“It’s beyond clear that the Sequoia Voting Systems machines in use in most New Jersey counties – and some in Pennsylvania, including Montgomery County – were an expensive mistake at best.”

It’s official: The Philadelphia Inquirer has joined the legions of those opposed to insecure, unauditable unrecountable direct recording electronic vote machines (DREs). And especially the Sequoias now in residence in Montgomery County and across New Jersey. Finally. Read their recent editorial here.

Are you listening, Montgomery County? You, who bought an additional 75 used machines for about $350,000 last fall, when you could have rented them dirt cheap. You, who bought 1,050 of them back in 1996. It’s time for a change.

You, too, Philadelphia, Delaware and Bucks, whose machines have the same basic architecture under the Danaher brand. When the facts are in and the arguments are weighed, the controversy pretty much dissolves into dust. Why employ questionable, expensive technology that is not only hackable but has no back-up when things go haywire. Why undermine voter trust?

Actually, The Inquirer has been a bit squeamish in its coverage of the DRE boondoggle in Pennsylvania. But fortunately, the weird vote machine saga playing out across the river in New Jersey has provided a platform for their renewed focus. We should be grateful.

Back in 2005 New Jersey took the enlightened approach of requiring that the state’s vote machines, almost all of which are Sequoias near identical to the ones we have in Montco, have voter-verified paper ballots. The idea was that, at the least, voters should have the right to a permanent record of their votes that could be recounted and audited. It was at the time a pretty strong statement by New Jersey lawmakers that the existing Sequoia machines were sorely lacking. And it was certainly beyond anything that Pennsylvania legislators could muster.

But since then, Garden State officials got way off-track and instead of ditching their machines for the sounder and safer technology of paper-based optical scan systems, they stuck with the vendor, Sequoia Voting Systems, in a deal to rig up printers to the existing DREs. Result: they’ve been dickering with Sequoia over unkept promises regarding minimum standards and delivery dates ever since. Deadlines were missed, extended and missed again. Apparently the technology doesn’t exist for a printer to be integrated effectively with the existing machines – something that the NJ election integrity movement had said from the outset.

Moreover, even if the paper trail worked – and there have been lots of problems with various systems around the country, it is a cumbersome process for voters to make their selections on a touch screen and then, in effect, vote again by reviewing a hard to read print-out that was supposed to reflect their original choices. And surveys indicate that voters forget or don’t bother with the print-out, anyway, thus defeating the basic rationale for the scheme. More to the point, the paper trail still doesn’t address the most fundamental weakness of the machines: firmware and software can be manipulated and that includes not only the internal memory of the machines, but the print-outs themselves. Paper is great, but without voter-marked paper ballots, paper trails are best saved for, well, parades.

Meanwhile New Jersey citizen-activists, led by the Coaliton for Peace Action, took the state to court in a case that parallels the one limping along here in Pennsylvania, Banfield v. Cortes (more on that soon). They are arguing, with pro bono help from the Rutgers Legal Clinic, that hackable untrustworthy paperless electronic vote machines violate the constitutional rights of NJ citizens. Princeton Professor Andrew Appel led a team of computer security experts in a comprehensive study of the Sequoia machines (hardware and software) that was ordered by the court. Read about it here.

Here’s what they found:

• The machines are easily hacked by installing fraudulent firmware, which can take as little as seven minutes to accomplish;
• Without even touching a single machine, an attacker can install fraudulent firmware into many machines by “viral propagation through audio-ballot cartridges.” The virus can steal the votes of blind voters, can cause machines in targeted precincts to break down or can cause the tally software to dysfunction.
• Problems with the user interface can result in uncounted votes and facilitate poll worker fraud.

What to do? According to Appel and his team, the responsible approach for elections officials is to acquire paper-based precinct-count optical scan voting systems along with a ballot-marking device for the disabled and employ them in conjunction with a scientific regimen of audits of the scanned paper ballots. The technology is far safer, more reliable and yes, cheaper, than DREs.

In fact, most states have gone in that direction. Only seven still use paperless DREs, including most of Pennsylvania and New Jersey. Since 2004, six states have passed laws to eliminate DREs, and none are still adopting the practice of attaching printers to the existing machines. Yet after more than a year late and a bungled policy, New Jersey is considering reneging on the whole thing. Let's hope that the court has a different view.

So what are the chances of Pennsylvania or, at least, counties in the Philadelphia area making the switch? It's difficult to say. There’s something about politicians and electronic vote machines. Think of it as in overcoming an addiction. Six states have conquered it since 2004 and others are on the verge, courageously taking their twelve steps. It’s hard work and subject to lots of back-sliding, but it’s doable. Citizens just have to support their officials to do the right thing – one day and one step at a time.