TALLAHASSEE - Using a newer laptop to do your taxes? Planning to try your luck at the slots in Atlantic City?

Chances are those devices had to meet tougher technology standards than the electronic voting machine you used to cast a ballot in the general election last month.

That's what a contingent of frustrated computer experts sees as the real problem with electronic voting. While lawmakers and attorneys debate the need for a paper record of votes, engineers and programmers say their concerns run deeper - down to the design and security standards under which voting machines are produced.

State and federal standards fall short of what industry or government requires for off-the-shelf commercial software, aircraft electronics, even casino slot machines, critics say. With democracy at stake, they say, shouldn't the bar for voting machines be set as high?

On Dec. 7, the U.S. Election Assistance Commission updated its standards for federal certification of electronic voting machines. Compliance by states, which set their own standards, is voluntary.

About two weeks earlier, scientists at the National Institute of Standards and Technology produced a report that assailed the design and security of voting machines. The institute advises the election commission.

"Much evidence has been produced that voting systems in general are not developed according to rigorous models of secure code development nor tested with the rigor of other security-critical applications," the institute's team reported.

The comments echoed a 2005 report from the Government Accountability Office, citing "weak security controls, system design flaws, inadequate system version control, inadequate security testing, incorrect system configuration, poor security management and vague or incomplete standards, among other issues."

Resulting problems, the authors wrote, have ranged from machine breakdowns to the ability to alter stored votes without leaving a trace.

The standards need strengthening, said Doug Jones, a computer science professor at the University of Iowa who sat on the Iowa Board of Examiners for Voting Machines and Electronic Voting Systems from 1994 to 2004. He was chairman for three terms. "The security standards, at this point, seem to be designed with deliberate loopholes in them in order to grandfather in existing products, and they ignore large areas of well-known practice in the world of secure computing," Jones said.

The issue is not just one of voter confidence, said Stan Klein, a software engineer in Gainesville.

"It's the means by which the consent of the governed is determined," Klein said. "In my view, it has to meet the same criteria as the computer systems that contain eyes-only information for the president of the United States."

By What Standard?
The Election Assistance Commission is a new entity and continues to revise its guidelines, Commissioner Gracia Hillman said.

Although insisting the machines in use now are secure, she said, "I think, within the next few years, the security standards for voting machines will begin to meet what people would feel very comfortable with."

Even among the severest critics of the status quo, there is little agreement on what better standards would be.

Vincent Lipsio, a Gainesville software designer conducting tests for aircraft giant Boeing, has suggested holding voting machines to a standard comparable to the one used by the Federal Aviation Administration for aviation electronics.

"The software sections of the [federal] voting-machine standards are only a few pages long; they essentially look like things extracted from an introduction to computer programming," said Lipsio, who sat in 2001 on a panel of engineers and industry representatives trying to come up with a standard for voting machines.

"FAA's standards are maybe 120 pages of nothing but how you have to review, re-review, proof, peer-review and sign off on the system."

The voting-machine industry representatives in the committee revolted against the suggestion, he said.

Of course they revolted, said Jones, the former elections examiner from Iowa. Applying FAA-style requirements to voting machines would increase by tenfold the price of $3,000 to $5,000 each, he said. "Voting machines would be used very sparingly."

Standards for electronic slot machines might offer more practical guidance. "They're extremely serious about the defenses they take," Jones said. "The audit teams make entirely unexpected visits to the casinos; they do spot checks that involve opening the machine, pulling the chip out, sticking the chips into their testers to make sure it's the chip that's supposed to be there. I don't know of any jurisdiction that does checks on voting machines like that. I wish they did."

Money remains the issue, said Hillman, the federal elections commission member.

"A lot more money is available for the highest technology with slot machines or with ATM machines. Heretofore, nothing close to that has ever been made available to spend on voting machines."

'Already Failed'
U.S. Rep. Robert Wexler, a Democrat from Boca Raton, is at the forefront of the campaign in Congress to require that voting machines used in elections produce a paper trail.

"Three years ago I'd have said, 'Yes, Congress should specifically ensure higher standards for these machines,'" Wexler said. "But in my mind, these machines have already failed - both in terms of their reliability and their inability to produce a voter-verified paper trail. I don't need any more standards."

But Jones said many models that create paper trails print on flimsy, fading paper. Other critics have complained that the printers jam or cause other problems.

Wexler said the paper-trail legislation he is pursing with U.S. Rep. Rush Holt, D-N.J., will include requirements for paper-trail machines that solve those problems.

Incoming Republican Gov. Charlie Crist and Kurt Browning of Pasco County, Crist's secretary of state, also said Thursday that they are open to requiring paper trails.

Rebecca Mercuri, a software security engineer who has testified in Congress about voting machine standards, has argued for applying the federal government's standard for defense and other security-sensitive technology to voting machines.

The federal government's litmus test for computer system security and reliability is a series of standards known as the Common Criteria. The Defense Department requires security-sensitive technology to comply with high levels of the standard, but commercial vendors - Linux, Apple and Microsoft - also started earning Common Criteria certification for their off-the-shelf products several years ago.

Mercuri said the voting-machine industry balked at adhering to the Common Criteria.

"If it's not their way, it's not acceptable," said Jill Friedman-Wilson, of Election Systems and Software, the company that manufactured paperless voting machines in Sarasota that are at the center of the contested Vern Buchanan-Christine Jennings congressional race.

ES&S does not oppose considering Common Criteria as a standard, Friedman-Wilson said in an e-mail. But, she wrote, "it is important to ensure that in applying Common Criteria - or any other standard - that it results in practical solutions that the marketplace can bear."

Government will decide what the market will bear, Mercuri said, because manufacturers will have to produce what the standards require.

Experts at the secretary of state's office wrote Florida's voting-machine requirements. Mercuri raised concerns about those standards - among them, inadequate requirements for software testing and review, insufficient memory storage requirements and insufficient climate controls for the machines.

Sterling Ivey, spokesman for the secretary of state, took issue with some of Mercuri's charges but also said the Division of Elections will review all its standards, beginning next month. Some may require technological updating, Ivey said.

"When that analysis is complete," he said, "Florida will once again be a model for other states to follow in election reform."