Here is the letter that I sent to Barbara Boxer:

I am writing you from Florida, but I volunteered in one of your campaigns when I lived in California and still receive mail from you.

I am writing to ask you to call for an investigation of the 2004 election.

The election was problematic in many ways. From my perspective, the most severe long term problem is the reliance on electronic voting technology that made recounts impossible, or on ballot scanning technology that allows recounts in principle but (at least in Florida) not in practice--when we recount, we re-analyze the same scanned image database as we analyzed before and, not surprisingly, get exactly the same result. We should be going back to the original paper ballots, but do not.

I am a Professor of Software Engineering at the Florida Institute of Technology. I am also an attorney, and when I lived in California, I served as a full-time volunteer Deputy District Attorney in the County of Santa Clara.

I am a member of the IEEE P1538 standard committee that is drafting the standard for electronic voting equipment that the Election Assistance Commission will rely on in drafting its regulations.

The IEEE committee is largely managed by voting equipment vendor representatives and, far from being a neutral technical voice, it is creating a standard that will meet the vendors' preferences. As a particular example, the standard will almost certainly not require voting equipment to generate hard-copy voter-confirmed vote verification or voter-checkable encrypted results. At the end of the day, voters and election officials will have to trust that the ESS and Diebold equipment gave us accurate and unbiased counts. We will have to rely on trust because there will be no independent way to check the truth of their reports.

The spin today is that the electronic voting systems were reliable, unbiased, and an improvement over previous systems. There were so few problems, according to the vendors, because the equipment is so good. It appears to me that there were few provable problems because the equipment makes it impossible to check the results.

Equipment like this is very appealling to some local election officials. If it is impossible to recount the ballots, we can save a lot of money on recounts.

On election day, I went to bed believing that the election had clearly and unambigiously gone to George Bush. I didn't like that result, but the data are what they are, and the margin seemed large enough to be a final result.

Over the next weeks, I read a variety of challenges to these results. Most were quickly dismissed, perhaps more quickly than carefully or thoroughly.

At this point, I don't think I know the actual result of the 2004 elections for President or Florida Senator (or several others). I know the reported result, but that comes from unverifiable technology that has been rationally questioned.

I have a probability estimate about the results (I currently think it is more likely than not that Bush and Martinez won their elections) but that's very different from a feeling of confidence. My discomfort appears to be shared by many other voters.

The sad part is that I can't know the result of the election. These other voters can't know it. You can't know it either. None of us can know the actual result because the reported results from the voting machines are not verifiable.

As a result, we all have to rely on spin and faith and power politics as the basis for accepting a final result. This is not an acceptable foundation for a societal structure that rests on popular acceptance of the rule of law.

We can do much better than this. We understand, technologically, how to create electronic voting machines that provide a good user interface and also provide strong security, voter verifiability, and later recountability.

We need hearings that can educate the public (and perhaps, other members of Congress) to the problems of unauditable elections.

It might be, today, that irregularities in the electronic equipment favored the Republican Party and perhaps, that some Republicans are less concerned about the equipment problems because they are satisfied with the result.

The problem is that insecurity can be exploited by anyone. If equipment can be tampered with, or programmed maliciously, the highest bidder is likely to be the one who determines the nature of the fix. I doubt that many Republicans would welcome the prospect of elections being fixed by the Triads or the Yakuza or El Qaeda.

A great deal of Federal money will be spent under HAVA to pay for new voting equipment. It appears that the new crop will be unauditable. At some point, Congress will wake up and demand auditability. Please understand this from a vendor's perspective--at that point, the government will have to buy a whole new generation of equipment, spending billions more to replace the equipment it just spent billions to buy. From a manufacturer's point of view, this is a winning scenario because it creates a longer cycle of high demand. From the public's point of view, it is the worst case and we should stop it now.

- Cem Kaner

Join with concerned organizations and individuals in encouraging Senator Boxer: www.ContestTheVote.org You can also write to her directly at: https://boxer.senate.gov/contact/webform.cfm

Cem Kaner (https://www.kaner.com/) is a Professor of Software Engineering, and Director of the Center for Software Testing Education & Research (https://www.testingeducation.org/) at the Florida Institute for Technology. He is the senior author of 3 books related to software quality, including Testing Computer Software, which is the field's best selling text (measured in terms of lifetime total sales). Along with his university teaching and research, Dr. Kaner has 17 years of experience in Silicon Valley developing software and consulting to software companies.