Slowly, Sunshine Creeping Into Texas E-voting Process
Prior to this week, Texas voters were effectively locked out of the decision-making process surrounding the certification of voting technology, electronic or otherwise. In an insular process that unavoidably sheltered them from the concerns of concerned citizens, state examiners previously met in private with e-voting vendors to discuss the pros and cons of proposed systems. Predictably, without counterbalancing voices available to grill vendors on system flaws, examiners have repeatedly found little reason to refrain from recommending certification to the secretary of state. And the secretary of state, relying exclusively on examiners, has always followed their recommendations.
Times change, albeit reluctantly. On Monday, voting examiners and three vendors appeared (courtesy of a lawsuit brought by EFF and the ACLU) and faced the public. "Enthusiastic" doesn't exactly capture the mood of the court-ordered participants. One examiner, Barney Knight, used his opportunity for opening statements to explain that he had no intention of responding to questions from the public or of participating to any greater degree than required. An Accupoll official rolled his eyes at early technical questions. In the very early going, the process seemed destined to play out as little more than administrative theatre.
Funny what happens when you crack open a closed administrative process to the public, though. Concerned citizens, bloggers, election officials, disability rights advocates, computer scientists, sociologists, and yours truly peppered the participants with detailed questions and observations about a wide range of topics concerning the proposed systems, using up no less than the full three hours of time allotted for public comment. Particularly active and effective was Rice University computer science professor Dan Wallach, who ensured that neither the examiners nor the vendors could avoid tough questions without consequence. A testament to the absurdly opaque procedures that for too long have shrouded much that is e-voting, even the most straightforward questions had particular resonance. Why aren't you encrypting vote data that is submitted by modem? Why can election officials alter election results with the right password? Why don't certain machines not automatically zero-out vote totals before voting begins? What's wrong with open source software? And to Diebold, what have you done since 2003 to fix the breathtaking security flaws exposed by computer science officials in 2003?
Off-script, emboldened, several examiners actually said that the forum was useful and brought an important measure of accountability into the process. Examiners Tom Watson and Nick Osborne went even further and made remarkable on-record observations that current Texas certification procedures were not good enough, noting that they were not able to inspect the proposed systems in detail, that they were forced to largely rely on what vendors told them, and that it would be extremely difficult for them to discover serious flaws given the current shape of the process.
As encouraging as this session was, it was not nearly as in-depth as it could have been. The public still is not able to personally witness or participate in the underlying examiner meetings. (The subject of Monday's forum was a meeting held in early January in which vendors demonstrated the actual technology at issue.) And even worse, Monday's gathering was a temporary measure, a stop-gap product of ongoing litigation that could evaporate if a judge rules against us in July. And the Texas legislature could decide on its own in the coming months whether the public has the right to monitor a process that does nothing less than decide how well the public's right to vote is protected. That this question is not already settled -- in Texas and much of the rest of the country -- is mind-boggling.