Friday, April 23, 2004

Voting Machines In Indiana Highly Suspect


Voting machine firm must audit system, Election Board rules

The three-member Election Board pummeled a company representative Thursday over why the incorrect software was installed and why the company did not tell county election officials there was a problem..
"We can't afford to make any more announcements about mistakes to our voters," said Marion County Clerk Doris Anne Sadler, one of three members of the Election Board..
Sadler said she had seen an e-mail from ES&S headquarters directing the company's local representative, Wendy Orange, to lie about the software and pass off an attempt to replace it as routine maintenance.
"I don't know that I can explain it," said Ken Carbullido, senior vice president of product and software development for ES&S, who was under oath Thursday. "It was a mistake not to be entirely forthcoming."


No....really?

There's one quote that I think sums up everything that's wrong with electronic voting today, from one member of the Marion Election Board to ES&S (Election Systems and Software, a company I have talked about much less than Diebold):

"What else don't we know about what you went in and changed?"


Precisely.
Gosh, You Mean, They're Not Secure?!



Jakarta vote monitoring centre shut down


There have also been several attempts to hack the system, including one successful attack on Sunday in which parties' names were changed.

Paper Ballot Thrown Out In Maryland



Discarded paper ballot draws ire of voter

...22 people in Howard County who filled out a paper ballot in the March 2 primary election rather than use the new touch-screen voting machines, and whose votes were thrown out.
Under the touch-screen system, those 22 paper votes didn't count, county elections officials said.


This is such bullshit. Didn't count?

I can tell you that in CA, the reasons we (election officials) were given as to why a provisional ballot might get discarded were that the voter wasn't actually entitled to cast that vote at that time in that place:

1. they'd already voted in their own precinct and were casting a provisional ballot in ours;
2. they weren't actually registered as the party they were trying to vote in (if there were partisan ballots);
3. they weren't actually registered to vote in the county or state;

etc.

NOT: "We wanted you to vote using a different method than you used with your provisional ballot."

So, here's my question: did the provisional ballots I processed in March's CA Primary get thrown out? What about my own absentee ballot?



Tuesday, April 20, 2004

Diebold Knew


Attorneys warned firm that use of uncertified vote-counting software violated state law

Diebold knew of legal risks



Well, big, fat, hairy, white, rich, republican, surprise.

Attorneys for Diebold Election Systems Inc. warned in late November that its use of uncertified vote-counting software in Alameda County violated California election law and broke its $12.7 million contract with Alameda County.
Soon after, a review of internal legal memos obtained by the Oakland Tribune shows Diebold's attorneys at the Los Angeles office of Jones Day realized the McKinney, Texas-based firm also faced a threat of criminal charges and exile from California elections.
Yet despite warnings from the state's chief elections officer, Diebold continued fielding poorly tested, faulty software and hardware in at least two of California's largest urban counties during the Super Tuesday primary, when e-voting temporarily broke down and voters were turned away at the polls.


I was there, kids.

More than 600 of the devices froze or displayed unfamiliar screens and error messages on the morning of Super Tuesday, for failure rates of 24 percent in Alameda County and about 40 percent in San Diego County.


Yep.

In fact, Diebold engineers were writing and rewriting the software at DESI headquarters in Texas and in Sacramento, supplying the latest versions two weeks before the encoders failed at high rates in the Super Tuesday presidential primary.


Yep. The night before we received the machines, as all of us (system inspectors and assistant system inspectors) were told during our training and supplies pick-up, Diebold technicians had been installing software on them. The UI had changed from what the trainer's script instructed.

What else had changed?







Monday, April 19, 2004

Machine Voting


Faking Democracy: Americans Don't Vote, Machines Do, & Ballot Printers Can't Fix That, by Lynn Landes.

Landes makes an important point:

In Bush v. Gore, the Supreme Court said that a "legal vote" is one in which there is a "clear indication of the intent of the voter." Voting machines (lever, optical scan, touchscreen, the Internet, etc.) produce circumstantial evidence of the voter's intent, at best.Think of voting as a three-step process: marking, casting, and counting ballots. Once a machine is involved in any one of those steps, the result is hard evidence of the machine's output and circumstantial evidence of the voter's input.


I'm not sure we can 'step back', once so large a group of officials have committed to moving 'forward', and that unconscious feeling may be why scholars and and political scientists aren't debating this question.

But is it true that "the only fix that will give Americans back their constitutional right to vote is to ditch the machines", due to the need for voting process transparancy?

I disagree -- any scenario we can construct where the voter, at the time of placing the vote, can confirm that this is the vote I meant to cast, and the person, process, or machine counting that vote can effectively reconstruct the vote the voter meant to cast, is a workable scenario.

The real question is, what is the best way to achieve that scenario? No technologist I've spoken with believes electronic voting machines in use in America are sufficiently robust to meet this criterion.

None. Yet we've voted with some accuracy and precision in the past. In science, accuracy means "the correct value", and precise means "reproducible". Voting with paper ballots may not be 100% accurate, but we know and can measure how inaccurate it is.

As long as we cannot do that with machines, they are not a step forward, but in fact, a step back. So, for now, though Landes and I disagree in means, we agree in ends -- we should return to using paper ballots, because the machines don't do the job.