I would say never.

Please show me a source where the machine did not record a vote it was supposed to.

It hasn't happened.

There have been undervotes, but that has nothing to do with the machines. The machines prompt TWICE before recording a non-vote. They won't allow you to not vote for something until you've verified that is what you want to do. Yet, this gets blamed on the machine.

I stand by "never". Prove me wrong.

Since you pretty much ignored the rest of my post, I'll assume you agree and move on. Now on to this assumption, so it is only democrats that are suggesting electronic voting isn't ready for primetime eh?

Well, you might be partially correct, President Bush has surely been trying to sell electronic voting as has Jeb Bush in Florida. This article paints a little different picture.




It appears the Republicans don't seem to have supreme confidence in the system either - I doubt they'd use the word "never" when there is so much at stake for them.


So, which is it guys? Do you support the machines? Or do you think there are problems with the machines? It seems that the Republicans can't even agree on this amongst themselves.

So, DR, I think it is safe to say that the Democrats aren't the only ones that don't trust the machines. I can say that the Democrats are the only ones that are consistent.

First, you simply cannot determine intent looking at hanging chads after the fact. You can make assumptions, certainly--but that is not 'determining intent', it is hazarding a guess. Basically, same line of reasoning as being careful of using the word 'never'. So, since you can't absolutely determine intent, you should not use that determination in deciding the outcome of a vote. That aside, what is it about electonic voting that precludes doing a recount? Computers are electronic--I'm going to make an assumption here that everyone here has, at some point in their life, printed something? If printed ballots are required (given the many other means possible of backing the data up, they don't need to be), then, print them! Nothing in an electronic system prevents this. To me, that is not the issue to be discussed at all--an implementation detail, perhaps, but not a conceptual one. Let's go back to the 'hanging chad' example you cite. How can that be prevented in a mechanical system? It can't--its a result of the mechanical process. How can that be improved in an electronic process? Simple--the voting device can display you choices you have made to you before they are finalized. The machine can't--once a hole is pressed, it is pressed. This is the main advantage of the electronic system, and I fail to see arguments (besides cost of implementation) against it. As for cost, allow me to ask a simple question: Would it have been worth it to have a more definite means of vote tallying in Florida during the last election? As long as costs are reasonable, even I as a conservative would favor federal spending to upgrade and standardize our voting system. It is simply far too important an issue to scrimp dollars on.

I am not advocating that, I was just citing that this is a requirement in Florida Election Law, I didn't write the law nor do I really agree with it.


It is not impossible, but that is one of the arguments being waged by the "Democrats" and argued against by Voting Machine companies. These companies are under the impression that it would be too expensive to print an auit trail of some kind.

I believe that some of the newer designs are incorporating this feature and I also know that congress is pushing for legislation making this a requirement. But currently, the majority of the machines do not have this feature. Cases like this Florida mistake, provie the case for why that is necessary.

I fully support e-voting, but if you don't make the changes to curb human error, then what's the point in bringing in another system where an entire precinct could be eliminated because of a stupid mistake; this compared to 12% of votes not counted, which would be measured over an entire state. A newer system should at least have a reliable backup and built-in fail-safes that will prevent human error. This is the problem e-voting, they don't have this backup, or the backup used is electronic too. Human error as is not the failure of the person using the device, it's the failure of the company creating the machine.

Even places that keep all their files in an electronic database at least have a hard copy of the files. Why? Just in case. Likewise, voting machines should have this. The problem is the company doesn't want to spend money on this, but that's the problem when you privatize voting, I think this should be a government headed project.

Again, I don't think this is a partisan issue, so don't bring up Republicans or Democrats stealing an election. I think conflict of interest is another discussion. What we're talking about here is whether or not we should fully embrace a system where a person can make a stupid mistake and wipe out an entire set of voting records.

You have to be careful when you use statistics. 12% loss of votes across a state is different if it is a count of votes in each precinct as opposed to 12% of those votes all lost in one precinct.

Agree with the first, disagree with the second. Absolutely there needs to be a backup system. Does this need to be a paper system? Not necessarily. I work as an IT Solution Architect--I have set up many different computer systems--every single one of them had backup plans--none of them were paper. It's a matter of defining what needs to be recoverable, and setting up proper procedures and equipment to make it recoverable. You can have a multi-tiered environment (machines, servers at each voting place, servers in each state, etc). Each tier can be set up with its own copy of data, and that data can be set up to copied to other places, if need be. Basically, the system can be designed to be as fault tolerant as you want it to be, with cost usually being the limiting factor. As for accidental deletions, this too can easily be minimized with proper design. First, you can make it very difficult for anyone to delete anything--very few people need the keys to the castle. Just as most data entry people in a large company simply cannot delete the entire accounting database. Combine this with replicated copies being sent simultaneously to other locations (probably up the chain)--and this basically becomes a non-issue, even without printing anything on paper. As for how to run this--don't have the government take it over, but they can certainly define the requirements. In fact, if the government is going to pay for it, they NEED to be the ones defining the requirements. This could then be sent out to bid, just like all the other government projects.

It is important to keep in mind that the flaws/advantages of any new system are only important relative to the existing system. For example, while it is possible that data can be lost in an electronic system despite any safeguards built-in to the system--is this not true also for the existing system? What happens if the box containing all the paper ballots gets destroyed by, say, a fire, flood, or accident? What is then relevant is whether the new system is more or less likely to have this happen, and more or less able to recover in the event something does happen. Properly designed, I think electronic systems, with their inherent ability to be redundant, have a much higher incidence of data protection capability than mechanical systems, and have a better user environment also. So, it really just boils down to an issue of cost. How much are we willing to pay to upgrade the systems? Should such an upgrade be subsidized federally? Statewide?

I agree with Inner City Blues--this is not a partisan issue. There's no way to determine any benefit specifically to any party from an improved voting system--who's to say in advance which way a mistaken vote is likely to go? Statistics would indicate only the minority parties are likely to reap any benefit from this, since mistaken votes are equally likely to end up on any of the candidates--thereby given Nader supporters a big advantage . So, its not a matter of partisanship--it's a matter of accuracy.

I will say I do see why Democrats are primarily against such systems. First, most poor precincts are Democratic--these are the ones least likely to be able to afford new systems (and also the ones currently with the most antiquated ones). Second, I think electronic voting would make Internet voting more plausible, and statistics show that higher income people (more of whom are Republican) are more likely to have Internet access. Therefore, down the road, electronic systems might make it more likely for more Republicans to vote. None of which is why I am in favor of it, btw. I think voting systems should be subsidized, so that they can be standardized--where you vote shouldn't determine how you vote. As for the Internet, more and more people are getting on it every day. I think any partisan advantage that might currently be gained would be fleeting.



CJ, I wasn't trying to pin that tail on you--simply pointing out that such recounts should be inherently better in an electronic system since they would take such issues out of the system. Electronic systems would have no 'hanging chads'--votes, once cast, would be deterministic and final. You could build in safeguards to prevent multiple votes, missing votes, etc. You could then allow the voter to review their choices. Once that button is pressed, end of story. Recounts would be essentially unnecessary, and, if done, would take just minutes (seconds?). Does this prevent any mistakes from being made? No. But those same mistakes would still be possible in the current system. And there would be far fewer of them.

http://www.verifiedvoting.org/article.asp?id=1261



amf

I wouldn't agree with you pertaining to electronic voting machines being a success in Georgia.

http://www.commondreams.org/views04/0212-11.htm




Searcy and Jekot say that if the computer software was not certified in the first place, the insertion of multiple uncertified patches further compounds the situation.

The inference above, of course, is the possibility that Georgia’s machines were manipulated prior to or during the 2002 elections. When asked whether the software could be manipulated from afar, Jekot said “they can be easily manipulated at multiple locations within the process, not just from afar”. Whether the machines were manipulated or not, however, the integrity of the process is in question.

also

http://www.commondreams.org/headlines03/1013-01.htm



Seems a little fishy to me. And nothing can be proven because there were no backup audit trail which Cathy Cox agreed to.

I agree with you Hobbes (I am also in the IT industry) that some kind of technology solution can be designed. What I have been arguing the entire time is that no one has bothered to design this solution. Companies like Diebold are more interested in "getting to market" to lock in profits, than they are interested in doing things right. If they had, things like some of these articles describe wouldn't happen.

I have also been arguing that especially in public sector government institutions, change does not come easily when it comes to IT (or anything for that matter). Implementing electronic voting is going to require a lot of folks to go get more training to support the machines and to conduct the elections. Process changes are needed. Finally, people need time to adjust to the machines and procedures and get comfortable with them. All of these things take time and it isn't something that can be rushed.

I also very much agree with your point about availability. Making voting electronic makes it much more available for all citizens - which is a good thing. However, it needs to be equally available for all citizens.

Finally, if the voting public is going to accept these machines then they are going to have to prove themselves in several elections. So far we have some success stories (a good thing) and we have a decent number of failures ranging in severity from minor to counties pulling the machines entirely and going back to the old system. In order for people to trust in the process, there need to be far more success stories than there are failures, the failures need to be in the 1% range I would think. This is especially true of this election. Knowing that the election will be incredibly close, and with the 2000 election fresh in mind it is extremely important that people trust the process.

Yet another news update on this from Wired.com - E-Vote Machines: Secret Testing. To continue my last point from the previous post:



Does it bother anyone else that all of this e-voting stuff is being done in secret here? Wired is not known for being a partisan publication, they care very deeply about technology.


This to me seems to be a pretty compelling reason to slow this push for e-voting down a bit and Udall is 100% correct here. With as tight as the election is this year, can you imagine how loudly the cries will be from both sides of the political spectrum if the machines fail in an important district?

More Flaws in eVoting, including an extremely easy to author hack found - article:




Granted this would take insider knowledge but Harris had this to say:


If you are a tech geek you really should read the entire article, it is pretty interesting. This proves once again what I have been saying all along, these machines are not ready for primetime and they need to work successfully in several local and state elections first before we even consider a national election.

Just an update.




This is shaping up to be the never-ending election.

I'm not as accustomed as others to using emoticons to persuade with, but let me try to respond in kind.

I found the following link to a Common Dreams Newscenter article, originally published Thursday, September 4, 2003, Will Bush Backers Manipulate Votes to Deliver GW Another Election? which sums up the public fears about electronic voting in a couple of quick references:


Yes Amlord, we do see Republicans at the head of most corporations. Remember, the haves and the have mores are what Bush considers his "base." Diebold in particular has been sending steady messages of "Ignore the man behind the curtain." and "Trust our software." They have fought leaving a paper trail that would allow a recount.

Meanwhile, we have to presume that their chief executive is still committed to helping Ohio deliver its electoral votes to the president this year; and in an election this close, why should we think he'll stop at Ohio? A union newspaper I received today had a list of 10 elections that were decided by a single vote, and as no. 11, it mentioned that John Kennedy was elected President by a margin of 1 vote in each precinct in Illinois. It occurs to me that software could pick the actual time out of the air with the signal from the atomic clocks, or GPS signals. It could then take the first vote for John Kerry, record it as being for George W. Bush, and then delete that program from its software. A test before election day would not activate the program. An examination of the software after the fact would not find the program that had altered the vote count, unless no one on that machine had voted for Kerry.

I'm not a professional programmer, I'm just paranoid.

I find that electronic voting seems to lack transparency.

I think in Florida the basic things have not been changed to rectify the disaster of 2000. Some of the reasons for this seem to include incompetence, corruption, political fenagling, and corporate influence.

The short of it is that I do not trust or electoral system as it stands. Autonomy seems to be theoretically loca, but these local officials also seem to be out of their league when in comes to interacting with state directives or resolving controversial events without deferring to a higher political authority.

Replacing confusing systems with computing machines created by E.S.& S. that have had a series of glitches and do not provide a voting record does not solve the problem. I want to have a paper record of my vote when I leave the polls. Actually I would rather have a piece of paper that confirms that I (verifies my name) voted for a certain list of candidates before I leave the voting booth.

Anything short of the seems to lacking accountability. I use a computerized grading system to report my grades, but I also still keep a hard copy to double check things if a discrepancy seems to appear.

Having an appointed Secretary of State who tried to overrule the state legislature and put in place an executive rule to prohibit manual recounts does not seem to be part of the process of fixing the real problems and the negative perception of electioneering in Florida.



Fla. judge requires manual recounts

Now there are two different systems being used, one optiscan system and one touch screen system. While assurances have been made that the latter are similar to the former in terms of accuracy rates, it has been reported that they are not.



This would seem to allow a system to continue that was used as the justification for interfering in the recount procedures by the USOC in 2000.



None Dare Call It Treason

We should be using a system that has clear double checks to it. Not secretive or opposed safeguards. These elections need to be cast out into openness and we need to shout down all voices that call for suppression of an open auditing procedure.

I don't think the electronic voting systems are infallible or more error prone. I think that they need to be designed to instill the greatest amount of trust in the way they record votes and send their record forward. At this point I am inclined to the Afghanistan method of sealing up a ballot box and escorting the box with an election official who monitors that the seal does not get broken until the donkey makes it to election central.

All systems can be rigged. Transparency and a genuine attempt to create a system to ends all flaws is important. This is a sacred trust and it should not be influenced by partisanship, lobbyists for a particular voting system, or anybody or anything else that is not directly aimed at gaining back the public trust in our election system.

May I make a note that on this board at 11:35 (central), every other post is by a liberal...just an observation, nothing else meant by it.

If I had to venture a gamble, Diebold doesn't hire based upon party affiliation, and the organization is flush with both democrats as well as republicans. Considering the recent polls, America is nearly split in terms of liberals and conservatives. I also believe that Diebold is a respectable organization with some safe guards to prevent question of tampering.

That's the biggest point people seem to miss.

They act like all three big evoting companies (Diebold, ES&S, Sequoia) only hire republicans ready to perpetrate felonies of the highest order. There must be some secret handshake or something I haven't heard about.

Now, they need to figure out how to put secret code in the software that will cast the votes they want without having a set configuration of who is democrat or a republican. They are generic machines for any type of election.

Then, they have to make sure that all the audits done in house and by the election officials work ok. Meaning, they randomly cast certain audit votes that MUST be produced accurately. And there has never, and I mean NEVER been once instance of an evoting machine making a factual error. They've all been related to human error.

Now, after they get through this conspiracy, they need someone on the inside to activate this trojan code - at all 40,000 precincts across the country (or however many there are).

All this without either democrats or honest republicans figuring this out.

We aren't done yet. Now, we have to believe that those three companies have invested millions and millions of dollars into these products where getting caught completely puts them out of business. Businesses screw each other over every day. But not themselves.

This fraud aspect is too absurd for words. It assumes the worst in people and assumes some people are smart enough to pull something like this off while others are too stupid to figure it out.

We don't get receipts now and you don't ever want one. And why? Because then we have proof of purchases for votes. People could sell votes and/or your boss might want to make sure you vote a certain way. At hire-at-will states (like mine), he can fire you for fun. So firing you for voting a particular party is easy.

Besides, if you don't believe the machines, why believe the receipts? Just because you got an accurate printout doesn't mean the results are tallied correctly if you believe in this fraud.

The best protection is the process. The evoting cynics continually leave that out. Bev Harris (a leading evoting cynic) shows a monkey deleting a log file to prove how easy it is to alter an election, but can't show how a monkey gets access to do it. With no controls, any voting process can be rigged. So, there are very, very tight controls in the process. If you are afraid that your vote may not be counted correctly, then you need to explain how every audit vote ever cast came out correctly. Election officials do 100 times more work to make sure evoting is ok versus punch ballots.

If you are afraid evoting will not count your vote correctly, why do you put so much faith in current technology? The same criminals would be running the store. Heck, for all you know, somebody could be calling in a made up number with the results he/she wanted.

Evoting is the most impartial and accurate game in town. With under and overvoting counts somewhere around 7%, evoting reduces it to .7% (due to undervoting - not overvoting). The disabled can vote in private and with dignity.

But mostly, we don't have to watch that bug-eyed guy holding up punch cards to examine dimpled hanging pregnant chads anymore.

You first assumption is that American is split between Liberals and Conservatives, I’d argue the split is between Dems and Rep’s, the majority of Americans are moderates for which they have no political party.

Your second:

Who’s minding the safe guards? THAT’S the problem with electronic voting that leaves no paper trail, there is no safe guard.

Just yesterday (Lesley's post), and imagine in all places, Florida, there were multiple failures with electronic voting devices for the early birds. An elderly lady was on the news this morning and God bless her, she summed it up quite nicely: “It sucks”. (I usually don't use this word but I'm just quoting grandma).

It’s one thing for any of us to debate, on either side, the merits or concerns of electronic voting. It’s quite another for anyone here to feel they’ve been horn swaggled out of a cast vote. If I were caught up in that mess yesterday I’d be on the steps of Jeb’s house this morning.

And aevans179, if you want to cast partisan stones then consider that in 2002 your Republican congress passed the Help America Vote Act, then proceeded to not provide funding. Hmmmm, would have made Jeb's task at hand harder?

Take a look at this article about some of the initial problems with electronic voting, if you thought hanging chads were a problem, this is much worse.


Meanwhile, the county clerk is blaming this on the voters, not the machines:


I think this backs up what I have been saying the entire time - yes we need this technology, but it is not ready for prime time and debuting it in a presidential election is not the best way to introduce it.

Whats funny about this is even the Bernalillo County Clerk Mary Herrera says people have said when they voted a straight ticket it went to the opposing party but still wants to blame people as being ignorant. Hopefully they have paper ballouts people can vote on if they feel they can not be confident in the results of the machine. I mean come on if I push a certain nominee and the other nominee gets a check next to it several times I would not trust the machine at all. And what happens to the people that this is supposed to help out with having a easier way to vote if they don't see the mistake like the elderly?