Talk:Electronic voting

Ronnie Dugger's excellent article

Those working on this article would do well to look at Ronnie Dugger's excellent (although polemical) article "How They Could Steal the Election This Time" in The Nation August 16/23, 2004. Aomg other things, it's a reasonably good survey of the more worrisome examples of failures (especially security failures) in electronic voting. Along the way, he also names most of the leading critics of electronic voting, although I think he gets the balance a little wrong (former ACM president Barbara Simons is mentioned only in passing, while Bev Harris gets several column inches). Anyway, a fertile source if someone wants to expand this article. -- Jmabel 17:44, Aug 22, 2004 (UTC)

This is a very technical article. Which is a good thing on some levels, I suppose, but misses some of the (not-really-covered by anybody in the media) issues inherent in the E-voting issue that bubble just below the surface. I've started a list, people can feel free to add comments or items as they wish. -- Penta 12:14, 20 Oct 2004 (UTC)

Disabilities and E-voting

1. Disabilities and E-voting: News flash for the able-bodied (the vast majority of you), but the current options for the disabled voter make a secret ballot an impossible dream, particularly for the blind. Absentee ballots are hard to use, and usually require another person, which makes the whole idea of a secret ballot, something most people would riot if they didn't have, a joke. Meanwhile, voting at the polls with currently-popular systems requires a bit of a clown-car effect, with at least 4 people in the booth besides the voter. Again, the idea of a secret ballot is sort of a distant dream (and a sick joke) when that is the case. This makes E-voting a very, very hard-fought, emotional issue among the disabled, and yet it is not being covered. (anonymously added Oct 2004)

Concur: Does someone want to take on writing this? In fact, this is why the League of Women Voters in the U.S. originally gave their unqualified support to e-voting before taking their later, more nuanced, stance. -- Jmabel | Talk 20:28, Oct 25, 2004 (UTC)

That's all well and good -- so let every district have _one_ ATM-style voting machine for the exclusive use of the disabled. It could even be voter-verifiable through the use of an OCR pen. Minority rights are utterly moot when there's a significant risk of majority rights infringement. Tell me: why can't Diebold make its voting machines half as secure as its banking machines? 66.75.250.175 05:09, 17 October 2006 (UTC) (actually, it was me, LeoTrottier 05:39, 17 October 2006 (UTC))

If that is addressed to me, I agree with you entirely. I'm entirely in favor of paper ballots, with any use of computers being limited to marking them (after which the voter can examine them) and counting them (with a simply programmed vote counting machine that is entirely separate from any voting machine) and then keeping the paper ballots for a possible recount. But the article isn't about what I think. However, both the original and modified stances of the League would belong in the article if someone would do the research and get some citations. - Jmabel | Talk 05:36, 20 October 2006 (UTC)

Cut from article

User:Nikita Borisov cut the following from the article with the comment "doesn't sound like NPOV, also not sure it's factual."

"(although some, such as Germany, have stubbornly stuck to hand-marked paper ballots until today)"

The only POV issue I can see is the word "stubbornly". However, the material does seem to be factually incorrect, since Germany apparently uses at least some electronic voting; however, here is a 1990 German paper ballot; their 1997 electoral law refers to paper, as in "A voter…who due to a physical handicap is prevented from marking the ballot paper, from placing it in the envelope, from handling the envelope over to the Electoral Officer or from placing it into the ballot box himself or herself may be aided by another person." Can anyone sort this out? Maybe someone from Germany? -- Jmabel | Talk 20:28, Oct 25, 2004 (UTC)

I've cut this section, entitled "Testing", from the article to here, for reasons I will explain immediately below:

Many people believe that the "butterfly ballots" did a bad job indicating voter intent. However, analysis shows that all the other voting methods in Florida 2000 were probably even worse ("Test the System", [1]).

The laws requiring a secret ballot make white box testing of electronic voting machines (and non-electronic voting machines) impossible while they are in use.

How can someone objectively test several types of voting machine to determine which type works best? Is it possible to design voting machines to make it easier to run such tests?

• "Many people" is very vague
• "Analysis shows" when citing one side on a controversial matter is contentious
• The questions at the end are not encyclopedic style.

The one statement about white box testing is valid, and I'd be happy to see it restored someplace in the article (probably fleshed out, since as it stands you pretty much need to be a computer type to understand it). -- Jmabel | Talk 05:49, Jun 16, 2005 (UTC)

There are actually three things that make a white box test providing full coverage impossible, lack of hardware schematics, lack of source code for the software and firmware, and complexity. The first two can be addressed by using a documented core like MIPS, and using software and firmware from reputable companies which provide the source code, but due to the complexity we don't have enough time left in the life of the universe to do a full white box test. The advantage of the old mechanical computers was that they could theoretically be whitebox tested, because they were open and simple enough. In practice it is considered too expensive to verify each gear and lever in each machine for every election, but a non-exhaustive whitebox test of is performed on each machine in NYC before each election, which is be repeated exhaustively if a candidate objects to some results and requests one. In 2004 a machine was found with ground down gears that resulted in the overturning of one election in the northern suburbs. The secret ballot did not prevent the whitebox testing, since the test could be performed before and after the election, and the election repeated if the bug is found after the election. Zenyu 12:29, August 10, 2005 (UTC)

Who is watching this article

I (User:Joebeone) thought it might be nice to have a list of people watching this page to connect usernames to credentials, if desired.

• User:Joebeone, Joseph Lorenzo Hall: a voting systems researcher from UC Berkeley's School of Information.
  • Generally, (informal) practice is that people's watchlists are their own private affair. -- Jmabel | Talk 17:06, August 7, 2005 (UTC)
  • That's fine, of course. -- User:Joebeone

Estonian E-Voting links

http://www.vm.ee/estonia/pea_172/kat_340/7025.html http://news.bbc.co.uk/2/hi/europe/4343374.stm

--3 Löwi 16:30, 15 October 2005 (UTC)

Not sure what to do with this

http://yro.slashdot.org/article.pl?sid=03/11/25/213206&tid=172&tid=103&tid=17 —The preceding unsigned comment was added by Floodo1 (talk • contribs) 23 Nov 2005.

Recently added (and removed)

Electronic voting in parliaments, for some reason, is only associated with the most inaccurate measure of collective opinion ever invented: the 2,500-year-old two-option majority vote. A computer program, Decision-maker, analyses any voters' profile according to eight different methodologies simultaneously: majority/plurality voting, two-round voting, alternative vote, approval voting, serial voting, a Borda count, a modified Borda count (or preferendum), and a Condorcet count. [www.deborda.org]

Can I just say "huh?" -- Jmabel | Talk 07:27, 3 December 2005 (UTC)

I second that "huh?" -- Joseph Lorenzo Hall 21:00, 3 December 2005 (UTC)

It seems to be some very incompetent voting system advocacy. A little knowledge of voting systems is a dangerous thing. rspeer 21:23, 3 December 2005 (UTC)

Got my Votomatic

Credit for inventing it (& so a %the blame for Florida: Fiasco 2000) goes to William Rouversol... Trekphiler 06:06, 23 December 2005 (UTC)

Walden O'Dell stuff recently cut from article

I just removed:

In the United States Walden O'Dell is the CEO of Diebold, a chief supplier of electronic voting machines. For years, O'Dell has donated to Republican candidates. In September of 2003, he held a $1,000-per-head GOP fundraiser. He has been a guest at President Bush's Texas ranch. He raised more than $100,000 for the 2004 Bush reelection campaign. O'Dell in the fall of 2003 wrote a letter pledging his commitment "to helping Ohio deliver its electoral votes to the President."

from a bullet point about large-scale fraud. This should be somewhere else as it's not evidence of fraud. It is evidence that the (former) Diebold CEO was a republican and that he assisted the campaign and made a dumb remark. -- Joseph Lorenzo Hall 02:43, 17 February 2006 (UTC)

Receipt language revert

I just reverted the following contribution in the advantages section:

Electronic voting may offer a printed receipt of important, and relevant information such as: whom an individual voted for, location where they voted, time of day they voted, date they voted, and a barcode. This ensures that a paper-trail is created in the event that technical difficulties hinder the voting process.

This is problematic for many reasons. I know of only one country where voting equipment issues some sort of record that is publicly available which indicates how you voted (Kazakhstan). That's problematic due to voter coercion and vote-selling. Some cryptographic voting solutions issue receipts, but they don't indicate how you voted. I think maybe this contributor was trying to get at VVPAT. -- Joebeone (Talk) 22:43, 16 March 2006 (UTC)

In quite a few U.S. states, there are laws against people being able to leave the polls with any sort of receipt demonstrating how they voted. This is intended as a measure against vote-buying: you don't want the vote buyer to be able to demand a receipt. - Jmabel | Talk 05:56, 23 March 2006 (UTC)

POV status of recent rewrite

The rewrite which was recently done by an anon changed the article a lot, and I can't tell what the purpose of the rewrite was. It worries me that the sections about opposition to electronic voting were condensed and moved down in the article.

There were some NPOV improvements in the rewrite, I will acknowledge; for example, a large paragraph of original research opinion about what humans can verify (which appeared to be citing a source, but that source turns out to just be Ken Thompson's classic "Reflections on Trusting Trust"). That did need to go.

However, a section about the disadvantages of Internet voting seems to be missing, as well as links to Bev Harris, Michael Shamos, and the Volusia error.

Most disturbingly, one section used to be NPOVised by the introduction:

People for the American Way cites as the principal advantages of electronic voting:

It now simply says:

The principal advantages of electronic voting:

User:68.50.103.212, if you're reading this, can you comment on the reasons behind your rewrite? rspeer / ɹəədsɹ 22:56, 23 March 2006 (UTC)

Maybe the Opposition section should be done right for once. I could take a shot at it this weekend. I've made some small edits. I think the opposition and problems sections need some work (parts of the problems section could use elaboration).

-- Joebeone (Talk) 01:25, 25 March 2006 (UTC)

User_talk:68.50.103.212 just reverted back to his rewrite which removes a lot of information. 68.50.103.212, we need to figure out a way to incorporate the stuff you want in Electronic voting without blasting a lot of good information out of existence. Could we please start a dialog here? (also posted to their talk page) -- Joebeone (Talk) 20:23, 26 March 2006 (UTC)

Response

I actually had an edit conflict with you Joebeone while posting this

Sorry for not commenting on the the discussion page for the rewrite. That was my mistake, but the entire article was riddled with inaccuracies and was very incomplete. The POV in the article was not the reason for my rewrite. I simply wanted to correct the innacuracies and expand the article to include correct information about the modern technology.

• "a section about the disadvantages of Internet voting seems to be missing"

That section is not missing. The section previously titled "Problems with electronic voting" was split into "Opposition to Electronic Voting" and "Possible malfunctions using electronic voting"

• As for the PFAW section, it is still linked. I thought more that that seemed unnecessary. I don't think that information is disputed, but if the consensus is to put it back, I will not object.
• Joebeone, if you find any problem with the oppostion section, please rewrite. I only ask that you remember that you keep the information relevant to the section. There are number of articles written about election controversies, and this is not one of them.68.50.103.212 20:31, 26 March 2006 (UTC)

Please explain what you mean by "blasting a lot of good information out of existence." I added information not removed. I have no problem with starting a dialogue. It seems you are removing my information.68.50.103.212 20:35, 26 March 2006 (UTC)

For anyone else invloved, Joebeone and I are working it out. 68.50.103.212 21:11, 26 March 2006 (UTC)

After re-reading this, I like your edits. I think it's much more coherent now. Rspeer, can we work to add back pieces that you think should still be there? (Such as having PFAWF in the text to clarify the source, etc.). -- Joebeone (Talk) 22:07, 26 March 2006 (UTC)

Sure, I've put that back, though I wonder why .212 didn't put it back himself after we objected. I think there has to be more to this, and I wonder what other POV is hiding in the rewrite. rspeer / ɹəədsɹ 05:25, 27 March 2006 (UTC)

Rspeer, could you elaborate? If you see anything missing that was valuable, we should add it back. I had always thought that singling out Shamos' recommendations was NPOV. -- Joebeone (Talk) 20:21, 27 March 2006 (UTC)

Rspeer, I'm sorry you thought I should replace the PFAW line. I thought we were going to discuss whether it was necessary here and once we came to a consensus edit the article. I still think that a reference would be better, but as I said before, I will yeild to the consensus od the group. Please feel free to search for any hidden POV. I would tell you that there isn't any, but I feel like I might just be waisting my breath. Thanks in advance for any helpful edits though68.50.103.212 21:16, 27 March 2006 (UTC)

PFAW

I think the test to PFAW is appropriate as it makes it clear where these advantages are coming from to eliminate any POV (many think that PFAW have a very specific POV, and they'd want to know where the info. is coming from). Thoughts? -- Joebeone (Talk) 00:08, 29 March 2006 (UTC)

Possible Problems

I just removed the following from the heading, "Possible malfunctions using electronic voting":

• Diebold machine review
• Diebold system discussion
• The U.S. League of Women Voters, who generally favor electronic voting, nonetheless point out that "HAVA allocates $100 million to make polling places physically accessible, but there is no national definition of 'accessible' or a deadline for implementation." [2]
• Because the software used in electronic voting machines is often not available for public review, it could contain undetected mistakes or deliberate cheating. Clint Curtis, a former employee of Yang Enterprises, stated that, in 2000, at the request of Congressman Tom Feeney, who was then the Speaker of the Florida House of Representatives, Curtis developed a "vote fraud software prototype" ([3]) that could alter machine results. [4] Both Feeney and Yang have denied the charge, however, and some critics of electronic voting have expressed doubt about Curtis's charge. [5]

The reason I did this is because these aren't possible malfunctions but items about other problems with voting machines... I'm going to try to work these into somewhere else in the article. -- Joebeone (Talk) 15:47, 22 April 2006 (UTC)

Added a benefit of electronic voting

This article listed a very short number of benefits of electronic voting so I felt enclined to add the following to the relevant section.

Further to this, what many people do not realise is that electronic voting can supply several benefits that paper-based schemes cannot. For example, an electronic scheme might be able to give each voter a receipt that she can use to check that the vote she cast is actually included in the tally. In today's paper-based ballots the individual voter simply has to trust that the ballot is included in the count. With an electronic scheme it is possible to create a receipt such that the voter can check on a web site that her ballot is included but cannot use the receipt to prove how she voted. If she could prove this of course, she would be able to sell her vote or be forced to vote in a certain way. Two systems that fulfill this are Prêt à Voter and Punchscan.

Also, Jason Kitcat, although he does speak on electronic voting, has no formal training nor has worked or does work for a company providing eVoting solutions so I took the liberty of changing the link to his blog from "UK-based eVoting expert". --Doje 20:40, 22 July 2006 (UTC)

I'm not sure what this type of paper receipt is supposed to "prove" here: an untrustworthy system's claim of its inclusion in the count still doesn't mean that it was counted at all, let alone accurately. - Jmabel | Talk 01:06, 28 July 2006 (UTC)

I think Doje was trying to accomodate the two new schemes that he mentions (Prêt à Voter and Punchscan) which do allow a person to know that their vote was counted correctly but would not serve as proof to t third-party sufficient for vote-selling (or coercion). -- Joebeone (Talk) 00:03, 30 July 2006 (UTC)

Exactly, there are now systems that can provide a receipt to provide voter-verifiability while still being receipt-free. The receipts are encrypted and can only be decrypted by a number of trusted parties that perform an anonymising mix. Thus, a voter cannot use the receipt to prove the contents of the vote to a coercer but she can use it to check that the vote has been included correctly in the decryption mix. Whether or not it has been decrypted properly is another matter and both the systems I mentioned are universally verifiable in the the decryption phase. This means that eventhough the voter cannot check that her particular vote is decrypted properly anyone can check that the full decryption is performed correctly. --Doje 20:51, 30 July 2006 (UTC)

I'm still confused. How can paper receipts taken away by the voters provide any guarantee that vote totals have been correctly tallied? At most, you can know that the (possibly untrustworthy) system gives you a (possibly untrustworthy) assurance that your vote was correctly tallied. No? - Jmabel | Talk 04:06, 3 August 2006 (UTC)

It turns on having some evidence on the receipt that the voter saw (or heard in the case of blind voters) in the voting booth next to the candidate. They then check a public message board or something like that to see if their piece of evidence is posted. If so, they know they're vote was counted as cast. The evidence is meaningless unless the vote-buyer or coercer was physically in the ballot booth when they cast their vote (or if the casting process was recorded by the voter). -- Joebeone (Talk) 19:44, 4 August 2006 (UTC)

Sorry, I still don't follow. I believe I see several issues:

• First and foremost, how does this in any way prevent (for example) the company that makes or maintains the machines adding several entirely fake ballots for their preferred candidate at each polling place? How do you verify that no actual voter cast those votes?
• Second, do I understand correctly that the idea is that all ballots are effectively visible but anonymous, but you can verify which one is yours by matching the receipt number? If so:
  • What is to prevent the machine from giving the same receipt number to two or more voters who voted identically, but only counting those as one vote? They would each see that their vote was "correct", but what they cannot see is that only one vote was counted, rather than the number actually cast?
  • If you can, after the fact, verify the content of your own vote, why couldn't you show this to the vote-buyer?

Perhaps I am misunderstanding, because the description of how this works is so sketchy; is there somewhere I can find a more comprehensive description? - Jmabel | Talk 07:12, 6 August 2006 (UTC)

I think this is talking about End-to-end auditable voting systems.

When I first heard a (sketchy) description of E2E, I was as skeptical as Jmabel -- it seemed completely impossible to me, also. Of the ones I've looked at, I found ThreeBallot the easiest one for me to understand, so let's use it for an example.

• First: ballot stuffing: The poll workers at the polling place count the total number of people who voted; anyone who suspects they might be part of the conspiracy can stand outside the polling place and count exactly how many people went in. There should be a corresponding number of ballots (3 times as many ballots with ThreeBallot).
• Second: Yes, every ThreeBallot (should) have a unique number on it. After all the ballot images are published, you should be able to find one image with the same serial number as the copy you took home as a receipt, and compare the image to your reciept to look for tampering.
• What prevents the machine from giving the same receipt number to 2 voters who voted identically? With ThreeBallot, you have paper ballots with supposedly unique numbers printed on them. After you mark the ballot, it's too late to change the serial number to the same number as someone else who voted identically. Before you mark the ballot, if the machine *guesses* that you are going to vote exactly the same as someone else, and illegally gives you a ballot with the same receipt number as that other voter, then they might get away with it if they always guess right -- but how could the machine (or a person) possibly guess perfectly accurately before you mark the ballot?
• Why can't you show this to the vote-buyer? Ah, this is the tricky bit. With ThreeBallot, you can show your receipt to anyone, and anyone who sees your ballot can confirm that the official published images exactly match your receipt. But the receipt alone says nothing about who you actually voted for. The receipt is a copy of only 1 of the three ballots; it takes at least 2 and in some cases all 3 of your ballots to prove who you voted for, and you are only allowed to copy 1 of the ballots. All 3 of your ballots are published, but there is no way for you to prove that the other 2 are yours.
• If someone modified the ballot you copied, I see that your receipt proves ballot tampering. But what prevents someone from modifying the other 2 ballots that you didn't copy? Yes, someone who modifies the other ballots will get away with it. But the trick is that they don't know which ballots are the "other" ballots. When you make the copy, no-one (other than you) is supposed to know which of the 3 ballots you copied. As long as you keep that a secret until after all the images are published, then any tampering with any one of the 3 ballots you cast has a 1/3 chance that you will have proof that tampering occurred. So we expect that, out of 100 tampered ballots, we will have about 33 pieces of proof that someone tampered with the ballots. Even if only half the voters go to the effort of checking their receipt against the published images, we still expect about 16 voters to have enough evidence to blow the whistle on this scandal.

Please help me improve the descriptions of each of the End-to-end auditable voting systems articles so they are less "sketchy" and answer all these questions for the next reader. --70.130.47.149 03:50, 17 September 2007 (UTC)

(Thanks for pinging my user talk page.) That is, at least, an intriguing system. I haven't thought through all of the implications, but the first obvious question is: what is its advantage over simple mark sense forms, the approach which most opponents of electronic balloting seem to prefer? - Jmabel | Talk 04:57, 17 September 2007 (UTC)

I'm not sure I understand your question.

ThreeBallot can be implemented using the mark-sense paper ballot counting machines currently in use where I vote. Such an implementation is a kind of mark-sense ballot form, so it has no advantage over every mark-sense form.

Perhaps you are asking "What advantage do end-to-end auditable voting systems have over other voting systems that are not end-to-end auditable?"

The whole point of all the end-to-end auditable voting systems is that they satisfy 2 seemingly contradictory requirements:

• (a) a voter can prove whether or not his selection was included in the final vote total.
• (b) a voter must not be able to prove to a vote-buyer that he really did vote the way the vote-buyer wanted.

The simple mark-sense ballots currently in use fulfill (b), but not (a). With simple mark-sense forms, what prevents a small conspiracy among some of the people who count the ballots from "flipping" votes, forcing their candidate to win? And if there is no conspiracy, how do you convince the losing candidate that there was no such conspiracy?

The "Everyone in favor, raise up your hand" style voting fulfills (a), but not (b).

With end-to-end auditable voting systems, you get both (a) and (b). If ballots are "flipped", then the voters who cast those ballots have proof that tampering occurred somewhere. If no voters blow the whistle, then the losing candidate must come to the conclusion that either (a) the election was fair, or (b) none of his supporters care enough to blow the whistle on his behalf.

Does that answer your question? --70.130.45.233 09:00, 22 September 2007 (UTC)

Criticism of E-Voting is off

The main criticism from me, and from slashdot.org as well isn't the lack of a person counting, but the lack of a paper trail. There's no possibility of a recount with Diebold machines, and some others. Once you've voted, you have no idea what was recorded, and there's no way that it can be verified that the machine's data didn't get corrupted somewhere along the way.

I have heard of some machines which do, however print a receipt out to show you what you voted for, and if it's correct you tell the machine that you are finished and it drops that receipt into a box. -khanjar24@nospam gmail.com

Indeed. One of the solutions that has been proposed to the problems of DREs is a combination of:

• A voter-verifiable paper ballot.
• Paper ballots preserved to permit effective recounts.
• A certain amount of routine auditing (comparing the paper ballots to the electronic results), partly on a random basis and partly based on allowing each candidate or party to single out (after the election) a small number of precincts to be audited.
• If that audit shows more than a certain (very low) level of discrepancy, a wider audit (up to and including a complete recount if wider problems are then observed).

I don't know from where this is citable, but if we can find it, it should be in the article. I've heard things like this from both Barbara Simons and from Eric Nilsson of CPSR, but I've heard them in conversation or in meetings; I'm not offhand sure what they've published (probably worth someone searching a bit).

A few more issues that are not, or are barely, covered in the article.:

• Proprietary voting machines are a bit of a license to print money. Whenever any adaptations are needed, the company that manufactures the machines has a virtual monopoly on the ability to do the work in question. Because of the sunk costs in the machines and the high expense of trashing them and getting different equipment, they pretty much have a captive market.
• It has often been suggested that many of the problems would be solved by open source approaches, in that any competent programmer could verify how the code works.
  • For what it's worth, I'm pretty skeptical on the value of this. Presumably, the code that would be examined is just source. You'd still have the problem of whether you can trust the compiler, the process by which it the code is compiled, linked, installed on the machines, etc. If you've got a paper trail based on voter-verified ballots, it's moot whether the software is open source or closed. If you don't, I don't believe you can ever have enough trust in the software to assure an honest election.

Again, I don't have references handy for any of this this, but I believe it is out there, and I suspect that someone with some time and diligence could follow this up and improve the article. - Jmabel | Talk 05:25, 23 September 2006 (UTC)

Incidentally: there is a difference between "voter-verifiable" and "voter-verified". In the former case, the voter can check that their vote was properly registered. In the latter case the voter does check that their vote was properly registered.

Cleanup?

Why was this page marked for clean up? There was no reason given on this page or any of the Wikipedia:cleanup related pages. It seems to me this article is one of the better Wikipedia articles availaible. I made some significant updates and I would say it's clean. I'm going to take down the cleanup tag, if there any objections, please lets discuss.

Electiontechnology 21:34, 13 October 2006 (UTC)

"Opposition to electronic voting" and "Possible malfunctions ..."

I find this separation to have a significantly mollifying effect on the arguments against the use of electronic voting, and if I were a lawyer trying to defend a voting machine company, this would be my first move when putting together my arguments.

Opposition and possible malfunctions are most certainly not separate issues. Indeed, most of the time if (the consequences of) possible malfunctions were remedied (e.g. by having a voter verified paper trail with random audits, etc.) much of the opposition would disappear. For these reasons I recommend that these two sections be merged, somehow, perhaps by the creation of a section called "Current criticisms of electronic voting in the United States".

In support of this merger, it should also be noted that problems with electronic voting are issues largely related not to electronic voting per se but instead to how the US is currently implementing electronic voting. Again, most if not all of the criticisms leveled against electronic voting could be relatively easily dealt with (at least when compared to the magnitude of the operation required to implement electronic voting in the first place). LeoTrottier 05:31, 17 October 2006 (UTC).

Also, why is the section labelled "possible malfunctions". Aren't the instances listed documented and verified problems caused by the use and/or misuse of electronic voting machines? Shouldn't it simply be "instances of voting machine malfunction". LeoTrottier 05:35, 17 October 2006 (UTC)

I disagree. Malfunctions (and I agree with you that we should not qualify this with "possible") are a technical issue. Opposition is a political issue.

Also, keep in mind: all election systems are in some degree flawed. I've heard Eric Nilsson (a computer scientist who has been involved in monitoring several notable elections, including the first election under universal franchise in South Africa) make a pretty strong case that even the best-run elections have a quite significant margin of error, typically about 3%. (Of course, in an honest election, it is random error, not systemic bias in one direction or another.) You point out the difference between electronic voting as such and how the U.S. is currently implementing it. It is also relevant to remember, to distinguish among (1) a theoretical perfectly run election, (2) best practices for other methods of election (such as mark sense forms), (3) actual practices for other methods of election. That is, we have to be careful not to compare corruptly run electronic voting to honestly run paper balloting.

Hold on: if you refuse to drink a glass of milk because it has gone bad, it's not because you are opposed to milk! By dichotomizing into "possible malfunctions" and "opposition" those in opposition are painted as being so for reasons other than possible malfunctions (or at least this is the implication). Current outcry, such as it is, is against shoddy, uncertified, unauditable, corrupt, etc. implementations of electronic voting systems. LeoTrottier 19:30, 22 October 2006 (UTC)

By the way, I completely agree with you on paper trail and random audits. So, do you feel like doing the legwork of finding citable sources on this? - Jmabel | Talk 05:36, 20 October 2006 (UTC)

I suppose I could visit the various websites and look at what they're complaining about. I have to say, though, that I've never seen just opposition to electronic voting, and framing it this way makes the 'opposition' seem dogmatic and ideological. Rather, all I've seen has been "electronic voting is easily manipulated and unauditable for reasons X, Y, and Z. Therefore we should use paper ballots which don't have the same problem." It's been my experience that those who are most opposed to electronic voting are not luddites and people scared of technology -- they are instead people all too familiar with the vulnerabilities of computer systems (see the various discussions on slashdot.org if you're curious).

Perhaps a better title for the section would be "difficulties in implementing electronic voting systems" (or something like that), because saying "posssible malfunctions" conjures up the image of one little voting booth going haywire. Another possibility for a better title might be "issues in current US implementation", a section whose content would would better reflect what's going on: people's deep mistrust for electronic voting in the US because some implementations slated for use in November are corruptible and unauditable and have a history of misrepresentation and abuse. This would also provide a venue for mentioning the alleged frauds perpetrated by some of the voting systems companies (it would also be good to have a venue where things like this could be incorporated).LeoTrottier 01:53, 23 October 2006 (UTC)

Finally: if we're willing to list the 'advantages' of electronic voting, shouldn't we also list the disadvantages? At the very least, it is certainly a disadvantage that electronic systems are more complicated and (at least up front) more expensive than paper ones. LeoTrottier 20:13, 22 October 2006 (UTC)

I feel that this article's sections are still too biased in favour of electronic voting. For example the section about Ontario municipal and provincial elections (where I live) implies that it's only a matter of time until electronic voting spreads. I don't know if that's true--there are doubts about its reliability, so we mostly still use paper ballots in Canada. One could equally rephrase it to say something like "Many municipalities still insist on using electronic voting methods" which would imply that it was on its way out. Dan Carkner 00:52, 14 November 2006 (UTC)

Dan, I don't mean to be dense, but what about that section "implies that it's only a matter of time until electronic voting spreads?" It seems to me to be an observational list of examples of electronic voting as the title suggests. Other than rephrasing to add a POV statement like you mentioned, what would be needed to correct the section? -- Electiontechnology 01:10, 14 November 2006 (UTC)

You're right, maybe I read into it too much, but I still think the article doesn't mention the skepticism that various electoral agencies in Canada feel about it. Oh well..Dan Carkner 23:59, 14 November 2006 (UTC)

Above in #Criticism of E-Voting is off I think I've given the outline of some of the most important case against. But someone needs to do the legwork to cite for this and get it into the article. - Jmabel | Talk 06:42, 17 November 2006 (UTC)

There is some information on the webpage of the dutch campaign against voting computers.

They point out that

1) A few very well placed individuals will be able to hack an electronic election by

replacing the binary file that goes into he machine or replacing the data file.

2) Electronic elections do not allow for recounting votes.

http://www.wijvertrouwenstemcomputersniet.nl/English nielsle May 2007

Why is there no 'electronic voting in the US' page?

We have "electronic voting in ..." pages for many countries, save the one in which e-voting is causing the most controversy. Why is this??

Because I got tired after writing all the other ones, the US one is going to be the most extensive, and more of the rest of the article is already covering much of the US. Electiontechnology 04:50, 31 October 2006 (UTC)

Dutch Decertification

Someone just added:

On October 30 2006 the Dutch minister of Home affairs withdrew the license of 1187 voting machines, about 10% of the total number to be used, because it was proven that one could ´listen out´ the voting from up to 40 meters by receiving radiowaves. National elections are to be held 24 days after this decision. The decision was forced by a group called wijvertrouwenstemcomputersniet we do´t trust voting computers.

I added a need for cite, fixed some grammar, and moved to a more appropriate section. Does anyone know more about this? Can anyone verify or update it? It's pretty poorly worded. Electiontechnology 04:50, 31 October 2006 (UTC)

Just added the requested information and made minor textual edits. Still not sure about the language as i'm not a native speaker. Also note that the decertification is based on the EM radiation from a normal computer screen which is used in the DRE equipment Van Eck Phreaking. The same attack will probably apply to all DRE equipment that features standard computer monitors. The remote read-out voids the secrecy of the voting process. Miw 15:39, 1 November 2006 (UTC)

Diebold AccuVote-TSx DRE

Does this model produce a truly voter-verifiable paper record? My understanding is the Dieblod models require the voter to use a barcode scanner to read the paper record, which is not voter-verification. The scanner is doing the verification, and the voter is merely reading what the scanner reports. If there is some agreement that this model isn't truly voter-verifialbe, the photograph of this model should be removed, or at least moved from the section about VVPAT.Awinkle 18:48, 10 November 2006 (UTC)

The Diebold AccuVoteTSX (with AccuView printe) does not require a barcode scanner or any other type of electronic device and is visually verifiable by the voter. Thought not particularly visible in the picture the visually verifiable paper is under clear plastic using a variation of the Mercuri method. If you take a look at the enlarged picture [6] the human readable printout is there, though mostly obscured by that small gray plastic door that covers the VVPAT during storage. I guess the simple answer is yes, and the picture is valid. (but a better picture is always welcome.) --Electiontechnology 19:44, 10 November 2006 (UTC)

Paper ≠ voter-verifiable

We mention voter-verifiable paper trail several times, but we don't even mention that not every paper-trail system is voter-verifiable, because not every paper-trail system is human-readable. For example, there are systems involving bar codes that sharee the advantage of a permanent paper record, but don't have the advantage of voter-verifiability. - Jmabel | Talk 06:50, 17 November 2006 (UTC)

I'm with you on that one. Feel free to make the edits, there is a lot more crypto and the like on the Voter Verified Paper Audit Trail page. I think the section is lacking, but expect feedback from a number of users... Good luck! --Electiontechnology

I've made a few edits, but the problem is that I don't have citable sources for most of what I know about this. A few of the key critics are people I know personally through CPSR; for the most part I've learned my way around this from meetings and talks, rather than from written materials. - Jmabel | Talk 19:46, 19 November 2006 (UTC)

That's an area I might be able to assist. I have access to a fairly sizable library of articles and papers. If you let me know a topic, I might be able to find something that covers it. --Electiontechnology 23:42, 19 November 2006 (UTC)

Electiontechnology, I basically laid out the case above in #Criticism of E-Voting is off. I did this on the talk page rather than in the article, because I figured that I would not really benefit an article on a controversial topic by laying out a view without citation for whose view it is and where it has been published. - Jmabel | Talk 03:15, 22 November 2006 (UTC)

VotersUnite Report

In January 2007 three e-voting activist groups published a report on 1022 seperate incidents made to a non-partisan hotline that operated the day of the November 7 midterm elections.

I think if this is going to be listed, it need some sort of explanation. This wasn't any sort of a scientific study and the "1022" incidents include many duplicates, second hand stories, news clippings, and some completely irrelevant stories that are not an "incident" of anything. Please check out the data: http://www.votersunite.org/info/2006E-VotingReports.xls Here are a choice few:

• Incident E-1302 is listed 3 times.

There are a lot of other duplicates like this.

• Incident E-1336 "While driving vehicle, caller reported that she heard on the radio that machines not working at the Princeton Dr polling locaiton in Dayton. Informed local EP contact who stated that a volunteer was present at Princeton Dr and that they would address the problem"

This is a second hand report of a news report...

• (Incident E-1565, Richland SC) "The caller reported that his friend voted earlier this morning, and told him about a problem he encountered. The caller’s friend intended to vote democrat down the line, but noticed that the republican senate candidate was listed as a democrat, and so he believes his senate vote was incorrectly recorded for the republican rather than democratic candidate. The caller did not know what type of machine his friend used. The caller votes at the same site and was worried that his vote would also be miscast.I asked the voter to tell his friend to call this number so we could hear about his experience firsthand. I advised him to vote and let us know if hie has any problem himself."

That reminds me of this friend I had who ate Pop-rocks and soda...

• (Incident E-3782, Union NJ) "Voting machine was unplugged and running low on batteries, but was plugged in by a poll worker."

This is another non-incident.

All is all the "conclusions" drawn by this study are based of poor data with zero scientific relevance. I think at best it would be misleading to include it in its current form. -Electiontechnology 05:22, 8 January 2007 (UTC)

where did You get the "study" from, the text says "report from a hotline data" ? i think that sums up pretty good that it's not a scientific study and no conclusions are drawn from it in the article. i don't know of any study which gives a comparable good overview on voting machine problems so unless You know one i suggest we leave it in there. the other "documented problems" are mostly news-articles, that's also not scientific.--Taintain 23:45, 8 January 2007 (UTC)

Conclusions are drawn from the data in the report, that's what makes it a study by the way. These conclusions aren't supported. If it's to be included, which I had and have no problem with, it needs to explained (which I have) otherwise it would be extremely misleading and POV. -Electiontechnology 00:30, 9 January 2007 (UTC)

it calls itself a report, that's where i got it from. i would rather call Your version POV. why do You have to emphasize it's unscientific, it doesn't claim to be. a lot of references on wikipedia are unscientific without a warning. and why did You delete it in the first place instead of improving the text as You've done now, would have been more productive.--Taintain 22:00, 9 January 2007 (UTC)

I didn't ask where you got "report" from. To answer your question I moved it to the talk page because in it's then current form was POV and inaccurate. I wanted to start a discussion about how to post it in an accurate form. It's clear that the there are duplicates and inaccuracies in the data for the report. We don't currently know how many or how inaccurate. I think without being qualified this link is POV. Since the section title is "Documented problems" I think it would be useful to link the data there and then the study in the "Advocacy, Commentary, and Criticism" section.

Absentee ballots

Absentee Voting is still a means of a secret ballot - the second person needed is to verify the signature on the sealed envelope containing the voter's marked ballot, which is then inserted into an outside envelope to be mailed to their registrar.—The preceding unsigned comment was added by Mozart54 (talk • contribs) 22:14, 9 February 2007.(moved by Electiontechnology 00:06, 10 February 2007 (UTC))

EIRS

The EIRS data needs to be appropriately quantified. It is a raw dump of unvalidated national hotline data including requests for sample ballots (#12167), second hand radio reports (#14216), any report of machines without papertrails (#16472, #28804) and dozens of blank entries as well many other similar "incidents" all labeled a "machine problems." Please view the data: https://voteprotect.org/index.php?display=EIRExportMapNation&tab=ED04&cat=02&start_date=&start_time=00%3A00&end_date=&end_time=00%3A00&search=&go=Search --Electiontechnology 01:46, 17 April 2007 (UTC)

E- Voting in Venezuela

There is very little information about recent elections in Venezuela. Since 2004, the recall referendum of Hugo Chávez, Venezuela has implemented e-voting technologies that have been controversial since the begining. Recall referendum results were considered electoral fraud by many. Including a study made by Ricardo Haussman along with Roberto Rigobón, professors at Harvard and MIT, one of them Dean, who claimed there was 99% of electoral fraud probability.

These unconfidence in the elections and democracy heavilly influenced Chavez's opposition in their determination to boycott the 2005 legislative elections, that were made and ended with the result of Chavez's coalition of parties winning all seats in parliament. Since then, critics have raised of authoritarism in Venezuelan government decisions, including the shutdown of RCTV, Venezuela's oldest TV network, and a constitutional reform that will transform Venezuela into a socialist republic with no term limits for President (both head of government and head of state in Venezuela). —Preceding unsigned comment added by 201.208.35.230 (talk) 17:18, 22 September 2007 (UTC)

DRE Merge Content

Taintain, please read more carefully before editing. Much of the content is already in the article. I will outline the changes below. Additionally following Wikipedia guidelines on Criticism and generally improving the article, the criticisms are disbursed throughout the article in the relevant sections.

1. Your addition, "Critics of DRE Machines claim that there is an increased risk of electoral fraud and if the security of the DRE software is compromised, election results could be tampered with in an undetectable fashion." existed as "Critics of DRE Machines claim that there is a risk of the DRE software being compromised in an undetectable fashion." I've replaced that with your version, but it was hardly "forgotten."
2. Your addition, "Other critics charge that foreign hardware could be inserted into the machine, using a man in the middle attack technique, and call for DRE machines to be physically sealed." existed as "Critics, such as the group "Wij vertrouwen stemcomputers niet" ("We do not trust voting machines"), charge that foreign hardware could be inserted into the machine, using a man in the middle attack technique, and call for DRE machines to be physically sealed." That edit will remain.
3. Your edits "These claims are countered by the position that review and testing procedures can detect fraudulent code or hardware, if such things are present, and that a thorough, verifiable chain of custody would prevent the insertion of such hardware or software." Will be added to the appropriate sections.
4. Your addition "Another method to detect fraudulent voting machines are parallel test elections which are conducted on the election day with randomly picked machines." Has been added to a new section "Testing and Certification."
5. Your addition "(The ACM published a study showing that, to change the outcome of the 2000 U.S. Presidential election, only 2 votes in each precinct would have needed to been changed.)" was intentionally left out. The study results seemed to be an aspect of elections in general rather than electronic voting, but it has been added back in.
6. Your edit "A workgroup of the National Institute of Standards and Technology (NIST) stated in a discussion draft, "Simply put, the DRE architecture’s inability to provide for independent audits of its electronic records makes it a poor choice for an environment in which detecting errors and fraud is important." The report does not represent the official position of NIST, and misinterpretations of the report has led NIST to explain that "Some statements in the report have been misinterpreted. The draft report includes statements from election officials, voting system vendors, computer scientists and other experts in the field about what is potentially possible in terms of attacks on DREs. However, these statements are not report conclusions."" existed entirely in the relevant section "Audit trails and auditing."
7. The section "Demonstrated Laboratory Attacks" already existed entirely. I moved it to "Documented problems" as it's previous location under "software" I realize the attacks were more than just relating to software.

--Electiontechnology 15:50, 9 November 2007 (UTC)

• i've reverted that merge since it doesn't make any sense, see discussion on DRE voting systems.

--Taintain 11:19, 10 November 2007 (UTC)