← Back to Archives

Comments: NYT's "New Standards for Elections" Editorial

elections

The 2004 election may not have an asterisk next to it the way the 2000 election does, but the mechanics of our democracy remained badly flawed. From untrustworthy electronic voting machines, to partisan secretaries of state, to outrageously long lines at the polls, the election system was far from what voters are entitled to.

It's patently obvious that presidential elections, at least, should be conducted under uniform rules. Voters in Alaska and Texas should not have different levels of protection when it comes to their right to cast a ballot and have it counted. It's ridiculous that citizens who vote in one place have to show picture ID while others do not, that a person who accidentally walks into the wrong polling place can cast a provisional ballot that will be counted in one state but thrown out in another. States may have the right to set their own standards for local elections, but picking the president is a national enterprise.

This is obviously a job for Congress, and it deserves the same kind of persistent, intense lobbying effort that reformers have given the issue of campaign finance. But improvements by the states may be easier to achieve, and will clearly help prod Congress by their good example. Advocates should push every level of government to be part of the solution

It's about time we started thinking about standardizing elections for federal office. That would probably mean making federal elections a separate affair entirely from local elections (at least in some areas, others would probably just let the regulations for conducting a federal election trickle down into their local elections and thereby be able to combine the two into a single event/day). We need to establish a deep, detailed infrastructure if we are to use as complex a system as electronic voting and volunteer poll workers. The NYT goes a long way in making some necessary steps clear... I think this - the formal separation of federal elections from local elections - will not be easy to do (probably requiring a constitutional amendment) but is essential to make sure that people in one part of the country can trust the results from another part of the country (not to mention trusting the result from one's own part of the country).

The NYT is short-sighted in a couple places. Specifically (some of this is to appear in a forthcoming paper, if you'd like to know more, just ask):

  • NYT says: "The computer code should be provided to election officials, and made public so it can be widely reviewed." They don't talk about the bugaboos that open-source holds. I've been thinking about this a lot lately and will have a paper draft out on this soon... get a rough peak at the background for the paper here: "Open Source in E-voting".
  • NYT says: "There should be spot-checks of the software being used on Election Day, as there are of slot machines in Nevada, to ensure that the software in use matches what is on file with election officials." What we really need is parallel monitoring - where randomly chosen machines are voted on in precincts as "demo" machines with set scripts and video-recording - which is only done in a few states right now (California). Doug Jones has some great insight into parallel monitoring from working with Miami-Dade with it's ES&S iVotronic system.
  • NYT says: "Voting machine manufacturers and their employees, and companies that handle ballots, should not endorse or contribute to political candidates." Moreso, there should be detailed rules about conflicts of interest as are the norm in gaming (gambling) and legislative employment.
  • NYT says: "Registration forms should be simplified, so no one is again disenfranchised for failing to check a superfluous box, as occurred this year in Florida, or for not using heavy enough paper, as occurred in Ohio." Not only that, but registration support systems (sometimes called poll books - the computers some areas use to look up people and tell if they're registered to vote and haven't voted already) need to be scrutinized just as intensely as the software that casts and counts our votes. There were widespread reports of these systems failing and resulting in voters being turned away.
  • NYT says: "Election officials should get professional help to design ballots that are intuitive and clear, and minimize voter error." Further, there should be usability / human factors standards... not to mention privacy, accountability and conflicts of interest standards.
  • NYT says: "Absentee ballots should be widely available for downloading over the Internet." I'm not sure if this is a good idea... I'm pretty sure counties can't be convinced to do this. Absentee ballots are very specific documents with a non-trivial printing process (watermarked paper, serial numbers, seals, colored inks). Not to mention that there are laws in a lot of states that specifiy the absentee ballot process, medium, etc.