Sunday’s short NY Times editorial, titled “Déjà Vu in Florida,” concerning the mess in Sarasota’s FL-13 U.S. House race between Christine Jennings (D) and Vern Buchanan (R) (369 votes difference, 18,000 missing votes on the paperless ES&S touch-screen systems in Jennings’s strongest county) gets much right. But on one huge point, they get it completely wrong.
First…Here’s the highlight graf amongst the stuff they got right:
Agreed. Wholeheartedly. ‘Nuff said.
But here’s the lowlight graf, in which they get it all wrong at the conclusion of the editorial:
Wrong.
A so-called “voter-verified paper trail” on Sarasota’s touch-screen systems would not have solved the problem in Florida. I’ll have more on this issue in the coming days, but suffice it to say for now: Paper trails, such as they are used with DRE/Touch-Screen systems, do not work. Voters don’t verify them, elections officials don’t count them, they are not accurate, they can be gamed, they jam the printers which leads to voters being turned away without being able to vote…among just a few of the reasons.
VotersUnite.org has now come out with a complete statement on why even they are no longer willing to support voter verified paper audit trails (VVPAT) and thus, they will not support Rep. Rush Holt’s otherwise well-intentioned HR550 legislation as long as it allows for VVPAT. Neither will we. More on all of that soon as well.
In the meantime, I’ve said it before, and I’ll keep saying it…louder and louder each time if necessary: America needs a paper BALLOT for every vote cast.
Period.
I hope you’ll join me in that call. I’ll try to make it easier for you to do so in the coming days.
If there’s one thing the events of Election 2006 have made crystal clear, it’s that the VVPAT train wreck ahead needs to be derailed immediately. The New York Times (no suprise) is clueless there. But, at this point, there is no longer any excuse for any self-respecting Election Integrity organization to keep their heads in the sand on this one.
BREAKING UPDATE: Please note, since posting this story earlier this morning, a blockbuster has occured. The National Institute of Science and Technology (NIST), the federal body responsible for creating the federal Voting System Standards (such as we have any), has recommended that all touch-screen DRE voting systems should be decertified! Further, apropos of my story above, they’ve said that so-called VVPATs do not work and should not be used with touch-screen systems! Complete breaking story posted here…
Amen, brother Brad!
Aw Brad, you can’t have a paper ballot. If you did that, the electronic voting machine would be nothing but a buggy, cumbersome, hackable $3000 pencil–and then the whole industry would be destroyed!
Lol. Good one #2.
I just wanted to ad that this call for a voter verified paper trail shows how ignorant the press remains on this issue. It can only be deliberate.
We all know and have seen the documentary that showed the vote counting done within the machine (the tabulator) can be rigged and yet it can’t be traced. Just because you vote for candidate A, and your paper receipt says candidate A, there is no guarantee that the machine actually counted your vote for the proper candidate.
This is why we need not just a paper trail, but the paper trail should be the standard OR the software should be fully auditable and open for examination. There is no other reasonable course of action.
I guess I have a different opinion on this. If an evoting machine outputs a paper ballot with the voter’s selections on it and the voter can review the paper ballot before turning it in, then I don’t have a problem with it. I see something like this as beneficial to people, elderly or handicapped, who physically have trouble filling in the circles on a ballot with a pen. In essence, the touch screen system becomes a glorified ballot printer.
On the other hand, if we’re talking about an evoting machine outputting a paper receipt that is not the actual ballot being counted, then I have a problem with it.
Your thoughts?
– Tom
Tom – Yes, that’s what we’re talking about. On the DRE’s and touch-screens that produce a “VVPAT” the dirty little secret is that those trails aren’t actually counted. The machine count is what is used, so all the voter has left is a false sense of security that that worthless little paper trail actually means anything. It decidely doesn’t.
Yes, Amen brother Brad.
I am so sick of hearing about these Voter Verified Paper Trails. It’s like they’re expecting that if anyone’s unhappy with the election results or questions the outcome, all they have to do is ask all the voters in the entire county to bring their receipts down to the county courthouse so we can have a recount.
What a bunch of baloney.
Give me a PAPER BALLOT, please.
Great report, but one question: even if everyone receives a paper ballot, whether through punch cards or DREs, aren’t those ballots counted by an optical scanner anyway? Unless these ballots are hand-counted, what’s to stop the final electronic total from being manipulated?
A big thank you to everyone at bradblog for keeping this issue in the public eye.
Thanks for the clarification Brad. Personally, I think of the paper ballots AS the paper trail (i.e. for recounts), so this is where this topic can get confusing.
Maybe a distinction needs to be made that the idea of a paper trail is good, but voter-verified is a superficial or worthless type of paper trail, since this trail is extraneous or irrelevant to the vote counting process. However, paper ballots provide an excellent paper trail that is integral to the vote counting process.
Or am I getting too pedantic about this?
– Tom
Tom –
You are not getting too pedantic at all. There are multiple issues here and one reason for opposing the VVPC is because it’s not likely that the democratic process could craft legislation that would address the issues sufficiently. Democracy does not handle technical complexity well – it would not be inaccurate to say that democracy doesn’t even rise to the level of handling it poorly. Anyway, some of the problems are:
1.The “paper trail” has nor been sufficiently defined, even in Holt’s bill. Is it a receipt? Is it a printout of a ballot that the voter verifies and drops in a ballot box? What does it look like?
2.Is there any requirement that the paper trail be counted? during a recount only? There isn’t, and that means that the fraud only has to rise to the level sufficient to ensure that the paper ballots won’t get counted. Obviously a paper trail DOES reduce the opportunity for fraud, but not enough to overcome the documented problems with the current systems.
3.No one is requiring that the source code be public. Without that, anything computer related should be suspect. Period. I work in Information Technology for a bank. Now, my company does use operating systems like windows for which the cource code is not available. But those are so ubiquitous that the risk of malicious code by the operating system designer is essentially non-existent. Aside from that, the review of ANY code that we use is exhaustive. I can’t imagine the ATM machine manufacturer telling a major bank to “trust them”.
4.Even if there was a uniform, decent paper ballot that printed out from the computer, could be reviewed, was kept in a ballot box and was always counted, this would create an extra step in the process. It’s NOT the same as a voter filling out the ballot and looking it over. No one has yet convinced me that this will do more good than harm.
Now, all that said, I differ slightly from Brad in my opinion about solutions. There are two main problems with paper ballots. One is ballot design – I’m always amazed at how badly this can be screwed up. This problem is not limited to paper, as FL-13 shows us, but it is one reason why Bush is president today.
The second problem is difficulty determining voter intent. Putting marks in the wrong place, etc, etc. This can usually be mostly overcome with manual counting like happens in recounts, but I’m not convinced we should accept that this is the only way.
It seems to me that it would be possible to create a combination of computer, mechanical device and paper ballot that would involve the person filling out the ballot with some level of mechanical/computer assistance. For instance, a ballot that had little “SAT” type circles to fill in that used some device to absolutely prevent you from drawing outside the oval or only partially filling in the oval. I don’t claim to have all the answers, but there are many many technically innovative people who have come up with solutions for far more difficult problems than the two problems with paper ballots that I mentioned.
The newspapers didn’t tell anybody here they could do it, but I guess the officials knew some people were on to the problems with the machines and might cause a scene at the polling station, so, at least some of the precincts let you vote on a paper ballot.
It personally felt great to me to be actually filling in a circle beside the actual name of the person I was voting for, for the first time in my life. It was easy, and gave me confidence in my vote. We’ve just got to bite the bullet and SCRAP these expensive, worthless machines that are designed to let each jurisdiction decide how it’s going to steal votes. Period!
The bottom line is simple. Everybody must know their vote will be recorded on a good, solid paper ballot and publicly counted with full transparency.
Those saying more is needed than a paper trail are correct. For over two years I’ve been designing a solution that would be as safe as a banking transaction. (my background is two decades of software engineering and internet systems design). I’ve created a web site to describe it at: VoteAndVerify.com It uses “online verification with bank-deposit reliability.” Here are some of the features of this system, but I suggest you watch the Flash demo, see the comparison table with other paper-trail systems, and go through the FAQ’s.
1) Paper receipt
2) Paper receipt the voter can keep
3) Safe so that no votes can be sold
4) Full recount of votes available
5) Efficient use of small amount of paper
6) Single voter can prove error or tampering with votes
7) Voter can verify that votes got recorded correctly in final official data base
8) Single voter could trigger an investigation and perhaps a recount
9) May show a recount is needed even in races that are not close
10) Can isolate the exact voting machine creating errors
11) Can lead to identifying person tampering with votes
12) Can allow use of intermediate electronic machines because the voter him/herself can verify the final results online. (i.e. you don’t care what machines a bank uses because you can verify the final result– and it better be right because you have a receipt that can prove them wrong. VoteAndVerify.com does that!)
I hope I’ve finally communicated all this in the web site clearly enough that those who can do something will take notice and properly evaluate this approach.
VoteAndVerify.com