If you knew specifics about voting processes, you would know that I'm not suggesting heavier workloads. In fact, currently, precincts using DREs (direct recording electronic voting machines) usually employ twice as many pollworkers as equally sized precincts using OS because of the specialized knowledge you need if something goes wrong. Maybe a pollworker has to feed ballots through a machine, which takes 10 seconds each. Is distributing and collecting voter access cards for DREs dramatically less work?
If you have DREs that print voter verifiable ballots, what happens if there is a malfunction on election day? If you have 4 electronic voting machines to serve the precinct population as the average precinct does, you have just lost 25% of your capacity to serve voters. I'm looking at this through the lens of thinking of an election day as a unique event that you have to get right. Therefore, everything needed for the election day should be in place, not created "on the fly" as with a DRE with paper trail. Pencil breaks? Get another one. Optical scanner breaks down? Save all the ballots and scan them later. It's a durable process. If a DRE breaks down, you've just decreased your ability to let people cast ballots. Any technology that requires ballots to be generated on election day as the potential to fail on election day. Hence, even if we had DREs that printed ballots, precincts would still feel the need to generate backup paper ballots prior to election day.
What if the arrival rate peaks at lunch time and right after work more than anticipated? What if you have a higher turnout than anticipated? You can expand OS capacity by adding additional privacy booths, whereas you are SOL with 4 expensive voting machines and no expandable service capacity.
I brought in "side issues" for two reasons. First of all, "ease of voting" isn't the only criteria upon which someone should judge a voting system. As I mentioned, there are many criteria. The combination of cost, reliability, ease of use, and security of optical scan far outweighs any gains from more complicated, more prone to fail, more expensive DREs, with or without paper trail. Optical scan ballots are NOT butterfly ballots, which haven't even been mentioned in this conversation.
Second of all, when someone makes the ridiculous claim
Electronic voting machines are in virtually every way superior to paper voting machines. then there are no "side issues"..."virtually every way" means every single relevant aspect of comparing OS to DRE are fair game. If you really think claiming there are no ways that optical scan paper ballots are superior to voting machines is "insightful commentary", you're cracked and know nothing about voting.
IMHO, there is no argument that justifies the costs and problems of DREs, in any iteration. I have heard some respectable arguments for DREs. But I don't think anyone who knows anything about the nuances of creating a voting process would argue that DREs are "superior in every way". And don't get me started on software upgrades, longevity issues, maintenance costs, replacement parts, licensing, and a boatload of other problems with DREs that most people don't even know about. Have you ever talked with ES&S or Diebold? Have you ever talked with an elections supervisor? Everyone thought DREs were a good idea until they actually realized what they were getting into. There is a reason why a lot of areas are abandoning this technology, and the reasons hold whether or not you have a voter verifiable paper trail.
As someone who's done some academic research on voting technology, I'd like to respond.
Electronic voting machines are in virtually every way superior to paper voting machines. Um...
They prevent you from accidentally submitting an invalid ballot. So do precinct count optical scan ballots (i.e. scantron). The way it goes is that you fill out your ballot and then a poll worker scans it through the machine to make sure you have no overvotes or doodles outside of the designated boxes. If you screwed up, your ballot is destroyed and you get a new one and re-vote. This doesn't happen for central count optical scan ballots (where they box them all up and take them to a central location to be scanned) but central count optical scan set-ups are being phased out.
They can be updated with a correct ballot much easier than actually printing ballots. Actually, precincts are required to print out backup ballots to use should touchscreen machines go down. So really each precinct is running (and paying for) a backup election with paper ballots even when they use touchscreen electronic voting machines. Even if they spit out a paper trail, a precinct is going to need backup paper ballots in the event of a printer malfunction. This kind of negates the whole argument of being able to change-up a ballot on the fly, because once those back-up paper ballots are printed, the precincts are committed to a set ballot.
They can more easily accommodate voting by the disabled. This is a legitimate argument, but one electronic voting machine per precinct specifically for disabled people makes more sense than buying several to serve all voters. And anyway, optical scan paper ballots can be easily adapted for disabled voters http://www.wired.com/science/discoveries/news/2006 /01/70036
They can randomly display the list of candidates, eliminating the 'first ballot position' advantage. You can also argue that a random listing of names would make candidates harder to find than an alphabetical listing. I don't think this is such a big gain when you consider the cost, security issues, and possible malfunctions that can occur with electronic voting systems. With optical scan, the worst case scenario is that the scanner goes down and ballots have to be saved and scanned once the scanner is fixed. With electronic voting machines, regardless of paper trail, if there is a malfunction, the machine is down and you've just lost a huge part of your ability to serve potential voters. Then you have long lines, people pissed off, people deciding they aren't willing to wait and not voting, etc.
I have yet to hear a reasonable argument for electronic voting machines over tried and true optical scan ballots for any criteria - security, cost, usability, convenience, etc. On election day, you only get one shot to serve all the voters. Best to have a reliable and secure voting system than a bunch of fancy machines that have the real potential to crash. Not to mention, based on some research done by a colleague, electronic voting machines cost over twice as much as an optical scan system per ballot cast. And the serving capacity for an optical scan set-up can be expanded by buying cheap plastic privacy booths rather than another expensive machine. I know slashdotters usually have a boner for technology, but learn a little about running elections before you bring that bullshit to the polling place.
Slashdot Mirror
If you have DREs that print voter verifiable ballots, what happens if there is a malfunction on election day? If you have 4 electronic voting machines to serve the precinct population as the average precinct does, you have just lost 25% of your capacity to serve voters. I'm looking at this through the lens of thinking of an election day as a unique event that you have to get right. Therefore, everything needed for the election day should be in place, not created "on the fly" as with a DRE with paper trail. Pencil breaks? Get another one. Optical scanner breaks down? Save all the ballots and scan them later. It's a durable process. If a DRE breaks down, you've just decreased your ability to let people cast ballots. Any technology that requires ballots to be generated on election day as the potential to fail on election day. Hence, even if we had DREs that printed ballots, precincts would still feel the need to generate backup paper ballots prior to election day.
What if the arrival rate peaks at lunch time and right after work more than anticipated? What if you have a higher turnout than anticipated? You can expand OS capacity by adding additional privacy booths, whereas you are SOL with 4 expensive voting machines and no expandable service capacity.
I brought in "side issues" for two reasons. First of all, "ease of voting" isn't the only criteria upon which someone should judge a voting system. As I mentioned, there are many criteria. The combination of cost, reliability, ease of use, and security of optical scan far outweighs any gains from more complicated, more prone to fail, more expensive DREs, with or without paper trail. Optical scan ballots are NOT butterfly ballots, which haven't even been mentioned in this conversation.
Second of all, when someone makes the ridiculous claim Electronic voting machines are in virtually every way superior to paper voting machines. then there are no "side issues"..."virtually every way" means every single relevant aspect of comparing OS to DRE are fair game. If you really think claiming there are no ways that optical scan paper ballots are superior to voting machines is "insightful commentary", you're cracked and know nothing about voting.
IMHO, there is no argument that justifies the costs and problems of DREs, in any iteration. I have heard some respectable arguments for DREs. But I don't think anyone who knows anything about the nuances of creating a voting process would argue that DREs are "superior in every way". And don't get me started on software upgrades, longevity issues, maintenance costs, replacement parts, licensing, and a boatload of other problems with DREs that most people don't even know about. Have you ever talked with ES&S or Diebold? Have you ever talked with an elections supervisor? Everyone thought DREs were a good idea until they actually realized what they were getting into. There is a reason why a lot of areas are abandoning this technology, and the reasons hold whether or not you have a voter verifiable paper trail.
I have yet to hear a reasonable argument for electronic voting machines over tried and true optical scan ballots for any criteria - security, cost, usability, convenience, etc. On election day, you only get one shot to serve all the voters. Best to have a reliable and secure voting system than a bunch of fancy machines that have the real potential to crash. Not to mention, based on some research done by a colleague, electronic voting machines cost over twice as much as an optical scan system per ballot cast. And the serving capacity for an optical scan set-up can be expanded by buying cheap plastic privacy booths rather than another expensive machine. I know slashdotters usually have a boner for technology, but learn a little about running elections before you bring that bullshit to the polling place.