that they use as it is about the equipment. That's one of the motivations driving the Secretary of State Project, which has gotten some positive press lately. As long as there's a perception that people in charge of elections have put partisan interests ahead of fairness, voting machines will be eyed with suspicion. And I can't say that I blame anyone for feeling that way, whether or not I like how they express that suspicion.

Posted by Charles Kuffner on September 18, 2006 to Show Business for Ugly People | TrackBack
Comments

I agree that the defeatism surrounding Diebold machines and the election process in general is unwise; it's a self-fulfilling prophecy of exactly the kind we don't need.

But (there's always a but) it's hard to ignore the incredibly dogged resistance to an incredibly straightforward idea of voter-verified ballots.

Take this Selker study. In a word -- well, two words -- so what? So some morons miss errors? That's good enough reason to deny everyone the chance to make sure their own ballot is correctly recorded? And I always thought MIT was a prestige gig.

And when you come up against resistance of this caliber over and over and over, you're strongly tempted to reach for the tinfoil hat after all: surely stupidity this persistent has to be bought and paid for? I'm reading Brave New Ballot right now, by Avi Ruben, the JHopkins prof who showed how hackable and shoddy Diebold technology was. And boy, the numbskulls this guy dealt with just beggar belief.

Posted by: Thomas Nephew on September 18, 2006 9:14 PM

I've always thought that electronic voting machines were a highy flawed solution. I voted in Alaska before moving to Texas and the entire state used optical scan ballots that were read and counted as you voted. You dropped your completed ballot into the big machine in the center of the room and it either read your ballot or spit it back out as unreadable. The results were then wired directly to the election headquarters when the polls closed.

With such a system you gain all the advantages of electronic systems. Immediate counting of the votes. You have voter verified ballots for recounts. And you even have ready solutions for disabled voters which is always a big selling point for the elecronic systems. For blind voters or other disabilities, you just set up a single Mac or PC attached to a printer with whatever accesibility features and software is appropriate. The blind vote using their braile keyboard or with voice commands and then print their ballot out which is then filed and read with the Optical scan machine with all the other hand-filled out ballots.

optical scan ballots also have two massive advantages over e-voting machines:

1. They are instantly scalable. If 100 voters show up at once you can hand each one a ballot immediately and they can fill them out sitting in a corner or on their lap if a booth is unavailable. No long lines like in Ohio.

2. They are unaffected by power or communication failures. If there are power or communication failures the precinct workers just collect the ballots and deliver them to the central location where they can be processed. Might delay results for that precinct by a couple hours at worst.

3. There are no electronic security issues. Sure there are always ways to cheat. But it's a hell of a lot harder to mess with thousands of original paper ballots to steal an election than with results that are only electronic.

This rush to electronic voting machines over the past 5 years has been the most misguided effort I have ever seen. A massive waste of money that ends up making the system worse than it was before in many instances.

Posted by: Kent from Waco on September 18, 2006 10:36 PM

Interesting info on the Secretary of State Project. Of course, it doesn't mean much for Texans, because Secretary of State is one statewide office that we don't get to vote on. At least, not currently.

Posted by: Jeb on September 19, 2006 1:51 PM

Evidence or No Evidence...

Evidence or Hidden Evidence...

Texas Constitution...quaint or not quaint...

http://www.dailykos.com/user/uid:67830

Now let's take a look at the Texas Constitution.

Article 6, Section 2 states:

"The privilege of free suffrage shall be protected by laws regulating elections and prohibiting under adequate penalties all undue influence in elections from power, bribery, tumult, or other improper practices."

Article 6, Section 4 states:

"In all elections by the people, the vote shall be by ballot, and the Legislature shall provide for the numbering of tickets and make such other regulations as may be necessary to detect and punish fraud and preserve the purity of the ballot box."

Now you tell me. How in the world can a system prohibit undue influence in elections from power, bribery, tumult, or other improper practices when there is no way to verify whether the things that are going on inside the machine are even recording the votes as the voters intend? How in the world can the purity of the ballot box be preserved when even the voter who cast the ballot has no way of knowing what actually got recorded on his ballot in the interior of the machine? Clearly, it can't, and it can't.

These machines clearly violate the Texas Constitution.

…………………………………………………………

Spend your time planning, organizing, and campaigning to defeat them. We have a Constitution to save.

David Van Os
Democratic Candidate
For Texas Attorney General 2006
Give me a hand at www.vanosfortexasag.com
Website: http://www.vanosfortexasag.com
Email: david@vanosfortexasag.com
1530 North Alamo Street, San Antonio TX 78215

........

Voting by absentee would seem to be abiding by the Texas Constitution for ballots.

However, since laws limit usage to those who will be out of the county for the duration of early voting and election day, being ill or disabled or over 65, or in jail but able to vote, many people cannot use this method.

see: http://www.sos.state.tx.us/elections/pamphlets/earlyvote.shtml

Plus, absentee voting is not perfect because there is the concern with the issue of chain of custody, and still the ballots are then scan counted, not hand counted for observable accuracy.

.........

Perhaps as tribute to the wisdom of the Texas Constitution requiring ballots, and as a measure that might help in the event of mishaps and errors, would be the suggestion by Lynn Landes.

This is a simple idea from Lynn Landes, for especially absentee voters, but also others, at:

http://www.ecotalk.org/VotingSecurity.htm

We must provide candidates with hard evidence of how we voted so that election results can be verified, or challenged, if necessary. Exit polls do not constitute hard evidence. Only voter affidavits can provide that. It’s time voters sign up and be counted.

..........

So, right after you vote, mail a letter to your candidate that has your name, signature, address, and precinct number with a note that states you voted for him/her. A candidate would then have evidence in hand, just in case.

It just might work.

I wonder what Van Os might think about this possibility. And, others.

.......

http://brainsandeggs.blogspot.com/2006/08/fixing-vote-and-then-fixing-that.html

In all, 1,218 voting machine complaints were filed in Texas in the 2004 general election with People For The American Way's Election Protection Division. In Harris County, 2,400 voting machine complaints were filed with a national voting advocacy group during that election.

In addition to these complaints, others were filed in Collin, Travis, Bexar and Wichita counties. Complaints included vote "transfers" (Kerry/Bush evidenced the same phenomenon reported in the 2002 and 2004 election in Harris County), lost votes, and machine and memory card failures. For the 2004 election, the Electronic Frontier Foundation and the Verified Voting Foundation received more complaints from Harris County than from any other voting jurisdiction in the nation.

........


Posted by: Prove Our Democracy with Paper Ballots on September 19, 2006 6:51 PM