This morning I heard on the radio of the problems with these electronic voting systems.
As an embedded systems developer, I find the lack standards alarming.
It seems to me that adding a printer or card punch unit to each unit would put all this to rest.
In collage I worked on IBM card punch machines. After college I worked for a company that repaired these same card punch machines. Kind of fun at the time.
Does anyone here know the real deal about these machines.
Why don't the powers in charge just put printers or card punch units on these things and be done with it.
hamilton
PS: Would you vote on a machine that uses a microsoft operating system ? Would you vote on a machine that uses a open source operating system ?
I must say find the very idea of conducting votes through anything more complicated than pencils and paper ballots dangerous. IMHO the very concept of a "voting machine" is a menace to democracy.
To put it bluntly: if there's *any* part of the entire voting process that the average voter can't thorougly check for himself, what you're conducting does not deserve being called an election. You have to be able to *prove*, not just in principle but in actuality, to every single voter who cares to ask, that yes, his or her vote was correctly counted. I see no way whatsoever how you might manage to do that with any kind of voting machine currently being planned or in used.
For heaven's sake, please, no card punchers. They'ld use cards with pre-punched holes again. "Hanging chads" and everything included. The makers of those machines used last time have proven beyond reasonable doubt that they don't know what the heck they're doing as far as punching holes into cards is concerned.
That business in Florida last time round was utterly ridiculous, and it may have damaged the reputation of democracy for a long time to come. You don't want anything even remotely like that to happen again, *ever*.
If those two were the *only* options to choose from, the open-source one, any time. But that's asking the wrong question. You should use neither.
--
Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.
Getting robust, secure, trustworthy results isn't easy, but many people, myself included, believe that a physical voter-verifiable ballot is the best way to go. Human readable paper and punch cards fit this well, but they need to be augmented with a good system for use, handling of exceptional cases, etc.
It appears that the brouhaha over the Florida recount in the 1999 US presidential election prompted a rush to "do something." Many jurisdictions thought that using the computer was the solution, without thinking through the issues of reliability and susceptibility to fraud. The companies, foremost Diebold, eagerly sold their touch-screen, automatic tally machines. A few offered results on a paper printout as well.
Part of the problem is that the commercial voting machine manufacturers have, to date, said "trust us, the machine is secure and reliable". Many selection committees accepted that. Some recent election results have shown problems with the machines. An analysis of Diebold code, unintentionally accessible on the web, showed several security flaws. The deficiencies have probably been fixed, but how do you know, and how do you know that others weren't installed without open source verification? The fundamental problem for me, however, is lack of sufficient tracibility if physical ballots aren't used.
There were a few people that have organized a campaign to demand voter verifiable ballots, basically meaning something human-readable on paper.
In Boulder County we have used punched cards for a couple of decades. The cards are not pre-punched. The punch machine has a lever that unmistakably points to the selection printed on the card. When the handle is depressed it sinks the die through the card, backed by a punch cutout. The punch was very clean and I saw no hanging chad. The machines got worn, though, through a couple of decades or so or regular use, so that the punch arm was not held at a perfect right angle to the slide rail. The result was that you sometimes needed to wiggle the punch to line it up with the corresponding hole underneath, otherwise you couldn't punch. Once line up, though, there was an unmistakable crunch as the punch traveled through, and the hole in the card next to the printed selection was obvious. The card was machine- and human- readable. Frankly, I thought the system was very good.
Boulder county is replacing their voting system. I don't know what they will end up with. They had a show-and-tell for voting machines a while back. Out of maybe ten machines, one or maybe two had a printed record. One machine I actually crashed when testing it. Right now, I think Boulder County is encouraging/requiring vote by mail, which has its own potential problems. I prefer looking the poll workers (often my neighbors) in the eye and affirming that I am who I say I am.
I would be happy with a straight non-computerized well-designed punched card system (as we had in the past) or a paper ballot. One of the reasons (excuses?) for using computers is to meet the need of handicapped voters. I don't know how the computer is more accommodating, other than perhaps allowing selection of a larger font, but that is part of the reasoning for selection of a computer. How has handicapped voting been handled up to now?
Lately there has been a lot of recent support for verifiable ballots, which I think is good. Unfortunately, many jurisdictions have apparently made commitments to non-verified systems. The question has become political, of course, with factions lined up on both sides.
More are starting to do that. There was an interesting editorial cartoon recently in which the vendor was explaining the use of the voting machine sporting all kinds of bells and whistles which basically said "and then it prints a ballot which you mark and place in this box (standard ballot box with a slot in the top)." The joke, of course, is that the fancy voting machine wasn't needed at all.
I could agree with that if not for the following aspects of it:
1) the governor of Florida being related to one of the contestants
2) said governor not having kept his hands out of the process
3) the supreme court getting involved in it, and the way that involvement took place
4) the eventual winner going on lecturing other sovereign nations about democracy
All in all, flipping a coin to determine the result of that election would have made more sense than what actually happened.
--
Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.
There has been tons of traffic on this topic in comp.risks over the past couple of years. I suggest you search for the items in that newsgroup/mailing list/website (it is all three).
Personnaly, I favour aimple pencil and paper approach to casting my vote. It should be possible to machine count the votes easily enough once the voting is completed and return the result in a reasonable amount of time. Heck, in the UK we still hand count and can usually have the results available about 4 to 12 hours later (soon enough methinks).
--
********************************************************************
Paul E. Bennett ....................
Forth based HIDECS Consultancy .....
Mob: +44 (0)7811-639972 .........NOW AVAILABLE:- HIDECS COURSE......
Tel: +44 (0)1235-811095 .... see http://www.feabhas.com for details.
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************
To tell the truth, I tried to create chad on a California voting card. It was very difficult. Anytime I did have hanging chad, it would fall off when put in a box with other papers and shook. I was unable to create the sorts of problems that were implied to be rampant. It was impossible to create a situation that appeared ambiguous in a manual recount. My conclusion was that Florida had lousy punchcards.
The real reason there is no paper trail with voting machines is that it is expensive. Ie, many voting districts have been ordered to use electronic voting machines, but given minimal money to accomplish this. Even in the California counties where the manufacturer had deliberately deceived the elections districts and were required to get new machines, they still still complained that they should be able to use the disqualified machines rather than waste time and money.
The machine defenders are a bit disingenuous as well. When asked why there was no paper trail, one defender answered "you can just print out all the votes at the end of the day and you'll have a perfectly good paper trail then." I wonder if she keeps all her ATM receipts or just trusts what the banks sends her at the end of the month.
Probably true. Even ignoring open source, if presented with a choice between a machine that allowed a select group of election commisioners to review the source code under a non disclosure agreement, and a machine where it was impossible to look at the code under any condition so that the maker had to be trusted implicitly, the choice should be easy.
--
Darin Johnson
I'm not a well adjusted person, but I play one on the net.
Easily solved. Have one touch screen machine at each polling place, and ask each voter if they would prefer to use that machine instead of the punch cards. But it only helps some handicapped people. What if I were blind so that the touch screen was useless? Should we require every single voter to say outloud to a poll worker what their choices are?
Which suggests the perfect voting machine. Use optical scan ballots, which have worked well for a long time, and which allow both electronic counting and hand recounts. But tack on a touch screen system. The voter makes their selection on the machine, then the machine prints out the optical scan ballot. The voter can then review that ballot if necessary. The ballots would be uniformly printed (not worrying about people not using number 2 pencils), the handicapped would be served by the touch screen, the voting machine manufacturing industry would still make money, etc.
--
Darin Johnson
Laziness is the father of invention
I don't remember exactly how he was involved. Was he? As governor, what job was he trying to do?
The US was in a very dangerous situation, thanks to the unbelievable incompetence of some in Florida. It could have been somewhere else, but it was Florida.
The Supreme Court did not want to get involved, as I recall, but they kind of had to. Blame those 12 if you do not like the outcome.
Gore and the formerly respectable Democraps would have counted and counted and counted and counted until they got the result they wanted.
Oh, come on.
No, Gore might have won, and the Democraps have lately not indicated a willingness to kill terrorists until there are none left.
I am sure the majority of election commisioners groups contain one or more such with the ability to evaluate source code in an arbitrary language running on arbitrary hardware with arbitrary mechanical and electrical failures.
Bear in mind that the rewards for beating the system are immense, and can include: money, power, women, life and death for many. The "winner" takes whatever s/he desires.
The primary requirement is the untouchable trail coupled with anonymity. Paper ballots and punched cards both can supply this. Maybe some other system can, but I have doubts. Without the trail a complete election repeat is required since recounts are impossible.
Electronics etc. can supply auxiliary systems, facilitating the counting and reporting process. They cannot go further.
--
fix (vb.): 1. to paper over, obscure, hide from public view; 2.
to work around, in a way that produces unintended consequences
that are worse than the original problem. Usage: "Windows ME
fixes many of the shortcomings of Windows 98 SE". - Hutchison
hamilton wrote: : This morning I heard on the radio of the problems with these electronic : voting systems. : : Would you vote on a machine that uses a open source operating system ?
formatting link
-- ****************************************************** Never ever underestimate the power of human stupidity. -Robert Anson Heinlein
The poor quality of that page (it fails to adjust itself to the width of the display) leads to grave doubts as to the quality of the proposals.
--
fix (vb.): 1. to paper over, obscure, hide from public view; 2.
to work around, in a way that produces unintended consequences
that are worse than the original problem. Usage: "Windows ME
fixes many of the shortcomings of Windows 98 SE". - Hutchison
AFAICT, he had no impact whatsoever on the result.
The unfortunate thing is that the SC was forced to get involved. Their decision made perfect sense to me, and I expect to most people. They simply affirmed that the Democrats could not pick and choose which districts to recount, but had to recount all of them if they recounted any.
They were immediately accused of giving Bush the election, but several subsequent (complete) recounts showed that Bush had indeed won the state.
That was apparently the plan.
--
Al Balmer
Balmer Consulting
removebalmerconsultingthis@att.net
The point about the time taken to do the count manually (or semi-automated) is quite on target. There's no pressing need for instant results and the cost of labor is insignificant since many of the people (in the U.S.) are volunteers. The election results will not be avail- able as quickly using manual methods, but the only people impacted by that are the "instant news" group from the local TV stations -- you can read about it in tomorrow's newspaper just as well (or better).
What has puzzled me about the supposedly more secure machine voting where a paper form is produced for the voter as a verification of his/her voting is how does the paper form improve things? If something is suspected or does go wrong, does everyone have to bring in their form for a recount? How do I verify that the ones and zeroes in the machine are counted in the results in the way I intend?
Sounds like the system used by Kern County, California. It worked quite well but the State mandated switching to "better" technology. Kern County did and now the State is mandating that the "better" technology not be used in the coming elections.
If anything, instant results and the reporting thereof can harm voting. If voters in Western states see how an election is shaping up, they may not bother to go to the polls. Of course, if the election is not going their way, they may be more motivated to vote. If the results weren't instant, voters might be motivated to vote as they don't know how the election is turning out and might not want to take the chance of losing.
I just heard on the radio that the State of Michigan has purchased optical scan voting machines for those districts that don't already have them. I'm not certain exactly what systems those are, but in my township, we have paper ballots with broken arrows pointing at the candidates. To cast your vote, you connect the arrow. Something like
The ballot is read by an optical scanner that stores the scanned ballots in a locked box. Invalid ballots are rejected, and the voter gets a fresh ballot to try again.
It works well. I'm not sure what they do for visually impaired voters...
The systems I have heard about display the resulting paper from behind a clear window. The voter then gets to press a button saying, "Yup, that's what I wanted" or "Wrong-o bison breath, let's try that again." Accepted ballots are then stored in a locked box. Rejected ballots are disposed of (a separate locked box, I assume, or perhaps just marked invalid by the system some way).
The Constitution has no requirement that election results be available instantly. If it takes a few days, or even a few weeks, for the results of an election to be verified and certified, well, the country can wait.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.