Why aren't touch screen voting systems designed to work like this?

You walk into the polling place. They check your name on the list and give you an access card. You walk up to the touch screen system, swipe the card to activate it, and make your selections. When you’re finished you press “done” and the system prints out a paper ballot that contains your selection. After you review your ballot to make sure it’s correct you walk over and hand the ballot to the poll worker. He feeds it into a scanner that tallies the vote, then places the ballot in a ballot box for safekeeping in case there needs to be a hand recount. The tallying box uploads its results over a phone line to election central, or is hand-carried there.

Advantages over the current touchscreen systems:

  1. Each individual voting station doesn’t need to be secure. It’s just a ballot-printing system. You only need to worry about security for the tallying box.

  2. Since the tallying box takes the paper ballots as its input there’s no opportunity for the voting station to print a false receipt.

  3. If there’s a question about the integrity of a tallying box it’s simple to test. Run the ballots through a different box and compare the results. Or do a hand count and compare the results. Fraud would be trivial to flush out.

  4. The technology for scanning machine-printed text is very reliable. Banks scan the tracking numbers on checks by the millions every day.

Basically it’s an updated version of the current punch-card system that eliminates that system’s major drawbacks. Ink on paper won’t be susceptable to the false negatives and positives caused by hanging or loose chad. It prevent voters from marking their ballots incorrectly. And since the ballot is easily readable by a human as well as a machine it makes hand recounts much easier.

Why aren’t electronic voting systems designed to work this way?

The technical reason might be the cost of the printing each vote- hardware plus paper plus ink. You have what I think is an excellent process but the bean counters wouldn’t want to spend the money.

I’m just cynical enough to believe that the reason it doesn’t work that way is that it would make it much harder to rig elections.

Here ya go.


Not exactly what you described.
You put the same paper ballot everyone else is marking by hand into this machine and vote using a touchscreen. (or keypad for the hearing impaired)
Then the unit marks the ballot for you and spits it back out.

The machine doesn’t tabulate votes though. It’s tabulated along with all the other hand-marked ballots with an optical scanner.
Basically it lets precincts comply with the HAVA (Help America Vote Act) while keeping their paper ballot solutions.

The OP has an excellent scheme. The printed ballot could have a checksum to make sure it is scanned correctly. Scanning in real time insures that if there is a problem you could destroy the ballot and re-vote. I like this a lot.

Source: http://www.verifiedvotingfoundation.org/article.php?id=6250

I’ve considered the cost issue. In order to create a paper trail many jurisdictions are demanding that electronic voting systems generate a paper receipt. Printing a ballot shouldn’t be any more expensive than printing a receipt.

There is the cost of having a “tallying station” with an optical scanner. However this should be more than offset by the amount saved by not having to make each voting station secure. I actually think this system would be cheaper.

Jurisdictions could even forgo the tallying stations and just hand-count the ballots. A standardized machine-generated ballot would be far easier to count than any current paper ballot.

How is this solution better than just having the voter fill in an Optical Mark Recognition form, like we already do here? I fail to see what the ballot-printing machine adds other than needless complexity and cost.

The touchscreen system can double-check to make sure the ballot is filled out correctly. And for elections with large numbers of candidates you can present them in a format that’s much easier to read if you don’t have to worry about cramming them all on a small ballot paper.

But, yes, I agree with you, low tech is generally better. However if municipalities are really eager to use touch-screen voting, they should at least have a system that doesn’t have major security holes … .

A touchscreen system can read the ballot to a blind voter.
A system can enlarge the text or present it in high contrast mode for a voter with visual problems.
A system can allow someone with mobility problems to use a foot pedal switch or a sip and puff device to vote.
A system can present the ballot in languages other than perhaps just the english that is on the paper ballot.

  1. How often are these obstacles preventing people from voting today?

  2. How much of an improvement would an electronic voting system that solves these concerns be over the existing low-tech systems that already addresses the problem of voting by the disabled/non-English/whatever?

Even as a dyed-in-the-wool geek, IMO electronic voting systems are a solution for a problem that doesn’t exist. The only thing they make easier is for untrustworthy people to steal elections.

Absolute drivel. :rolleyes:

If even 1 person found it easier to vote using the technology touch-screen machines offer, it would make them worth it. As it is, they offer considerable advantage to elderly people with poor vision, as well as most sight-impaired people period. Go talk to blind people and ask them how they like having to make their vote less-than-secret to be able to vote.

What I don’t understand about the OP is why all the seperate paper stuff is needed. When I voted in November, I first received an electronic review of my choices, then I saw them printed out on a paper journal, behind plastic where I couldn’t tamper with them. Then I “cast” them by touching the correct button. This way, there is a paper trail, there isn’t any need to worry about added machines to secure, and I can’t be accused of stuffing the ballot box with bogus versions of the paper the OP would propose.

It was simple, easy, and not worth screwing around with, IMHO.

Obviously you are not blind or disabled in a way that prevents you from voting with whatever system is presented to you. That’s fine, but there are many people with disabilities that up until recently have been unable to go to a voting location and vote in secrecy because they need assistance with the ballot.
The HAVA act gave states the 2006 deadline to have systems in place that allow people with disabilites to vote independently.

I have the distinct feeling nobody checked out the website I linked to.
That system uses the same paper ballots that everyone else in marking by hand. It’s really just an intelligent pen that helps out the voter.

Actually I think that electronic voting makes us a lot safer than the old fashioned systems.

Way back when I was a volunteer for McGovern I supervised the transfer of the vote totals from the machines to a piece of paper. One guy would yell out the numbers and someone else would write it. At one machine he made a mistake of 100, coincidentally against McGovern. I was astounded and froze for a second. No one else said a thing. Then I yelled out: “Liar!” and the number was changed. The point is: How do you currently know that after you vote, that it’s counted and passed on correctly?

With electronic voting, you could choose a random number/password which would allow you to check up on your vote after the fact. If you don’t want to do it, fine, you don’t do it. But enough people would - me for instance - to keep the cheaters honest. And of course, anyone could look up the votes anywhere and check up on the counting. They just wouldn’t know who voted for whom.

And besides, if Brazil can do electronic voting without problems, we can do it here.

In the Florida fiasco there were a lot of errors where people voted for more than one candidate in a single contest. There is also the chance of not completely filling in the little circles.

I understand, but having a system that makes it easier for the disabled to vote is meaningless if the system can also be easily subverted. It’s like requiring televisions to support closed-captioning text, but the mechanism has a chance of catching fire and burning down your house in the process.

The system cannot be “easily subverted.” That’s propaganda from people who don’t like the result of the 2004 election. :rolleyes:

Chimpanzee alters audit log on Diebold/GEMS electronic voting system (video)

Ohio 2005 referendum results swing 30%+ from pre-election polls

Nonpartisan watchdog group logs over 600 complaints with e-voting systems in 2004 election.

Diebold insider claims upper management ignored undocumented backdoor in software before 2004 election
…oh, yeah, it’s all propaganda. Just trust the numbers the machine gives us, eh? :rolleyes:

People have been expressing concerns about e-voting since well before the 2004 election.

There are probably anti-corruption laws that ban your getting what is in effect a reciept stating who you voted for, so this wouldn’t be an option.

I’ve been involved in the HAVA implementation here in Minnesota, and a couple of years ago, went to demonstrations given by nearly all the vendors of electronic voting systems.

Minnesota decided to go with the Automark system (already mentioned), but as a supplement to the Optical Sensor machines already used in much of the state. This system does nearly all of what the OP suggested.

But this is a supplement – most people will continue to vote via marking the regular Optical ballot. The Automark system just lets disabled, elderly, or tech-happy people use it to print out a Optical ballot with their marks on it. That ballot is then fed into the same optical scanner/counter as all the other ballots. Ones with errors will be kicked back out, and the voter can try again. (Like DanBlathers examples of voting for more than one candidate.) And the accepted ballots are retained in a locked box if needed for later recounts.

Note a couple of features of this:

  1. votes from disabled people are treated the same as everyone else. There is no separate tally for them, which could be forgotten, and which could endanger the privacy of their vote. And absentee ballots, too. All voters need at home is a pen to mark their ballot, then they mail it back, and it gets counted with all the other ballots.
  2. Recounts are done just by running the ballots thru the scanner/counter again. (Or usually, a carefully tuned one at city hall.) So recounts are very, very fast. Typically take a few hours at most.

Many of the proposed electronic systems produce ‘receipts’ that are NOT machine-countable. They could only be counted manually. And the design makes them harder to count than even an old-fashioned paper ballot from 100 years ago. Recounts would be slow and prone to error in such systems. So they would tend not to be done.

Finally, this combination of low-tech (marks on paper ballots) with high-tech (optical scanner/counter machine) works well under pressure. (‘Degrades gracefully’ in systems terminology.)

I’ve seen it at my own precinct when there was a huge volume of voters. People can vote whereever they can use a pen to mark their ballots. All the voting booths were full, but people sat down at the cafeteria tables around the room, or even right on the floor, and filled out their ballots, then brought them up to the scanner. That worked real fast, so only a small line there. With any of the other electronic systems, you can only process as many voters as you have machines. If you have an unexpectedly large turnout, you will have big lines, and some people will leave without voting.

And if the scanner breaks, people can continue to mark their ballots while the workers send for one of the spare scanners from city hall. Voters just turn in their completed ballots to the election judge to be scanned later. They lose the feature of the machine kicking out a ballot that is marked incorrectly (like voting for 2 candidates), but they can still continue to vote.

We’ve even seen this working when the electricity went out at a polling place. The poll workers got out flashlights to read their voter lists, and even put candles at the polling booths so people could see to mark their ballots. (It was in a church meeting hall, so plenty of candles were available.) Rather dark, but voting went on, despite the power failure! That wouldn’t happen with any of the electronic systems.

Finally, there is the matter of costs. This Optical Scanner system is far cheaper than the others. For each voter, all you need is a paper ballot and a pen to mark it, and someplace flat to write on. We use small voting booths that are basically a small table with privacy flaps on 3 sides. They fold up completely into an attache-case size. And they’re pretty cheap. The major expense is one scanner/counter machine needed per polling place (plus a few spares). You can even have more than 1 precinct voting at the same location, because the machines can automatically keep separate counts, even if each precinct has a different list of candidates. All the electronic systems need one machine for each voter, greatly increasing the cost, both in purchasing, maintaining, and transporting these machines.

Having seen pretty much all the other systems, I think the Optical Scan Ballot works best of all in practice.