Slashdot Mirror
Touch Screen Voting Trouble in Florida
usn2fsu03 writes "Here we go again with
another election controversy in South Florida. Touch screen voting was used in a State House election that was won by twelve votes. Unfortunately, there were 134 people who went through the process of checking in to vote, but either did not vote or cast a vote that was not counted. Without a paper trail it is anyone's guess as to what those voters' intentions were. Obviously, there is work to be done in the Election Supervisor's office before November comes around."
And we all know what happens when electronic voting goes bad.
I mean seriously, what will it take for these people to realize some things are just better done the old way, one of them being voting.
I can see it now, in the future major media conglomerates will consolidate and choose the president based on which is the most popular in *their* opinion. I guess that could be called a 'representative democracy' too
Representation of corporations *shudder*
I think each slashbot should think carefully about this and write to his congressman.
-Voter walks into booth
-Voter touches appropriate button on screen
-Voting machine records the vote electronically and also prints the vote on paper (maybe in like a scantron type format so it can be easily recounted)
Done?
If you don't press vote, you didn't vote. You have to live with this. The instructions are available, so if you don't complete the transaction, you really can't complain. (and I'm sure your local poll worker will help if you have trouble reading the instructions.)
There will never be a technology that will prevent idiots from making mistakes. Now we have a system that is more expensive, no better than hanging chads at determining voter intention, and HAS NO AUDIT trail. We are worse off then before. Yet another example that technology will never turn an imperfect collection of humans into a utopia.
In the UK, the loser would have the right to go to court and ask for (and probably get) a new election. It happend in Winchester in 1997.
If they couldn't punch a voting card they also shouldn't be voting. My 4 year old could correctly punch a ballot.
Right. Uh-huh. We never saw *this* coming. No sireeee. Electronic voting is *reliable*, *safe*, and *fun for the whole family*, and anything else is against the word of the Fuehre...er, I mean, is Anti-American.
--
I Hit the Karma Cap, and All I Got Was This Lousy
I'm sorry, but since when was any vote-counting system designed to interpret what a voter's intent was, beyond correctly-cast votes?
If people don't/can't vote correctly using even the simplest methods, then perhaps even they did not know what their intent was.
"Ask not what your country can do for you." --John F. Kennedy
I know it's scary that we can tell there's something wrong but there's no way to know the right result -
But the worse scenario is one where there's no way to tell anything's wrong. No reason to request a manual count, no reason for trusting fools to question the results.
Most people, it seems, have an "I haven't verified this system, therefore it must be secure" mindset. But don't worry; this particular problem will be fixed and people can go back to assuming everything works until the next time something is obvious wrong.
Remember - it can't be a problem if nobody knows about it.
Democracy in inaction...
Still, USA is not a democracy. Its a republic. People seem to forget that...
All the groups calling for voting reform can point there and say "Electronic voting without proper auditing tools is worse than hanging chads."
The Canadians will just keep laughing, as more people ask why their pencil and paper system works more smoothly, and in many cases faster, than ours.
I don't care if we have a fancy electronic system with proper audit trails, or if we go to a pencil & paper system with proper audit trails. I just care that we get there quickly.
frob
//TODO: Think of witty sig statement
I kind of get this kooky conspiracy theory feeling where say every 3 votes for the "wrong" candidate is excluded and it's a part of the closed program code. You kind of get that feeling when you see stuff like this: Bogdanoff had a ready explanation for the mystery. She theorized that some of the people who cast nonvotes were among the county's true-blue Democrats who were appalled to find a ballot with only Republicans. Did this really happen?
I'm otherwise (still) surprised that paper receipts were never given in the beginning, but it's a very good idea for the future. If anything, it should be a requirement.
People say I'm crazy, I got diamonds on the soles of my shoes...
Can someone please explain to me when this became a land where we had to determine what a voter intended and not what he actualy voted for (or in this case didn't vote for). Ballots are fairly simple things, and most of us learned about them in 4th grade. If you are unable to comprehend how to work a ballot, by law, polling places are supposed to have someone there to explain and assist you. If you don't take advantage of it, that was your choice. Vote right, or don't vote at all, but don't be bitching when your incorrect ballot isn't counted.
T Money
World Domination with a plastic spoon since 1984
You're missing the point.
It's not whether those individuals voted or not.. it's that there's no way to go back and check whether they did or not. There's no way for people doing a recount to go and look for the equivalent of "hanging chads" and such.
The article even addresses that, it's fine if someone doesn't want to vote. It is NOT fine that there is no way to go back and identify the voter's intent.
Without any audit it is impossible to tell if the problem was their stupidity. I have no problem expecting people to be smart enough to do this, but for all we know those votes could have been lost through any number of technical errors, there's absolutely no means to check this with no audit trail and secretive software practices. The only available audit, tallying people showing up vs casting votes shows a significant discrepancy. That is cause for concern, and indicates the need for a better audit trail. Something that is simply being ignored and denied at every request.
If you acheive the first goal, but fail to address the second, you create an increasingly angry and restless population, and that's unhealthy for any democracy. A lesson many politicians seem to have taken from the Florida debacle is that most people will "get over it", and go back to driving their SUVs and watching TV. So far they've been right about this. Unfortunately, that only works if we're talking about an isolated incident; if people begin to develop even the impression that they're being repeatedly screwed, our society will suffer.
Because hopefully it will bring attention to how important a voter audited paper trail is. Hopefully this will gain widespread attention, so that before a more important election (say a national congress seat or presidential election) the people who administer elections will get it right.
This has been said before on /. and elsewhere but is worth repeating:
Paper receipts that stay at the polling place = good. Allows parellel count of small sample to check machine accuracy; allows recount in the event of a problem.
Paper receipts that go home with you = bad. Potential for intimidation and vote buying.
Hows the recount going to be fair if they can't recount the individual votes? About all they can do is tabulate the total from each voting machine again.
As many people have already stated, this is exactly an audit trail is necessary with electronic voting.
...and my three-year-old can mark an X in a box!
I do not understand why you Americans go for these Rube Goldberg methods of casting a vote come election time. I can understand the need to be able to count the ballots quickly, so go for cumputerized voting if you must, but why not use the KISS approach for what should be a required paper trail? Seems to work just fine in the rest of the voting world, and there's no silly assed questions concerning "hanging chads".
Make the voting booths a bit more substantial, like the "man-traps" that are in some banks now.
Voter enters the booth, booth closes and locks. The booth will not re-open until the person has voted properly or if they page a pollworker to let them out. If the latter occurs, the pollworker can give them additional instructions or let them out and note the incident for any subsequent legal challenges to the election.
Of course, in all fairness a "none of the above" entry should be made for any one-party election.
I vote in all local and national elections and my local incumbent "representative" is not of my political party. My party (or any other party for that matter) does not even have a candidate on the ballot! In those cases, I leave the entry blank if I cannot vote "NO" to abstain. Since in the Florida election all the candidate choices were Republicans, I would think that some voters seeing their party was not represented at all on the ballot would abstain in a similar fashion.
So there's nothing to see here.
I tried every decent and legal way I could think of to resolve the issue w/the business before I rented the chicken suit
If they can't figure out to push the VOTE button to count their selection, maybe they shouldn't be voting anyway...
True, but out of all the voting systems, computer systems could be more idiot-proof than any of them. I quickly thought of several simple ways for the system to prevent a luser (I mean voter) from leaving the booth before they actually voted. This same non-voting problem may have happened with the chad-machines. And even pen and paper isn't immune from UI problems.
Adidas To Bring Back Sneakernet
It's not keeping more of your own income; it's continuing to accept the services you formerly paid for with taxes (in fact taking more services), but now paying for them with a cash advance from a multitrillion dollar credit card. You're still going to pay it all back one day with money from your income, but with interest.
I disagree
Voting should be so easy and so simple to do that it is hard to screw up.
A key part of a fair election is that if someone makes the effort to cast a vote, the system should record that vote.
Making it unnecessarily difficult risks making it an unfair election.
...by attaching a *printer* to the voting machine.
So, how is this better than a paper ballot with a stub you detach as proof of voting?
It gives the machine makers millions that should have gone to public schools.
Hooray for demcracy.
In Soviet Russia, articles before post read *you*!
"No.. no.. you don't want to pick *him* he's the wrong candidate." ;-)
There is no reason to believe that votes were lost through any technical error. The machines simply claim that about 1.5% of voters chose no candidate. This is a valid option, and MUST be allowed.
It is a little surprising that so many people chose no candidate, given that this was a special election with only one question on the ballot. (Why bother voting if you're not going to vote for anyone?) But on the other hand, voter stupidity and the fact that the only candidates were Republicans would tend to increase the number of nonvoters.
I think that we need to get a verifiable hardcopy system where the voter gets a chance to verify the hardcopy token before it disappears into a secure receptacle. This will not only make recounts better, and it will also help technology-challenged voters to vote the way they intend.
The one issue here is that the token cannot display more than just a small amount of information. Otherwise very few voters will bother to check it for accuracy. And in a regular election, the level of State House Representative is surely below the cut that should be made. Otherwise (in most states) you will have at least six different names that the voter has to check: President, US Senator, US Representative, Governor, State Senator, State Representative. If you think that more than a few percent of voters are going to pay attention to a long list of names after they are already done voting, you are crazy.
Some, if not all of these 134 people are what are referred to as "fled voters". I work on my local election board. I'm one of those knuckleheads who spends 13 hours twice a year explaining to the exact same people how to use the exact same voting machine that they've used for 10 years. Every three or four elections we end up with more people signed in then actually voted. It's pretty simple actually... someone comes in, waits in line for a bit, then decides that it's too much of a pain in the ass to wait any more and simply leaves. Since I started on my local board I've done maybe 10 or 12 elections and we've had at least 4 or 5 fled voters. Multiply our average of a third of a fled voter per election over the several hundred polling places in this election (I have no idea of the actual number) and 134 fled voters is not out of the realm of possibility.
What would REALLY worry me is if there were more votes than voters signed in. Now THAT is definitely voter fraud.
NOTA gives voters the opportunity to actively state that they don't like any of the candidates. With a binding NOTA, if the majority of votes go to NOTA, no one is elected and the process begins again. In a non-binding NOTA, the populace get to express their opinion, but the candidate with the most votes still wins.
Nevada has had non-binding NOTA on the books since 1976. This past summer, Massachusetts passed the first binding NOTA. It goes into effect in 2005.
Jeez. Maybe we should stop trying to sell "Democracy" to other nations until WE can get it right. This is getting just a bit absurd...
FUCK BLAIR!!! and I'm not talking about the fat girl (which one?) from 'Facts of Life'...
That's not entirely true - otherwise we wouldn't have any use for ECC or parity. Computers can make "mistakes" in as much as data can be corrupted by physical processes that having nothing to do with the intended or programmed operation.
Technicalities aside, none of the election problems are about counting accuracy, neither human, nor mechanical, nor electronic. That's not the point. All measurements have an associated accuracy. It's how we deal with it that counts. If the margin of the election is of a size that given the error rate of the system there's a "reasonable" probability that the outcome is in error (1 sigma, 13% probability of error, say, given the error rate of the technology used) then a run-off election should be automatic, even if there's only two candidates in both elections. No matter what the voting technology. A 5% threashold would be statistically supportable.
All sampling systems have a margin of error. It's a 9th grade science mistake to get an F for submitting a graph of plant growth or whatever without any error bars. We seem to suffer from cognitive dissonance in refusing to admit there's an inescapable margin of error, and thereby not accommodating for it.
In 2000, FL and several other states should have held run-off elections between W and G after the first election found them at a "statistical tie". It's not clear which way it would have gone after that, but whoever thereby won would actually have been a democratically elected president, rather than one technically appointed by a divisive judicial coup.
Anyway, the critical failure regarding DREs is the lack of recognition that they are fallible. How do we deal with critical systems that might fail? We create an audit trail so if something goes wrong, we have a chance of undoing the error, or at least figuring out what failed and fixing it, and at the very least knowing that something did in fact go wrong so we can try again.
The systems shipped by Diebold and ESS etc are both intrinsically fallible and intrinsically inauditable, which is intolerable. Further, if a voter has reason to doubt the impartiality of a company that has, for example, pledged to deliver it's electoral votes to the republican in the next election to be run on it's own vote counting equipment, they might have some reason to doubt the veracity of the black-box tallying process and that undermines the authority of democracy. It is important, therefore, even if it were proven technically unnecessary, to provide voters with the familiar indicator of fairness provided by a human-readable, authoritative, tangible ballot.
We've gone through a lot of effort convincing ourselves, and by force much of the world, that having a brainwashed electorate choose one or the other corporate flack as titular head of the country is the best and fairest form of government on the planet (and it may well be, alas); at the very least we can apply basic 9th grade science to finding out whether tweedle dee or tweedle dum won the popularity contest.