Leter from Board of Elections about Electronic Voting Machines..Help Reply

[masmdu]

During the primary elections here I voted under protest and filled out a form at the time to tell my board of elections that I don't like voting on an electronic voting machine without a paper trail.

Their response follows.

After reading it, if you have any specific rebutals or counterpoints to the points he makes in the letter I'd appreciate hearing your views as I want to respond to this letter.

Also, is there anyplace in the US that has decided to switch to DREs with voter verifiable paper trails?

"Guilford County
Board of Elections

Dear X:

Direct Record Electronic (DRE) machines have been in use for 20 years in the US and since 1988 in Guilford County. In that time, no board of elections, the Federal Election Commission nor any other individual or group has ever produced a documented case of electronic voting system programming fraud.

The published speculations by various achedemcis and other "computer experts" have, in my view, failed to take into account the procedural safeguards boards of elections employ to insure security during any election. Without proper procedural safeguards, all voting systems are vunerable to fraud. Paper systems in particular, have a long and storied history of fraud asociated with them.

We can certaianly be grateful to these individuals for heightening the alert to potential fraud with DRE systems. We have, of course, been following the debate over the DRE machines since it started and will continue to protect the interest of Guilford County voters as new information and facts call for such action.

Our DRE machines can produce an audit trail that can be reproduced on paper. They do not automatically produce a paper audit or receipt for each vote. Again, I would suggest that once paper is produced it can more readily be used to perpetrate fraud that if it is preserved in the, more secure, electronic form.

At this point, the NC State Board of Elections has indicated that it will continue to certify and allow the use of DRE machines as they have been used in the past in NC, without a paper receipt, but with an audit trail that can be reproduced on paper.

Thank you for expressing your concerns and being an active citizen.

Sincerely,
XX
Director of Elections"

[masmdu]

1. kick

.

[Stephanie]

2. "An audit trail that CAN be reproduced on paper"

What is he TALKING about? He means he can print out a copy of the totals. You need to explain to him Voter Verified Audit Trail.

[masmdu]

4. Yes...essentially he is saying that they can provide a printout

to show results.

[rfranklin]

3. What brand of machines are they using?

And the reason that fraud cannot be documented is because there is no traceable trail left by a program that changes the votes and then disappears. And no one can check what the programming is because the code is "proprietary."

[masmdu]

5. Thanks...I will collect the ideas on this tread to form my reply

Sorry, I don't know the brand.

[masmdu]

6. By the way...has any place in the US switched to using DREs with voter

verifiable paper trail?

[0rganism]

12. There's at least one state in New England, Connecticut perhaps?

One of the smaller touch-screen vote machine companies, Avant, IIRC, specializes in paper-trail oriented e-voting.

The idea that having the machines produce receipts makes the system more easily defrauded is... laughable. I'll leave it at that.

Ideally, the machines would produce a legible ballot printout which would be counted by an optical scan machine in parallel with the electronic voting machine. The two counts would then be compared, precinct by precinct, for consistency. In the event of a mismatch, the optical scan ballots would have precedence.

[masmdu]

7. Or...Links to refute/rebut his points would be helpful...Anybody?

TIA

[Eloriel]

8. A couple of rebuttals

Proof of election fraud? (I'm going on Georgia situation, you'll have to adopt it, as necessary, for NC.)

How, exactly, could one provide proof of computerized election fraud?

1. Currently, the certification process at the federal level (and state?) provides insufficient examination (if any!!) of the code to even have discovered the glaring security faults outlined in the Hopkins and SAIC reports. How could it have detected fraudulent code??

2. No one can legally examine the code after the fact to look for malicious code.

3. "Recounts" and after-the-fact ballot images or audit trails are useless because they will reflect the tampered-with-vote (if there was tampering) rather than the voter's original intent. (And we know that audit logs can be tampered with, btw.)

4. With no voter-verified paper ballot, there is no truly independent audit mechanism.

How do you "prove" fraud in these circumstances?


Audit Trail
See above.


Paper = fraud
Much more difficult, much more localized and contained. Computer vote fraud can affect a whole state, and a whole nation. There's no excuse for not safeguarding our vote, even if it is inconvenient.


Procedural Safeguards
There are two threats that computer experts warn against:
1. For any system that has telecommunication capabilities (and they pretty much all do), there is no way to safeguard against internal access and manipulation or external hacking. It can't be done with our current technology.

2. Certification isn't catching the more glaring errors, it's certainly not going to catch well-concealed malicious code.

All the physical security in the world can't completely protect against these two threats. If we want to have elections we can trust -- at all -- we MUST have voter-verified paper ballots. We need other things as well (random automatic recounts with discrepancies triggering full recounts from the ballots, open source code, etc.), but voter-verified paper ballots is the absolute minimum. It will not prevent fraud, but if there's a robust enough audit (5 - 10% of each precinct), it could help discourage fraud.

Dan Wallach, one of the authors of the Hopkins Report, actually convinced me the other day when he appeared at the public forum at GA Tech that full manual counts of computer-generated voter-verified paper ballots is the right way to go. But I don't know how realistic that is.

Eloriel

[masmdu]

9. Great points...thanks

.

[masmdu]

10. kick

.

[Robb]

11. Not that it would be helpful

...But I'd tell him he meant "ensure", not "insure" in his second paragraph.

[masmdu]

13. kick

.

[Bushfire]

14. CT started using TSVM with a paper trail

http://www.nj.com/news/ledger/jersey/index.ssf?/base/news-0/106783995473540.xml

System that prints out voter's choice put to test

Monday, November 03, 2003

BY KEVIN COUGHLIN
Star-Ledger Staff

As elections go, tomorrow's may not seem terribly sexy in Connecticut. There are some municipal races; there are some referendums.

But these elections are big news to a small New Jersey company that hopes to show America the future of electronic voting.

Avante International Technology Inc. of Princeton Junction will deploy its new electronic "touch screen" system in four Connecticut towns. Avante is vying with three larger rivals -- Diebold Election Systems, Sequoia Voting Systems and Daneher Controls -- to impress voters and election officials in the Constitution State and, by extension, across the nation.

New Jersey and other states are scrambling to replace outmoded voting equipment before next year's presidential election, using millions of federal dollars aimed at averting a repeat of the 2000 fiasco in Florida.

What sets apart Avante's system is a printout that lets voters verify their electronic ballots.

more...

http://www.nj.com/news/ledger/jersey/index.ssf?/base/news-0/106783995473540.xml

[masmdu]

15. kick

.