Election
Integrity

Was the 2004 Presidential Election Stolen? Exit Polls, Election Fraud, and the Official Count by Steven F. Freeman & Joel Bleifuss / Foreword by U.S. Representative John Conyers, Jr.

Read a PDF excerpt
Report on a hearing from 2 years ago
  About Blog Exit Polling Needs/Opportunities Reports/Presentations News/Commentary         Forum

Report on a hearing from 2 years ago

by junling 1/21/2008 3:11:00 PM

(Copied from Junling' blog on Yahoo)

February 16, 2006

Today California Senator Debra Brown held a hearing on California's Voting System and Federal Test and Certification Process. The hearing took place in Menlo Park this afternoon.

Four computer security experts gave testimony on their view on the securities of current voting system and their recommendations. These four experts are:
1.Aviel Rubin, Professor Computer Science, John Hopkins University;
2.Dan Wallach, Professor of Computer Science, Rice University;
3.David Dill, Professor of Computer Science, Stanford University and Founder of VerifiedVoting.org;
4.Peter Neumann, Computer Scientist, SRI International.

The first question asked by Brown was: What's the difference between electronic voting machines and other public machines such ATM, Slot Machine? If those machines can be made reliable (you always got correct dollar bills from ATM, and you trust the winning odds in a slot machine), then why is it hard to make electronic voting machine reliable?
Here are the answers from the panel:
1. ATM machines are subject to the scrutiny of the users. The Bank has incentive to make it right. If not, the user can switch bank right away.
2. The Gambling company has the incentive to maintain certain incentive in slot machines. Otherwise they cannot get customers.
3. It's much harder to know what the check and balance of a voting machine. There is no incentive for the manufactures to be accountable to the users. Therefore it is much harder to have reliable machines.
4. A machine is not trustable in the way that a human is not trustable. This is why we have a democratic system that allow different parties monitor each other, thus providing check and balance.
5. Machine are built by humans. We need to scrutinize a machine as much as we scrutinize humans. All parties who are involved in building, shipping, operating the machine should be subject to each other's monitoring.

Senator Brown asked: Is it possible to build a reliable program like those running in airplanes? How much does Boeing spend on their software testing?
David Dill answered: Boeing spent hundred of millions of dollar on their software testing. This cost is too high for an election software. The more fundamental problem is: No matter how much we spend in testing or certifying a program, we cannot trust that it will not be tampered after testing. Therefore an auditing process has to be in place. Having an auditing process allows us to have cheaper software and deliver it in time for election, and still ensure the integrity of voting.

At the end of the session, Brown asked each expert to give us their take-home messages.
Aviel Rubin: 1. A system has to be auditable so that we don't rely on testing or
certification.
2. It is very easy to put a bug in a computer program without being detected by others. Therefore testing or certification of a program is not effective in the sense it is hard to find problems there.
3. Auditing (such as paper printout) or parallel testing (doing mock election on the real machine when the actual election is going on) is far more effective.

Dan Wallach:
1. Voting integrity is about the process: Every step from manufacturing, to deploying, to counting the vote has to be checked.
2. Transparency is critical. The computer source code has to be open to the public. Therefore should not be any trade secret. The company can have copyrights, but no trade secret.

David Dill:
1. The Federal certification guideline for E-machines is too weak and allows many loopholes.
2. We need to have auditing for machines.

Peter Neumann:
1. The assumption that electronic voting machines are safer than punch-card machines is flawed. E-voting machines is no safer, and can be more dangerous.
2. All the source code for voting should be made public.
3. Every link involving the voting machines is weak. Therefore we must look at each link. The system can be tampered when creating the code, shipping the machine, operating the machine, or counting the vote.

In the end, Brown thanked the panel of experts. Next week, California will hold a hearing on Audit process for electronic voting machines.

Tags:

Add comment


(Will show your Gravatar icon)  

  Country flag





Live preview

1/6/2009 9:23:59 PM

 

 

Home | About Us | FAQ | Contact Us | for Candidates
©Copyright 2008, Election Integrity. All rights reserved.

 

 


Make payments with PayPal - it's fast, free and secure!


Recent comments

© Copyright 2009

Sign in