California to Require Paper Voter Receipt

DDumitru writes "Wired reports that California Secretary of State Kevin Shelley will require all electronic voting systems be equipped with a voter-verifiable paper receipt. This receipt will not be retained by the voter, but deposited at the polls and may be used to audit electronic election results. All new voting system installed after July 1, 2005 must include the new printers. Existing systems, including the systems already installed in four counties must be retrofitted by July 2006. It looks like the public outcry about Diebold and other voting equipment manufacturers has been heard, at least in a very major market for these machines in the US. It should be very difficult for other states to not follow suit."

Comments

It's too late

[SpaceLifeForm]

This needs to be implemented *before* the elections next November to avoid a mess again.

Re:It's too late

[bussdriver]

The cheaters behind all this do not really want it fixed. Everyone talking or actually doing something is STILL putting it off long enough for them to gain working control of the government. At which point, they could manage to set things up so they never lose control. Any case, they will do plenty of harm until the problems are fixed. This is not STUPID mistakes, its INTENTIONALLY STUPID mistakes!

Re:It's too late

[cravey]

It's not too late. Now the machines can fudge the ballots that they print to prove that the ballots they fudged electronically are valid. What a great idea.

Saying it's too late suggests that this might actually make a positive difference. I see it as a large step back. Now the candidate can say 'See? The paper ballots PROVE that I got all the votes and the other candidates didnt get any! The other candidates must have voted for me too. The paper receipts PROVE it.'

Re:It's too late

[cravey]

ok, scratch that. I need to remember not to post before I wake up.

Re:It's too late

[SpaceLifeForm]

Your argument is actually the reason why the machines need to be open source. How do you know that the paper record isn't already fudged with the current state of the software? You don't.

Re:It's too late

[cravey]

No, my argument was stupid and about something totally invalidated by the article. I wasn't awake yet.

But now that you mention it... If the machines are opensource, aren't the people loading the code into the machines ALSO the ones who've been loading uncertified code into the machines? How do the voters know that the source code they've seen is what's in the machine?

Re:It's too late

[anthony_dipierro]

How do the voters know that the source code they've seen is what's in the machine?

md5sum?

Re:It's too late

[cravey]

ok, so the voters are going to offload the code installed on the voting machines and assume that they're not only being given the code that's actually running, but that every machine is the same?
Yeah, That'll work. I can just see the election judge stopping everyone from voting because some guy with a laptop wants to personally verify each voting machine. If not, then who does the verification? Who watches the watchers? It's not Diebold or M$. And it's certainly not the government.

Re:It's too late

[anthony_dipierro]

ok, so the voters are going to offload the code installed on the voting machines and assume that they're not only being given the code that's actually running, but that every machine is the same?

You could just install md5sum on the voting machine itself.

I can just see the election judge stopping everyone from voting because some guy with a laptop wants to personally verify each voting machine.

No need to stop people from voting. These checks can take place before the voting begins.

If not, then who does the verification? Who watches the watchers?

Anyone who is listed on the ballot or registered as a write-in candidate certainly should be allowed to appoint someone. Really, I'd say anyone registered to vote in that district should be allowed to.

Re:It's too late

[anthony_dipierro]

How do the voters know that the md5sum source code...?

Look at it. The source to md5sum isn't that complicated. Then check it against a bunch of known test cases.

I wonder

[Anonymous Coward]

if diebold CEO is still promising (and meaning it) to deliver W..
Oh, wait.
The printer was delayed until AFTER the next major election.

Re:I wonder

[squiggleslash]

California would be a difficult one to rig, even if Diebold wanted to. It would look exceedingly suspicious if it went to GWB (yes, they elected the Republican candidate for Governor, but that's a Republican who is far more liberal than Bush is.)

Re:I wonder

[Daniel Dvorkin]

I hope you're right. The problem is that when things get screwed up at the state level, the only recourse is to turn to the federal government -- and who runs the federal government right now? That's right ...

Well, okay, there is another recourse -- "1776 is the cure for 1984" and all that. But I really don't expect to see that happen, no matter how egregious the vote fraud gets.

Re:I wonder

[nyseal]

That's the problem with CA; too many Dems and too many reps. Maybe we should even the playing field so the US is not represented by a lot of illegal aliens and a governor who has the MPAA in mind. What a joke.

2005? 2006?

[Anonymous Coward]

Way to deflect the issue, kids. "yeah yeah, we have to be accountable... but in two years". Too bad they're going to have a little thing like "presidential election" first before all that comes about, huh?

Democracy works?

[Hackie_Chan]

Democracy works?

Re:Democracy works?

[quigonn]

It does, but (currently) not in the US.

Re:Democracy works?

[epicstruggle]

What democracy!!! Last time i looked, we lived in a republic. Wow, schools have certainly gone from bad to hopeless, when even the type of govermnent is not know. Just so we all know, the public does not directly elect the president. Its the electorial college that does that. So its possible to have candidates who have more total votes and still loose the election.

Learn the type of govermnent you have and then youll be able to properly complain about it.

later

Re:Democracy works?

[YOU LIKEWISE FAIL IT]

It does, but (currently) not in the US.

The United States is not a democracy, it's a Representative Republic. The distinction is important, because if electronic voting ever realised its full potential, the argument that 'the people can't elect their own president directly because it would be too difficult to coordinate, too hard to count' would be difficult to defend.

YLFI

Re:Democracy works?

[clifyt]

Its a little more than that.

Having a Represenative Republic ensures that the medium guys get as much attention as the smaller guys.

I might be wrong, but isn't over 50% of the population in 3 states? New York, California, Florida? In such, these 3 states COULD control the presidency, find a candidate that is willing to sell out the other 47 states and give these guys anything they want and there ya have it. The logistics of giving everything ya promised is a little tough, but it could be done with enough behind the doors deal making.

Face it, if we could elect our own presidency as a people, Gore would have won hands down. Ironically, it was said his people were preparing a legal strategy that said his presidency was legitimate even though he had a minority of votes, because the popular liberal thought was that Gore would win (8 years of great economy, nothing really bad happening except a little oral sex, technology happening at a pace thats never been seen in 100 years...that sort of thing), but that the conservative states would rise up in anger and far more would vote against him than for him.

Ironically, it happened the other way :-)

Even though I voted against Bushie, my vote meant nothing to the total votes. My state went for Bushie, and thus ALL these votes went that way. Its a small price to pay to make certain that the individual chunks of peoples get represented...remember we are the United STATES. It happens at State level and THEN nationally.

Personally, I think the way electorial votes are counted should be modified a little. Allow states to split votes. If you have 20 electorial votes to give as a state, and 45% vote one way and another 55% goes the other way, why should you give ALL your votes to the man with the winning total. As long as there is some granularity within the state, why not set it up that way. All your people are represented, your state is represented, and you suddenly become more important nationally as you can't just push it over the edge, ya gotta REALLY make certain its in your favor.

Then again, it probably goes against the idea of a unified state making its decision as a whole...

Re:Democracy works?

[geoffspear]

I might be wrong, but isn't over 50% of the population in 3 states? New York, California, Florida?

No, you're very wrong. If that was the case, Gore would have moved his entire campaign to Florida, campaigned for Nader in every other state in a deal to keep him from being added to the ballot in Florida, and would have won the election easily.

Also, states are allowed to split their electoral votes. Some states themselves forbid their electors from voting for anyone but the candidate with a plurality of the popular vote, but even in those states (I believe about half of them) that allow the electors to vote for whomever they want, they don't.

A better system, and one that would actually preserve the intent of the Electoral College, would just be to just have the electors themselves run for election and let the voters choose electors whom they trust to make the best decision. Right now, the parties pick a group of electors for each state that they trust to elect their guy, and if they win a plurality in the state they get to have their electors in the College.

Re:Democracy works?

[jc42]

Ah, the old false dichotomy between a democracy and a representative republic. The US is, of course, both. Check your dictionary.

But the electoral failure three years ago was a result of something else that few people other than historians ever mention: The US Electoral College was in fact set up by the Founding Fathers as an explicit check on the power of the masses. They were afraid of a popular demagogue winning an election and overthrowing the established order. Not an irrational fear, as illustrated by several cases in the 20th century where this happened in some other countries. So they devised that peculiar scheme whereby the voters choose "electors", presumably well-to-do members of the established parties, and those electors then decide amongst themselves who should be the president. This system can overturn the wishes of the masses, and that's exactly what it was designed to do.

In this case, it did have some help from a court that ordered a halt to the vote counting, so that one state could "choose" the desired set of electors. This is something that the Founding Fathers apparently didn't anticipate, and has thrown a major monkey wrench into the works. But this isn't the first time; check out the 1876 election for a precedent. ("Rutherford Tilden election" is a good set of keywords for a search site.)

Now we have the have the phenomenon of new voting equipment being widely installed, from a company whose CEO has brazenly promised one party that he can deliver states to them in the next election. Information about this equipment supports his claim fully.

So maybe we can truly remove the US from the list of democracies ...

Re:Democracy works?

[Anonymous Coward]

Having a Represenative Republic ensures that the medium guys get as much attention as the smaller guys.

I might be wrong, but isn't over 50% of the population in 3 states? New York, California, Florida? In such, these 3 states COULD control the presidency, find a candidate that is willing to sell out the other 47 states and give these guys anything they want and there ya have it.

This is a great myth, but it's simply not true. Because of the electoral college and the way they vote (all votes go to one candidate even if the candidate was elected by 51%) the electoral college actually makes things worse. Consider the following (electoral votes data, census data) (assuming everyone can and does vote, doesn't matter, divide all pop numbers by 4 or whatever):

Plurality of votes:
Total US population (July 1 2002): 288,368,698
Votes needed: 144,184,350
Pop 51% (see below)
CA: 35,116,033 17,909,176
TX: 21,779,893 11,107,745
NY: 19,157,532 9,770,341
FL: 16,713,149 8,523,706
IL: 12,600,620 6,426,316
PA: 12,335,091 6,290,896
OH: 11,421,267 5,824,846
MI: 10,050,446 5,125,727
AZ: 5,456,453
Total: 144,630,484

So you'd need all votes of 9 states (assuming a minority who don't are offset by a minority who do in other states).

WITH the help of the electoral college:
Total Votes: 538
Needed to Win: 270
Pop 51% pop
CA: 55
TX: 34
NY: 31
FL: 27
IL: 21
PA: 21
OH: 20
MI: 17
GA: 15 8,560,310 4,365,758
NJ: 15 8,590,300 4,381,053
NC: 15 8,320,146 4,243,274
Total: 271

So you'd need the electoral votes of 11 states to win. The problem here is that each state only needs to be won by 51% of the vote. You could potentially have a situation where a candidate promises half-assed to sell the rest of the country out for these 11 states, and she would win 51% of the votes in these 11 and none in the other 40 (DC). Wonder what that would look like population-wise? 83,968,838. In our current system, an election could potentially be won with 29% of the total vote! So are we assured that "the medium guys get as much attention as the smaller guys"? No -- not with 51% of a state winning the vote. Make it so the electoral college splits their votes according to the way the state voted, and then you have protection...

Re:Democracy works?

[squiggleslash]

The United States is not a democracy, it's a Representative Republic. The distinction is important

*sigh*

What do you think democracy is?

Look, there's a bunch of people who insist that "government types" is like something out of Sid Meier's Civilization. You get Despotism, Communism, Republicanism, Democracy, et al, and they're all contradictory.

This is not true. Most of these are measures, or they describe a constitutional arrangement. For example, a monarchy and a republic are opposites (or rather, mutually exclusive), however, a republic can be democratic, communist, or a whole host of other things. Indeed, a monarchy can be democratic.

Confused? The latter sounds impossible?

A Democracy is a regime where the legislature is answerable to the governed. This is currently the case in the United States and a whole host of other countries. The US achieves this by having a directly elected congress and, currently, a directly elected senate. Thus, both are answerable to the people, and as a law cannot be passed without being approved by both houses, it furfills the definition of democratic. "Aha", I pretend to hear you cry, "But the US also has a constitution with a bill of rights in it preventing laws from being passed that the people's representatives might be in favour of!" Well, sure, but it's a constitution that can be changed, again, by the people. There's no office that can veto changes to the constitution, and currently, with constitutional changes requiring the consent of bodies (states, senate, etc) that are all answerable to the governed, it remains democratic.

Does this mean that the US is not a Republic? Far from it. Indeed, even Britain and most of the other European monarchies are democratic, because, for now, those monarchs have agreed to let their elected legislatures be responsible for all lawmaking, and the executives in those countries, however formed, are bound by those laws.

Thus, a "Representative Republic" is not an opposite of a democracy, it is a democracy.

People tend to think that democracy means more than it actually does. I regularly read people who think that democracy means "rule by plebicite" or some other such nonsense. Bollocks.

Further, I also read a lot of too-clever-for-their-own-good types who propose that America isn't democratic because they read the Federalist Papers and, boy howdy, those papers say it's a Republic and not Democratic. Well, sure. And back in the late 1700s, there were no guarantees that individual states would be democratic, and the constitution left the choice of how to appoint Senators up to the states.

The world has moved on since then. Senators are now directly elected. No states are undemocratic. The US is currently a democracy, and thanks to the 14th Amendment, that's going to be difficult to change except by changing the current constitution.

Re:Democracy works?

[SoupaFly]

Currently only two states allow their electors to split their votes: Maine and Nebraska. Votes are split based on congressional districts and the remaining 2 votes are given to the candidate that won the majority of the districts.

Re:Democracy works?

[CaptBubba]

THis system doesn't represent the middle people. What it does is get the canidates to make one or two stops in the largest cities of the states. All they care about is getting more of the vote than the rest of the canidates. Those living in rural areas never get addressed, or talked about because they take work to get to. Why should you bother if you are effectivly going to get their votes anyway when the state's electors vote?

A popular vote would compleatly throw off current caimpaigning practices. Because those who are in charge are the ones who are campaigning, there will never be a change unless there is a huge public outcry that can't be silenced by replacing the voting machines or a similar measure that realy doesn't change anything but looks like they did.

Re:Democracy works?

[NegativeK]

A Democracy is a regime where the legislature is answerable to the governed. This is currently the case in the United States and a whole host of other countries. The US achieves this by having a directly elected congress and, currently, a directly elected senate.

Not so sure about that one. If you really want to get pedantic about it, the US is a representative democracy. We trust certain people to enact our will, but do not do so directly on the federal level. Our founders placed specific restrictions on majority rule, on the basis of Greece. We have a lot of trappings intentionally in the process that slow the democratic process down. Yes, through the current processes, a majority can end up reigning over the minority. If a super majority had enough cajones(sp.?) to do it, they could completely abolish the Constitution. Of course, in that case, the US wouldn't be the US in all but name.

In short, the point I'm trying to make is that you can't just call the US a democracy, and leave it at that. It's a Representative Democracy, and was (for the most part) intended to be that way. Take either part away, and you have something that is arguably different.

P.S.: Another thing that distinguishes the US from a direct democracy is the possibility for _massive_ agency loss. Some (not necessarily myself) would point to the 2000 election, but you could also point to issues like RIAA lobbying, etc. Does the majority _really_ believe in what the DMCA actually states?

Re:Democracy works?

[geoffspear]

Actually, only 2 states do split their votes; that is, Maine and Nebraska require the electors besides the 2 at-large electors to vote based on the vote in the congressional district they represent (which, as you'd expect in small, largely homogenous states, have always gone for the same candidate anyway).

About half the states actually require the electors to vote for the winner of the popular vote in the state, while the others (as the system is supposedly designed to permit) allow the electors to vote their conscience (even though they were picked by the winning party specifically to vote in their candidate.) These electors do not represent a specific district that may have gone against the plurality in the state; the point is that an elector from these states, once sent to the Electoral College, could, if he or she wanted to, vote for absolutely anyone. Of course, one would expect the parties to have a rigorous enough screening process that none of the electors would do anything but what they were sent to do.

replace the printer

[Anonymous Coward]

If a machine runs out of paper, he said, Sequoia would recommend that poll workers remove the entire printer component and replace it with a new one so that workers do not need to touch the receipt roll.

Yeah right, so his company makes even more money...

Re:replace the printer

[Politburo]

This is yet another example of the ridiculous double standard. If the machine could have the paper replaced like normal reciept printers, you would be clamoring about the security of the paper record. They make it so that they never have to touch the thing, and you complain about the cost. It's one or the other, guys. Things cost money.

Re:replace the printer

[drinkypoo]

Why is it one or the other? Why can't you have a not-readily-opened paper tape cartridge like a 110 film cartridge that autofeeds the paper, and is replaced when it's done printing? If it breaks, you just yank it out, and replace it with another one. You can have both, without spending a load of money.

Re:replace the printer

[TRACK-YOUR-POSITION]

These guys are taking public tax money and, if allowed to escape scrutiny, have the power to fuck with the democratic process. I think the public has the right and responsibility to give them as much shit as fucking possible. We the people are supposed to be the boss of these guys. Being a government contractor means you need to take a lot of shit from the public--it's just part of the job description. You've got a problem with that? Then go back to Russia and leave those of us who actually enjoy democracy and free speech the fuck alone.

Re:replace the printer

[Politburo]

Because there isn't a design like that (that I know of). It takes money to design hardware.

Re:replace the printer

[Politburo]

I'm not saying you can't raise shit, but it's really fucking dumb to raise contradictory points. Look up the "Contractor's Triangle" if you still don't understand.

Re:replace the printer

[TRACK-YOUR-POSITION]

What?

What to count

[Councilor Hart]

But what will be counted?
The electronic votes or the printed votes.
Who says they are the same?
Who says people will even bother reading the piece of paper?

Re:What to count

[Yokaze]

- Unless there is a complaint, surely the electronic votes, otherwise the printed votes.

- No one, that is why one can complain and request a recount, should there be a doubt.

- Well, those who care to vote, surely care how it will be counted. I mean, it is not like one goes there just for fun.

Re:What to count

[StenD]

Well, those who care to vote, surely care how it will be counted. I mean, it is not like one goes there just for fun.

No, they go there for a promised pack of cigarettes, or a bottle of 40/40, or...

Re:What to count

[hazem]

I would imagine the electronic votes would be counted.

There would also probably be random-sample checks of various polls to make sure the tallies are the same. And if someone contests a tally, or it's really close, the paper could be used to verify the electronic count.

New warning labels

[eap]

It looks like the public outcry about Diebold and other voting equipment manufacturers has been heard, at least in a very major market for these machines in the US. It should be very difficult for other states to not follow suit.

Will Diebold voting machines should now carry warnings that state, "This voting machine contains technology known by the State of California to be harmful to Democracy"?

Re:New warning labels

[Liselle]

You'd have to paste all of those signs on the members of Congress, too, because the United States has a representative government, not a democracy.

Re:New warning labels

[Ctrl-Z]

Please, do tell how the representative government of the United States is not democratic.
Among others, www.m-w.com defines "democracy" as

1 a : government by the people; especially : rule of the majority b : a government in which the supreme power is vested in the people and exercised by them directly or indirectly through a system of representation usually involving periodically held free elections

Re:New warning labels

[Skjellifetti]

To elaborate, I always thought that a democracy was a government system where every eligible citizen voted on every issue, while a republic was a system where the individual citizens chose representatives to do the work of government for them.

Re:New warning labels

[Ctrl-Z]

Sure, but I don't quite understand why you are implying that a republic cannot be democratic. Is the United States not a democratic republic? The fact that it is a representative democracy does not make it undemocratic.

Re:New warning labels

[Liselle]

I know you are making a jab at the justice system, but it's funny you mention the 2000 presdential election. If it were a true democracy, Al Gore would be president, because he won the popular vote. The business down in Florida would not even have mattered, it's the electoral college that threw everything out of whack. :D

Re:New warning labels

[uncadonna]

This "republic" vs "democracy" nitpicking is some sort of wierd semantic quibble popular in certain right-wing circles in the United States. I think it leads to some core ideological point or other. Perhaps it is simply that "Republican" is a better name for a political party than "Democrat". Who knows? What's wierd about it is not that they make the distinction, but that they insist that everyone else make the same distinction.

Some advice to the nitpickers, then. Words mean what they mean. Saying "the dictionary is incorrect" is silly. If you insist on the distinction, you should say, instead, something like "many of the writers I am interested in make a finer distinction in using these words, specifically treating 'republic' and 'democracy' as mutually exclusive attributes of a social order, and it is in that sense that I use them as well." That would be more polite and reasonable, since most of us do not believe that anything prevents a republic from being a democracy.

If you try being more polite and reasonable, maybe the rest of us will pay more attention to your ideas.

Re:New warning labels

[Liselle]

The Republican vs. Democrat thing goes all the way back to Jefferson. It's an interesting read, but it has nothing to do with any of this, heh.

I agree that saying "the dictionary is incorrect" is silly, which is why I didn't say it. I think the definition is not 100% accurate, meaning that it's misleading, if that makes you feel better. It comes down to calling things what they are: speak plainly. Maybe it's nit-picking, but I think the distinction is important. This isn't Greece, we don't have all people eligible to vote having a direct impact on laws. I didn't vote on the DMCA. This isn't about the difference between "potato" and "potatoe", it is two completely different things being bandied about using the same word, which is wrong.

I think that about makes my point as clear as I can make it. Geez, now I feel like RMS. :)

Re:New warning labels

[David Jensen]

Sure, but I don't quite understand why you are implying that a republic cannot be democratic. Is the United States not a democratic republic? The fact that it is a representative democracy does not make it undemocratic.

A republic is a country that has an elected head of state, not a king or emperor. A representative democracy is what the US and all of the other democratic nations have. It has an elected legislature. So, yes, the United States is democratic and a republic, but it is not a democratic republic. A Democratic Republic is a pseudo-Marxist dictatorship, so it is neither democratic nor a republic.

Re:New warning labels

[Rotten168]

So you'd like for the court to be elected? The supreme court nominees had to be ratified (by the then Democratic congress) you know.

Re:New warning labels

[Rotten168]

That is not my understanding of a republic and the difference is important to me. IMO a republic is a highly decentralized union of city-states who run themselves and elect representatives who have a say in the federal government. Think Ancient Rome before Caesar.

This is different from a more centralized democracy. IANA political scientist so correct me if I'm wrong.

Re:New warning labels

[Malcontent]

Democracy is a dozen wolves and a sheep arguing what to have for dinner.

Re:New warning labels

[Malcontent]

It would be great for democracy if they had to re-confirmed every so many years. How about a no confidence vote?

I guess nothing would help. As long as judges are determined to first look at your political affliation and then look at the merits of the case we are in trouble.

Re:New warning labels

[Rotten168]

Yeah but don't you see the problem with that? You'd have supreme court justices campaigning on how tough they were on terrorists or something, other than interpretting law.

Direct democracy is not a good thing. Hell, I'm in favor of repealing the 17th amendment as a matter of fact.

Re:New warning labels

[Ctrl-Z]

I didn't say "Democratic Republic", I said "democratic republic", which is to say that it is a republic that is also democratic.

how?

[Lord Kholdan]

How does the voter know that the line printed and the vote saved are the same one? It would be trivial to make the program print a vote for candidate X and mark it as a vote for candidate Y?

Re:how?

[Anspen]

How does the voter know that the line printed and the vote saved are the same one? It would be trivial to make the program print a vote for candidate X and mark it as a vote for candidate Y?

True, but at least it would be possible to hold a paper recount, which would show such a deception.

Re:how?

[NoOneInParticular]

Just make sure that the paper votes are counted after the elections for a random selection of voting machines. If large discrepencies are found, redo the elections, put Diebold in jail.

Re:how?

[pentalive]

Repeat after me
It's not the voters, It's the Vote counters who decide.

Unless you follow the whole chain of custody yourself how do you know your vote is properly counted anyway.

I still say that the voter should have several human readable stubs to send to various vote counting groups. Sure some of the stubs will be used to get paid for the voters vote, some of them. I would rather have that and a fully verified vote.

A local talk radio show had a guest on who had seen a voting machine. The machine had gotten stuck somehow
and when the workers at the poll opened it it was full of already cast votes. This was in the morning, just after the polls opened.

If I had a stub that was proof of how I voted, and I sent my stub to a group that I trusted for a paralel count that would go al long way for trust.

There is ofcourse the non electronic way. The polling place is divided into two rooms. You identify yourself and fill out the ballot in the first room. then you go through the door into the second room where your ballot is immidiately counted. Judges from all the partys and other interested groups monitor the process at each polling place. The polls are combined under the same supervision when polls close.

Re:how?

[StenD]

If I had a stub that was proof of how I voted, and I sent my stub to a group that I trusted for a paralel count that would go al long way for trust.

How? Unless that group was sent stubs by all (or a sizable percentage) of the voters, all they can do is establish a floor on how many votes a candidate should have. They wouldn't be able to argue voting trends, because the self-selecting sample of voters who would send them their stubs would make their trends different from the electorate as a whole.

Re:how?

[TheLink]

"But in the event of discrepancy, which one is considered correct? The machine record, or the paper record? Considering one makes the other."

Neither.

If you are really serious about having a Democratic Republic, in event of a significant discrepancy, you find the people responsible and put them in Guantanamo, PERMANENTLY.

And then you start again. From scratch. Otherwise why don't you just forget about the whole thing.

I mean it's not like the USA can't afford decent voting systems and machines, it's not some 3rd world country where you'd be glad if not too many ballot boxes are stuffed or go missing.

How much is being spent to remove Iraq's WMD^H^H^H Gov? From the resources spent it would seem that who's the leader of Iraq matters more than who's the leader of the USA.

Re:how?

[pentalive]

The various "disinterestd parties" would have to colaborate on thier tallies.

The stub itself need not be human readable. but it shoudl be machine readable. And then it only releases the info to
the voter himself (password...) or the disinterestd party.

Re:how?

[StenD]

The various "disinterestd parties" would have to colaborate on thier tallies.

And if all of the groups don't collaborate (probable), or all of the voters don't provide their stubs to one of the groups (even more probable), you're still talking about a self-selected sample of voters, which makes any apparent trends meaningless.

The stub itself need not be human readable.

Huh? You said "I still say that the voter should have several human readable stubs to send to various vote counting groups.". Why the flip-flop?

but it shoudl be machine readable. And then it only releases the info to the voter himself (password...) or the disinterestd party.

And how does the "disinterested party" get access to the info? If the government sets standards for outside groups to get access to the votes, then we're back in the same situation where only the groups in power receive access.

Re:The real question is...

[terraformer]

To rig an election like that (actually fudging the paper ballots) you need to do it on such a scale to make a difference that enough people would see it and it would become clear it was systemic. Keep in mind, the machine is not going to know who will be dilligent as to look at their ballot and who will not be. It will just randomly fudge ballots.

Hey...

[s20451]

What exactly is wrong with taking a piece of paper with every candidate's name on it, and making an "X" beside your choice? This is the way things are done in Canadian federal elections, no fancy-pants touch screens or butterfly ballots or any other nonsense. Everyone gets a ballot with a standard design, from Victoria to Halifax.

Sometimes the best solutions are the simplest. If technology doesn't simplify life, what use is it?

Re:Hey...

[miruku]

because the results take hours and hours to collate and analyse

Re:Hey...

[hey]

In Canada the election results are always out soon after the polls close in the West. In fact some people in the West complain that the results are known before many people vote after work in BC and their vote doesn't count as much as others. I don't buy that complain but it shows that results get tallied up very quickly.

Re:Hey...

[fname]

Not just directed at you, but: Earth to Slashdotters. This does not require electronic voting. Marking a piece of paper with an "x" would be perfectly valid. This simply requires electronic systems to work properly and to be audit-able. Bubble cards, etc. are still ok.

One nice advantage of electronic voting is it has the potential to be very easy/quick to set up an election; there are very many other positives. This decision addresses the one giant negative associated with the process.

Re:Hey...

[damiam]

As opposed to US systems, where the results weren't known for weeks in 2000, and some results (on electronic machines), are never truly known.

Re:Hey...

[oogoliegoogolie]

It's not like they have to wait two weeks to find out who has won. Results are counted within a couple of hours after polls close. Hardly what I would call an excessive amount of time.

Re:Hey...

[Ulven]

Isn't that everywhere?

Re:Hey...

[Skjellifetti]

Too labor intensive. Columbus, OH where I live often has problems finding enough poll workers. Ballots can be very large when you have races for municpal, state, and federal offices (judges, county engineer, auditor, dog catcher, council, reps, executives - its a long list) plus ballot isues to deal with.

Re:Hey...

[1ucius]

Other reasons I've heard cited are that it is tough to accomodate blind people and multiple languages.

Re:Hey...

[GQuon]

Because somebody is bound to make that "X" accross several boxes. And then we'll have discussions about who they tried to vote for.
No. Let's have machines with names and pictures of the candidates. And if the voter tries to vote in an invalid way, give them a little electric shock (like you get from sheep fences) untill they make a valid choice. Let me see you do THAT with a paper ballot.
Of course, i totally agree with you. They could have ballots with names and pictures, black out a box, and then run them through fast scanners. If the scanner can't understand the ballot, an election official has to interpret it. This is a lot like the old chad system, only using a pen/marker to vote. Cheaper than machines, no chads falling off, and almost as fast results as voting machines.
I've done mulitple choice tests and applied for student loans in this way. No problems.
Also (a pete peeve of mine), all states should distribute their Electoral College members in proportion to the votes in the state. The Democrats should make it an issue. Ow, that's right: They don't want to change it, only complain about it after the fact. (And I guess some Democrats and Republicans want change, but they are minorities within their parties.)

Re:Hey...

[saforrest]

Canada... you mean that country that harbors terrorists intent on killing americans?

Been watching Fox News, I see.

Re:Hey...

[drinkypoo]

One nice thing about Californian ballots is that you vote by punching out holes. This lends itself readily to inexpensive or expensive optical scanning. You can do it with some equipment readily available in any old 5.25" floppy drive, or you can do something moderately sophisticated with bitmap processing; either way the results are easily machine-gatherable from the paper ballots. Frankly I think we should stick with California's existing paper ballots, and just scan them electronically. They work fine. However, electronic voting has the potential to be cheaper; I just don't think it's worth it when the voting system has so much room for fraud. If it's not open source, then I want a paper printout listing my votes. You could just break it down to a simple string and offer a voting string decoder on the state website in order to save paper, if you want, but I want something that says how I voted that I can take with me. I know which holes I punched out, when I punch out holes.

Re:Hey...

[SoTuA]

because somebody is bound to make that "X" accross several boxes. And then we'll have discussions about who they tried to vote for.

We don't care who they tried to vote for. That's a null vote, and that's that.

At least, that how we do it in Chile. Ballot with names and horizontal lines, you draw a vertical line on the horizontal line beside your candidate's name, making a pretty "+". If you touch more than one line, you vote is null and void. If you do anything but a line but your scrawling only covers 1 candidate (as if, for example, you write "This guy sucks" over his name) your vote is to that guy. But it you write "candidate 1 for president over his name, and touch another candidate's name with the "t", for example, your vote is null.

Re:Hey...

[cpeterso]

How do blind people read the touchscreen ballots? At least a paper ballot could have Braille dots.

Re:Hey...

[Jeremi]

Reasons to use an electronic voting terminal:

1. An electronic terminal can error-check your ballot and give you a chance to correct any mistakes before you turn the ballot in. If you mess up a paper ballot and don't catch the mistake yourself, your vote won't count and you'll never know that you were disenfranchised.
2. Electronic terminals are able to provide additional assistance to blind or illiterate people, by showing pictures of the candidates and reading the text aloud into headphones, if necessary.
   
3. Electronic terminals make more sophisticated forms of voting (like IRV, Condorcet, etc) practical, and thus more likely to eventually be approved.

Re:Hey...

[Jeremi]

We don't care who they tried to vote for. That's a null vote, and that's that.

That is the standard solution, but throwing away someone's vote is undemocratic and undesirable, so if there is now a better solution available (e.g. a touchscreen that makes the voter's vote is valid before it gets submitted), why not use it?

Re:Hey...

[GQuon]

if, for example, you write "This guy sucks" over his name) your vote is to that guy
LMAO. People marking with "X" could do this too: Socialists could draw little pentagrams or swastikas next to Bush, and they have voted for him. Heh, heh. Then they could go to the courts claiming their vote should be counted as a "anybody-but-him vote".

Re:Hey...

[Dr. Blue]

Because elections in the U.S. are orders of magnitude more complex than Canadian elections. In the 2000 elections, there were 150 million votes cast in the U.S. In the 2000 Canadian elections there were 12 million votes cast.

My ballot had over 80 different things to vote on, from president to state and local offices, ballot initiatives, etc. How complex was the Canadian ballot?

So with 80 different things to vote on, and 150 million people, we're talking about tabulating 12 billion votes. Hopefully you can see the benefit of some automation here....

Re:Hey...

[JimBobJoe]

As a pollworker in Columbus (actually I work in Dublin, but it's still Franklin County) I can say with confidence that counting the 80-300 ballots (per precinct) would be just as labor intensive (and perhaps less so) than dealing with the Shouptronic machines (which have a fairly cumbersome closing down procedure, and are heavy as all hell.)

Municipal races are in odd years, state/federal in even (all over Ohio), so that keeps ballots from getting too big.

Columbus, OH where I live often has problems finding enough poll workers

a.) no need to add the "Ohio" tag. we're the biggest, so let people assume it
b.) the problem is slightly exaggerated
c.) well if you're not a poll worker already, go volunteer :-)

Re:Hey...

[Lord_Breetai]

How do blind people read the touchscreen ballots? At least a paper ballot could have Braille dots.

Perhaps with a pair of headphones, it could be read to them. And how do the make their selection, you ask? A row of buttons can be included on any side of the screen with braille on them. They could say: selection A, etc.

Re:Hey...

[SoTuA]

if, for example, you write "This guy sucks" over his name) your vote is to that guy LMAO. People marking with "X" could do this too: Socialists could draw little pentagrams or swastikas next to Bush, and they have voted for him. Heh, heh. Then they could go to the courts claiming their vote should be counted as a "anybody-but-him vote".

It makes for lively discussion when its time for counting. Each voting table is manned by a "presidente" and two "vocales" (kinda witnesses) that are picked at random from the people who have to vote at that table. Also, every candidate has the right to a witness-representative to the counting, in every table. So people always argue those "defaced" votes. A friend of mine was representative for the counting on the last presidential election, and tried to argue that some guy who had drawn a dick over the other candidate's name wasn't expressing his preference :D (the guy got the vote)

Re:Hey...

[SoTuA]

Chile? You're holding up Chile as an example of a government that represents its people? Liberty? Democracy? A model for the US to follow? God help us all.

Dude, read the news. The plebiscite that got Pinochet outed was held in 1989, and Patricio Aylwin assumed his duties as president in 1990.

BTW, thanks for the US-instilled coup. (and you don't have to take me on faith on that, check the files that CLinton declassified before he ended his last term)

Re:Hey...

[jjhlk]

Bah, multiple languages. If the ballot has only names on it, then people of any language should be able to recognize the names. They'd need to recognize the candidates name somehow, when listening to information in their language. If you can't be bothered to figure out how to read someone's name as it would appear on a ballot, maybe you don't deserve to vote.

Re:Hey...

[ElectricRook]

I'm with you on keeping the punch cards. That technologys been in use for more than 120 years, and most of the bugs have been wrung out. I'm not saying that new could not be better, but new paradymes have new problems. And after that last bunch of BS, we don't need to go changing everything at once.

but I want something that says how I voted that I can take with me

Actually the reason we don't do this, is to prevent vote buying, and extortion. Imagine this: "Bring in a ballot stub that says you voted for ElectricRook, and get a free beer." or "Hey Arnold, you little skinney twerp, if you don't have a ballot stub that sez ElectricRook, I breaka you face."

Re:Hey...

[drinkypoo]

Well you're right, it's a bad idea. Besides, the free beer thing would be rad. I wouldn't do it, unless I got the beer ahead of time though, so that's not going to work. Anyway the solution I think is as others have pointed out, some kind of hash that doesn't show how you've voted, only that your vote hasn't been changed if you rehash and compare.

Re:Hey...

[JimBobJoe]

I'm a driver's license privacy activist/specialist/expert. Right now, the big fight is/will be, in Jersey.

Re:Hey...

[miruku]

the whole point is to get the machines to work properly, by having a paper receipt which can be checked

My system can do that!

[Asprin]

They should look at my voting system idea, which I outlined in my journal.

Re:My system can do that!

[balloonhead]

Important Guy Running Elections: "This system is lame. We need a better one."

Staff #1: "The Think Tank has lots of ideas, we're going through them case-by-case for viability"

Staff #2: "Look at this other country, they have a good implementation, we can modify it and try and solve the problems it has."

Staff #3: "Here are a list of commercial vendors, they all seem to have quality products except for one - Diabolic I think their name is?"

Important Guy: "No, we've tried them, what a shambles."

Staff #4: "I read some guy's Slashdot journal. How about you all shut the hell up and we do what he says? Screw your multi-million dollar deals"

Aye, fucking right. Why the hell would they be scouring slashdot for some idea of pompous preacher who thinks "I thought of that first. I'm so l33t."

PS this is a joke, although it does seem to come across as quite abrasive.

Re:My system can do that!

[Effugas]

Everyone thinks of this.

Everyone thinks they're the first to think of this.

Everyone's got the right idea -- except Diebold.

--Dan

Re:My system can do that!

[HeelToe]

What your system cannot do is prove to the voter that its internal storage records exactly what the screen does in terms of the votes placed.

No electronic system can do this.

Of course, some of our current non-electronic voting machines cannot, but some can. The voter can actually see what physical piece of evidence will be later counted to determine their votes.

Since you cannot write a proof to guarantee code does something, you have no mechanism to tie any artifact (whether electronic on-screen results or a printout of voting choices) to the actual vote being counted.

A physical artifact counted by people is all that can do this.

Re:My system can do that!

[TheLink]

There are tons of ways to do better than Diebold. I mean it's harder to get crappier than the current e-voting systems the US has.

Even a show of hands would be more accurate - you are unlikely to get less than zero votes.

Just get a bunch of crypto and security guys, they'd be able to figure out a good system that has anonymity and auditability.

But it sure doesn't seem very important to the US.

If anyone put such shoddy code in a banking system they'd be sacked with extreme prejudice. But hey no big deal, this is just the US elections after all.

It's not like it's the Iraqi elections.

Why So Long?

[Databass]

Why does the bill allow such a long timeline? By requiring a paper trail in 2005 (not in time for the next presidentail election), the legislature is clearly saying there is a problem that needs to be addressed. Why does it not need to be addressed in time for the Presidentail election?

A year is plenty of time short of deliberate sandbagging.

Re:Why So Long?

[borkus]

Because if George W. Bush does not become re-elected, they can send Governor Schwarzenegger back in time to terminate the Democrat president.

Re:Why So Long?

[Politburo]

Because that is only a year away. You have obviously never worked for government. Design (if applicable), procurement, setup and training for an election system could never occur in under a year. It could be possible, but with fundamental changes in the system. I would rather see it take 2 years, and have it done right, then have them rush a shoddy system into place for 2004.

Re:Why So Long?

[symbolic]

My guess: probably for the same reason that this was ever allowed to happen in the first place. Politics and integrity have nothing in common.

Re:The real question is...

[balloonhead]

That's irrelevant. The important thing is that the audit trail is now possible. The majority of voters don't even bother to vote - should they all be made to?

I don't think that electronic voting is really an advantage over traditional methods, especially as it's so open to abuse. But if it is implemented, then at least the possibility of verifying results is now there.

I'm sure some smartass will just claim their voting receipt is different from their vote just to kick up a stink though... enough of these could throw the thing into more doubt.

um...useless?

[amichalo]

I see two possible scenarios which make this an unrealistic solution:

(1) The receipt includes a voter ID and the results of their vote. This totally violates the anonymity of the voting process but does allow for counting.

(2) If the receipts include no voter ID but just some form of transaction ID, then why print them off at all? Just run some report at any point during the voting process to see the tally? Why not? If the voting system is compromised, then there is no way to ensure the paper votes with the transaction id, generated from the compromised system can be trusted either.

As I see it, this solution does not add value without removing rights.

Re:um...useless?

[3th3rn3t]

but wasnt the whole point to get rid of paper voting and vast ammounts of it accumulated? the receipt will just be the same thing imho. massive useless paper amounts

Re:um...useless?

[srleffler]

You missed the point. The reason to print the receipt after each person votes rather than printing off a report later is so the voter can see the receipt and verify that the machine has correctly recorded the vote. Even if not every voter bothers to check the receipt, enough will that a malfunctioning machine will be detected. The receipts than allow for a recount to be done later if there is some doubt about the machine's accuracy or if the machine crashes.

Re:um...useless?

[kasperd]

I see two possible scenarios which make this an unrealistic solution

With an appropriate cryptographic solution the receipt doesn't have to reveal information about the actual vote. And still it is possible with the right algorithm to verify, that this vote was actually counted in the final result. Unfortunately I don't remember the rest of the details about how this should work.

Re:um...useless?

[StenD]

No, the alleged point was to avoid errors caused by people confused by the notion that "vote for only one candidate" meant that they should only one candidate, and confused by the concept that a candidate's name on the same line as a spot to indicate a vote meant that in order to vote for that candidate, they should mark/punch out the spot on the same line. The "butterfly" ballot gets all of the bad press, but other paper ballots caused similar problems, if not to the same degree. Now, I'm not sure how anyone who has had errors in their bank/credit card/telephone/etc statement would think that computerized voting would avoid errors, but...

Re:um...useless?

[AlinuxNCSU]

How about option 3?

(3) The reciept includes a timestamp (perhaps), and a output of results in human readable form. This gets deposited into a locked box on your way out (as paper ballots are now). It doesn't have any indication of who the person was who voted.

This method doesn't violate anonymity in any way, and it allows the voter to verify before he or she deposits the slip whether or not the vote was tallied correctly.

Later, you can take this locked box and count reciepts like you count ballots. Randomly audit a certain percentage of polling places and you can check for discrepancies.

Is there anything I missed?

-Alex

Re:um...useless?

[martin-boundary]

You missed the following: suppose the voter votes for one candidate, but swaps the printout receipt just before placing it in the sealed box. Now the results in the machine are different from the paper results. Which one will be trusted?

BTW, the article states that you don't get to touch the receipt. You simply view it inside the printer, behind a piece of glass.

Re:um...useless?

[martin-boundary]

You're missing the point. Anonymity is lost in case the machine is flagged as untrustworthy. Every person who complains that they didn't see their vote on the printout can be tracked, since they must divulge their identity before voting.

So if the machine is deficient on purpose, you could get a list of all people who didn't vote for candidate X, including name and address. The list won't be complete since not everyone will complain, but it'll be large enough to send Vinnie over for a "talk" later on if you so choose.

Re:um...useless?

[AlinuxNCSU]

Well, serves me right for skimming the article.

For the sake of arguement though, if you wanted to, I'm sure you could figure out some (expensive?) physical way to guarantee that the receipt isn't tampered with even if it's handled. For instance, you print some kind of verifiable signature (which doesn't include the voters' ID) or you print on special paper...

Either way, to answer your question, I'd trust the paper. Why would someone modify their vote other than to be malicious? That would just be changing one's own vote (and being the idiot that causes the recount). However, to protect against such possibilities, you don't mandate a complete recount unless the differences between computer and paper are statistically significant (which depends on the total results).

-Alex

Re:um...useless?

[Guppy06]

"The receipts than allow for a recount to be done later if there is some doubt about the machine's accuracy or if the machine crashes."

Like hell they do. They're receipts, which means the voter leaves the polling place with them in hand. That makes them even less reliable than the machines we're talking about. As soon as the state asks for receipts to come back for a recount, each party will turn in a half a million such receipts they "found" somewhere.

This new law requiring these machines to print receipts is nothing more than feel-good legislation that accomplishes nothing. The machines are still the only word you have on who got how many votes, and they're still using the old, buggy software that local governments aren't even allowed to see, let alone fix. Everybody here who's posting "Hey, they finally fixed it!" is going along with the carrot these politicians are dangling in front of the voters' noses while ignoring the pork-barrel politics that's paying for these machines to begin with.

All the problems of mechanical voting machines, none of the advantages of physical mechanisms, for more money. And somehow spitting out a piece of paper is supposed to fix this?

Re:um...useless?

[martin-boundary]

I don't know why they would do it. I was just pushing the idea. What's troubling is that they could in that type of system. Sooner or later, groups of people would get the idea to synchronize their vote tampering and then it would be significant. Maybe they want to just sabotage the election, delay a result by a couple of weeks. Where there's a way, there's a motive, I guess.

Re:um...useless?

[vidarh]

They will NEVER, EVER allow voters to leave with printed receipts from the voting machine. Take one guess why.

Sale of votes and intimidation of voters.

If people are allowed to leave with receipts, we WILL see attempts to buy/sell votes because voters now suddenly have a way to prove that they have voted a particular way. Currently buying votes is unattractive, because nothing is stopping me from taking your money and voting however I please.

Similarly we WILL see attempts to intimidate voters from people who will make it very clear that if you can't show them a receipt indicating the "right" vote you will be in trouble.

I doubt any election official will be stupid enough to even think about letting voters leave the polling station with definitive proof that they voted a particular way, preventing the above two is one of their most important jobs.

Re:um...useless?

[vidarh]

This is bullshit. There is a number of ways to anonymously indicate that the vote is flawed. For instance, voters could be asked to indicate on the receipt whether or not the receipt is correct, and put it in a ballot box. A quick scan of receipts would then reveal whether or not the number of errors would be big enough to be an issue.

There is NO reason why you should need to show your receipt, as they can't adjust the numbers based on complaints anyway as it has to be assumed that a lot of people won't check their ballots, so the number of complaints won't be an accurate measure of error rate.

Re:um...useless?

[John Hasler]

> Is there anything I missed?

Yes. Just feed the "receipts" (i.e., paper ballots) into a scantron, count them, and announce the results. don't bother to have the touch screen machine do any counting at all.

Re:um...useless?

[drinkypoo]

If you wanted to solve the problem of bogus receipts being brought in, all you would do is assign them serial numbers and record the serial numbers in a database (and on the record tape.)

Re:um...useless?

[Jeremi]

This is bullshit. There is a number of ways to anonymously indicate that the vote is flawed. For instance, voters could be asked to indicate on the receipt whether or not the receipt is correct, and
put it in a ballot box.

Hmm.... this would lead to an easy way for a group of people to throw an election into doubt, by anonymously indicating that their ballots were flawed. i.e. If you think you can't win the election, you could nullify it by getting people to check the "machine messed up my vote" box even when it didn't.

Re:um...useless?

[townmouse]

And they will NEVER, EVER allow voters to vote by post. Or by proxy. Or let anyone in the building without strip-searching them for hidden cameras. These days cameras in mobile phones are preferred, but 19th-century technology is perfectly adequate for the purpose.

Re:um...useless?

[martin-boundary]

RTFA, you don't get a "receipt". You visually inspect the printer's outpout behind a piece of glass or something. There is no way to indicate anything unless you tell someone and they come and check. Sure you could just tell someone and they might not be allowed to check. But then anybody can simply claim it's broken without proof, and you *will* get a statistical percentage of jokers doing that in each election.

Yay

[Ancil]

Woohoo! Score one for the good guys.

you're missing the point

[professorhojo]

the point isn't that people will get the receipt and double-check it. although that will be a nice side-effect.

the point is that we'll have a complete paper record of who voted for who. the system will be accountable for its results instead of just numbers in an access database that could have been tampered with.

that's what "paper trail" means.

prof.hojo.
my site.

Re:you're missing the point

[k8er]

IMHO, paper records aren't that important. It's probably a false sense of security anyway. I want the system to be as open and visible as possible, which eternal vigilance in checking and double checking the processes. A little faith is always required, but I would have more faith in a highly visible, proven electronic technology than I would ever have had in a paper only system. Look at how inaccurate that system seems when they do a recount. And accuracy isn't the only issue. It is sometimes a pain in the ass to go to the polls. I want to be able to log into the polls from my pc, enter my unique voter ID, vote for my unpopular candidate and see the vote show up on my precinct ticker in real time. I don't really care if they keep a paper record with my unique ID or not. If someone can prove it's me, what are they gonna do? Is George Bush going to kick my ass because I didn't vote for him? I don't see any reason that the system has to retain your personal information when the keys are being generated. One set of keys for each eligible voter in the US. Could people sell their keys? Sure, I guess. Would it be worth it to the politician? It would be pretty hard to keep that kind of think quiet. What if some bad politician or is supporters try to strong arm people into giving up their voter keys? I don't know. Once again, there would have to be a significant amount of this going on to swing an election and they would probably get caught. And if there are people out there that bad, they are going to be doing bad things anyway. You have to stand up to them or they are going to push you around anyway. Most Americans do not vote anyway. If we make the process convenient, open, and visible enough I think more people would vote and it might make a positive difference.

Re:you're missing the point

[Ancil]

The point isn't that people will get the receipt and double-check it.

Actually, that is the point. If no one looked at what was on the paper ballots, they would be no more reliable than the electronic ballots. The basic problem with electronically recording votes is that people can't SEE their vote be recorded.

Being printed on paper and stored in a locked box is worthless if people never look at the paper and see what's printed there.

Re:you're missing the point

[vidarh]

You're completely wrong. There will be a set of ANONYMOUS receipts that should match the electronic records. If anyone demand a recount, it will be trivial to find out whether they match (just count the receipts).

Recording who voted for who would never be accepted because it would lead to massive buying and selling of votes and intimidation of voters. A system that doesn't have anonymous voting is not, in any sense, democratic because it more or less guarantee that the votes will not reflect the intent of the electorate.

paper receipt?

[ejaw5]

What's so hard about using a sharpie to fill in a (relatively large) bubble next to the canidate you want to vote for? Then use any computer technology you want to count the bubbles. Sounds cheaper to me. The paper trail is there, and only what needs to be automated (counting) is.

Maybe setup a few touchscreen kiosks for those who really need it. For the rest of us, I want my pen and paper.

Re:paper receipt?

[SemperUbi]

This is pretty much the way San Mateo County voters have been voting for years. We use black markers to connect two dark lines for the candidate we want, and then feed our ballots into an optical scanner which records our votes. It's a simple, elegant solution.

It's surprising that this technology hasn't gotten more media attention. People following the news would think the only three ways to vote are old voting machines, punch cards and DRE!

Re:paper receipt?

[David Price]

One of the biggest reasons? Blind people. DRE machines are *huge* for blind people, because they can be supplemented with an audio interface that help them vote unassisted. Computer-mediated voting allows any number of interfaces to be presented in order to overcome various disabilities.

Some of the biggest advocates for DRE machines have been advocates for the blind and other disabled people who have previously required help at the polls.

Re:paper receipt?

[TheRealNecator]

Hm, not really a reason. Blind people can use special election-papers -- which have to be counted by hand. No problem here. And: Manipulation is not nearly as simple as with these great Diebold thingies ... Why is it necessary to use a digital solution, if the old (really old -- not that punching-holes-thingies) analog one is much better (and safer)?

Re:paper receipt?

[John Hasler]

> One of the biggest reasons? Blind people. DRE
> machines are *huge* for blind people, because they
> can be supplemented with an audio interface that
> help them vote unassisted.

A touch screen that prints a paper ballot can also be fitted with an audio interface.

> Computer-mediated voting allows any number of
> interfaces to be presented in order to overcome
> various disabilities.

All of those interfaces can be used with machines that print paper ballots.

Re:paper receipt?

[David Price]

True, and true. Please note that I was responding to someone advocating pen-and-paper, non-computer ballots.

Re:paper receipt?

[Ungrounded Lightning]

What's so hard about using a sharpie to fill in a (relatively large) bubble next to the canidate you want to vote for? Then use any computer technology you want to count the bubbles. Sounds cheaper to me. The paper trail is there, and only what needs to be automated (counting) is.

Except that:
- Optical character readers are ALSO error prone, resulting in the equivalent of "hanging chad/dimpled ballot" recounts.
- Optical ballots are easy to accidentally mismark - much worse than "butterfly ballots".
- There's no assistance to check for accidental invalid voting choices (i.e. voting for four in a "vote for three" election).
- There's no assistance for the sensory (or literacy) handicapped. (As another poster already pointed out.)

Re:paper receipt?

[nurbman]

I agree. The vote (data) should be separate from the counting (processing).

The voter shouldn't have their vote sucked into a black box to become part of a proprietary system that can't be recounted by another system if needed.

This year's Toronto elections had a large 14"x10" or so ballot that was scanned in. The scanner rejected invalid vote combinations and the total was phoned in by voice as well as dialed in using a modem in the scanner as a double check.

You had results in 20 minutes with fewer errors and an anonymous paper trail if needed for a recount.

Re:paper receipt?

[Idarubicin]

- Optical ballots are easy to accidentally mismark - much worse than "butterfly ballots".
- There's no assistance to check for accidental invalid voting choices (i.e. voting for four in a "vote for three" election).
- There's no assistance for the sensory (or literacy) handicapped. (As another poster already pointed out.)

Actually, all of these problems can be solved quite readily. In a recent municipal election in Oakville, Canada, I understand that they used optical card readers. Next to each candidate's name was an arrow, with a break in its shaft. To vote for a candidate, one filled in the gap. I don't see the accidental mismarking being a problem here, unless someone does something incredibly stupid with the ballot layout (see Florida). As long as there is a single column on each ballot, there shouldn't be mismarking issues.

When you returned your ballot, it was inserted into the machine and read immediately. (There was a sleeve or some other mechanism to prevent the election workers from seeing the ballot markings.) The elector was immediately informed if the ballot had been completed correctly; if there were problems, he or she could try again (the original ballot would be marked as spoiled.)

For the visually impaired, a template marked in Braille can be placed over the ballot, with openings where they can mark their votes. I don't know if this was provided in the Oakville election, however I do know that this is done for Canadian federal elections. It's certainly not difficult to implement.

These ballots are also, obviously, very readily hand-counted. In any close race, I would expect the machine count to be checked by hand counts.

Still the potential for abuse

[corebreech]

It's easy to envision an instance where an individual dedicated to corrupting the vote stations himself within the voting station, observing the voters as they leave the booth and deposit their receipt.

Those voters that don't bother examining their receipts can be easily discerned, and the voting machine could conceivably be instructed remotely to change that voter's vote.

This is a very good step being taken by California, but I think they need to go one step further and mandate a recount for every election, regardless of whether it is seen to have irregularities or not.

Re:Still the potential for abuse

[NoOneInParticular]

Or you could dig a tunnel under the vote station and use a saw to make a hole under the box where all the paper votes are kept. Then when a paper gets fed to the box, you will take it and replace it with another vote of your liking. Don't forgot to wear a tinfoil hat during the operation.

Re:Still the potential for abuse

[corebreech]

It is funny. I mean, we all know that nobody would ever go to such lengths to obtain power.

Nobody would ever do that.

Not once in the history of this nation, or of the world for that matter, has anyone ever conspired to steal power.

All those stories you hear about fraud and elections in the past, they're all made up. Nobody ever stuffed ballot boxes. Nobody ever denied the right to vote to minorities. Nobody ever bought votes, or intimidated voters on their way to the booth, or bribed election officials, or tampered with ballots.

I mean, that's just crazy, right? It's all a great big conspiracy theory.

LOL

Better a tinfoil hat than a blindfold made of iron.

Re:Still the potential for abuse

[vidarh]

That just plain doesn't make sense, since you won't know whether those voters looked at the receipt before leaving the booth. I certainly would, as I wouldn't like people to be able to look at the receipt from I left the booth until I deposited it.

Re:Still the potential for abuse

[bussdriver]

Not funny, you discredit the many possible situations by creating an over-the-top one.

Just think if you worked for a company, and a billion dollars (literally) was being spent on elections; and you wanted some of that money.... You'd find ways to do stuff...

Have "bugs" in the software, where 50% of the people do not verify and re-vote to correct the "bug". This "bug" I personally KNOW about. took 10 reboots of the system before it let the vote correctly register....well it was put onto flashRAM, so I still don't know...

Printing 2 times, 1 for the voter goes to the trash, and 1 for the ballot box. Must be a dozen ways of just doing that trick.

Re:Still the potential for abuse

[corebreech]

Ah, it depends on the meaning of the word booth, doesn't it?

If we're talking about the classic style, where you have a curtain and all that, then yes, I'd agree with you.

But isn't likely that we'll migrate to booths similiar to those used by banks for their ATM's? Just enough privacy to ensure that nobody can peek and see what your passcode is, nevertheless you are still in full view the entire time.

Re:Still the potential for abuse

[NoOneInParticular]

My response was mainly there to point at the far-fetchedness of your particular conspiracy theory. Obviously there are many ways to tamper with any election scheme, only yours was no more likely than mine. Another response pinpoints the cause: what if people check the receipts in the ballot box?

As to making sure that the counts between machine and receipts aren't different, this could easily be done by performing random checks, prefereably with random machines/receipt combinations being selected *after* the election. You could probably get away with checking no more than 1% of the machines/vote boxes, to get an indication of the honesty of the elections. This should of course be done by an independent body consisting of all interested parties. If discrepencies are found, start recalculating all and find the fraudster.

Unfortunately this does not seem to be the plan in California. Given the shoddy history of elections in the US, it's time to make some real advances in voting. Maybe let the UN in to check procedures, like they do with other countries that have a poor track record of honest elections.

Couldn't voters insist on using the old machines?

[corebreech]

I know it would probably create long lines at the polls, but I for one would be more than happy to wait an hour or more if I could know that my vote wasn't being rewritten by some unseen entity.

Re:2005? 2006?

[leerpm]

You have to give the counties an appropriate amount of time to purchase voting machines that work this way. Not all of them have money falling out of their pockets that they can spend on brand-new voting machines (again), if they happened to recently purchase some machines without these features. Granted, those counties probably should not have purchased such machines, but if you force this on them too soon, you will get a backlash because the counties will have to pull the money from other parts of their budget.. AND that would piss voters off.

how is this better than paper voting?

[martin-boundary]

From a procedural standpoint, if people are required to take the receipt and bring it to an official stating "hey, I didn't vote for that guy", then anonymity is effectively lost. How many people are going to think twice about complaining in that case?

Voter: Sheriff, I just voted with that machine over there, and it said I voted for Bubba Smith.

Sheriff: Yeah, what's the problem? Don't like my cousin?

Voter: Uh, no everythin's fine. Forget it.

Re:how is this better than paper voting?

[argent]

This *is* paper voting. The question isn't "how is this better than paper voting", it's "how is this better than electronic voting". And of course it's better because it provides a physical ballot. If the paper ballot doesn't say what the voter expects, then it's a spoiled ballot, it must be discarded. When he tries again, if it still votes for Bubba Smith, then he calls the TV station and they break a big story about Sherrif Bohunkus and Bubba Smith and vote fraud.

With electronic votes, nobody would ever know.

Re:how is this better than paper voting?

[martin-boundary]

Fucking AC, you're the one who should read the story. Who said anything about voters taking receipts home? It's about some technical oversight guy looking you in the eye and asking "did you vote for xxx?". It's about intimidation. You've got two choices if there's a problem: either you complain that the machine is wrong and you didn't vote that way (and if the machine is rigged, those people have your name and address now), or you keep your mouth shut and you've just voted for somebody you didn't want to vote for.

Re:how is this better than paper voting?

[martin-boundary]

This isn't quite paper voting.

With traditional paper voting, you keep the piece of paper in your hand until it's in the box. The only visual verification is that somebody saw you put a piece of paper in the box. Any piece of paper, it doesn't matter. When the votes are counted later, if your vote is disqualified, then no-one knows you did it.

With this system, the votes are printed and visible to you. If you're going to complain that the machine stuffed up, you have to tell someone. This person will ask you who you voted for, and will want to verify that the printout contains another candidate's name. Once they've verified that your candidate and the the one on paper are different, some action will be taken to fix the machine. But by then, the official will know how you intended to vote. Your vote is no longer anonymous.

Re:how is this better than paper voting?

[John Hasler]

> With this system, the votes are printed and
> visible to you. If you're going to complain that
> the machine stuffed up, you have to tell someone.

The machine could have a REJECT button which would drop the spoiled ballot into a shredder and let you try again.

Re:how is this better than paper voting?

[argent]

There is no reason a spoiled ballot from a computer should be treated any differently than a spoiled ballot from a punchcard machine, or you accidentally messing up an optical scan ballot.

This isn't to say that the election officials or the voting machine manufacturers won't try and get ANOTHER broken design past the community, but if they require anyone to reveal how they voted that's because they screwed up... not because there's anything inherently wrong with the model.

Re:how is this better than paper voting?

[TRACK-YOUR-POSITION]

This is intriguing, but I think in the end this is actually more secure though possibly slightly less anonymous than paper voting.

For a conspiracy to do what you describe, it would need to have cooperation between a major corporation that produced or maintained the machine, and corrupt local officials to intimidate voters. The local officials acting alone would not be enough, because if the machines are acting properly, the number of misvotes shouldn't be large enough to completely control the outcome of the election. Because even if what you say is true--that correcting a misvote would require losing anonymity--I don't think that alone is enough to conduct an intimidation campaign.

Essentially, you have an almost-paper system acting alongside a purely-electronic system--both systems have to be corrupted in order to corrupt the final outcome. That is the brilliance of the new system.

Re:2005? 2006?

[Anonymous Coward]

What did they use before these trojaned voting machines? Did those all disappear? Canada seems to have elections without using machines at all and they do okay.

And it needs to be ...

[burgburgburg]

a National Standard

This is about all of the electronic voting machines (even though Diebold is most suspect) and it's about the whole country.

Re:And it needs to be ...

[wart]

Unfortunately, election rules, including voting machine standards, are governed by the state, not the federal government. States are free to follow California's lead or not and the feds can't do crap about it.

Re:And it needs to be ...

[BurritoWarrior]

No, a national standard for voting is a bad idea. It would allow a national exploit as well.

It is much better, if more expensive, to allow counties to implement the voting system they see fit.

Re:And it needs to be ...

[FreekyGeek]

Anything required by California is almost de facto a national standard. It probably isn't worth it for voting-machine manufacturers to make two different models, one for sale to CA and one for sale to other states.

You see this in lost of industries: low- and zero-emission vehicles are available nationwide primarily because CA required them. And that's why the banking lobby fought so hard against privacy regulations in CA: because if they had to redo their IT systems for CA, then basically it becomes available to their customers in all states. Cheaper to do it for eveyone than just people in one state.

Re:And it needs to be ...

[ibbey]

No, a national standard for voting is a bad idea. It would allow a national exploit as well.

It depends on what the national standard describes... I don't think that the parent poster was suggesting that we need a national standard for the actual voting software, only a standard saying that a VVPAT is required. How that paper trail is implemented, and the software used, does not necessarily need to be standardized.

Re:And it needs to be ...

[anthony_dipierro]

Unfortunately, election rules, including voting machine standards, are governed by the state, not the federal government.

The national government can and does regulate voting standards. In fact, the 14th, 15th, 19th, and 26th amendments, as well as article IV, Section 4, require it to.

Re:And it needs to be ...

[nyseal]

That's why CA sucks. I, for one, do not want CA setting my standard of living....especially in the political arena. They may seem 'cool' but it's a farce. Their 'feel good' attitude is all but destroying this country, especially when a person visits Los Angeles, which isn't so nice. The last time I was there I felt like I needed a shower just walking through their airport. Personally, I wish CA would just fall into the ocean. Mod me down as a troll but it still wouldn't change my opinion.

Amen

[djupedal]

...as goes California, so goes the nation. Smog laws; consumer protection laws, etc. Not always, but usually. Too bad CA can't stop shooting itself in the foot when it comes to business and health care.

A paper trail is just a sanity check, and a completely reasonable way of keeping things in line.

At what point

[mindstrm]

did "republic" and "democracy" become mutually exclusive?

http://www.google.com/search?q=define:republic&i e= UTF-8&oe=UTF-8

The US is a republic. Eire is a republic. The soviet union was a bunch of republics. China is a republic.

Re:At what point

[Daniel Dvorkin]

Yes, exactly.

To elaborate: in the US and in many other countries, republicanism is the mechanism of democracy. We are both (in theory) a republic and a democracy -- a republican democracy, or a democratic republic (that latter term, unfortunately, having been recently hijacked by some very undemocratic republics.) Anyone who says "the US is a republic, not a democracy" and thinks it proves something is an idiot.

Examples of undemocratic republics: USSR, China, Cuba, Iraq (yes, still), Iran, North Korea, etc. The list of unrepublican democracies is shorter, but includes countries such as the UK, Japan, and Sweden. If the grandparent poster and those like him had to give up one -- republicanism or democracy -- I wonder which he would choose?

Re:At what point

[smagruder]

Anyone who says "the US is a republic, not a democracy" and thinks it proves something is an idiot.

Correct. It's those who think they "learned" that in their high school civics class who are the problem. The Constitution clearly lays out a republic, but with democratic mechanisms. And several of the Amendments extended the democratic aspects of those mechanisms. And on top of that, roughly half the states have implemented citizen-sponsored ballot initiatives, further expanding democracy. You can even add the recent ability for ordinary citizens to comment on federal regulations as a recent democratic expansion, amongst other e-democracy initiatives.

Democracy has yet to stop expanding in the USA.

Re:At what point

[benzapp]

Democracy = government rule by majority vote of the people

republic = government rule by a select group, can be elected by the people or not.

I agree in principle with what you are saying, but most people who make a point of saying the US is not a democracy are basing on that majority rule concept. A lot of people think that if the majority wants it, it should be law... whereas there is no such principle in the country and ample evidence that is what the founding fathers were trying to avoid.

As far as unrepublican democracies, how is the UK a democracy? All legislative power is in the house of commons which is elected by the people. The people have no power on their own. Further, the queen can become a de facto dictator if the need exists.

I don't really know anything about Sweden or Japan. With such a small population, I wouldn't be surprised if Sweden could pull it off. Japan however... I dunno.

Re:2005? 2006?

[jackb_guppy]

So those counties do not believe the voter need to know their vote counts, what is bases of county that has required to buy the RIGHT equipment, in the first place.

Electron Day 2004:

Yes, we use paperless machines here still, this is saving you money...

Wait! Stop! Please, do not pay attention to man behind the curtain, he is the repairman... ...and the new next president of United States!"

Iowa has the best voting system.

[acoustix]

In Iowa to vote you go inside your own booth will nothing but a pencil and a scantron sheet (like the ones you fill out on a standardized test). Fill in the circle and you're done.

Of course, the circle has to be completely filled in. But the again, if you can't fill in a circle then you probably shouldn't be voting.

Counting the votes is relatively fast. We usually know within 2 hours of the polls closing who has won.

Why do we even NEED an electronic system? What is wrong with the paper ballots?

-Nick

Re:Iowa has the best voting system.

[Politburo]

But the again, if you can't fill in a circle then you probably shouldn't be voting.

What if I had my arms blown off in nam?

Re:Iowa has the best voting system.

[barawn]

Because a scantron system can screw up, and destroy all of the ballots, so paper is dangerous.

Plus paper is expensive. Plus counting is only fast if you have the people (or the machines, which are dangerous) to do it.

Plus scantrons are ambiguous. There's a recognition issue there, and while they're pretty good, the margin of error is nonzero (as it is with all counting systems, but here it's measurably non-zero). And then you'd get into "pregnant chad" lands again, just with, I dunno, "pregnant bubbles".

Look, the paper trail isn't the important part. The important part is that a hardcoded audit trail is available, and that it can be easily spot checked to ensure that the machines are working as they are supposed to be working.

Electronic voting is the right way to go, in the future. As you scale the number of people, the logistics get insane, and wasting money on elections is not what I want a government to be doing. We're talking about *counting* here, something that's been done since the first person looked at his fingers.

What you need, though, is a foolproof system. A system without friggin' software, a system running on bare metal, just logic gates, writing to a verifiably safe write-once-read-many storage medium.

Unfortunately, in order to develop that, you need to have some technical expertise, which Diebold and co definitely don't have. Come on. Commercial off-the-shelf crap? Jeez. Take out a damned electronics CAD package and design something that doesn't suck.

Re:Iowa has the best voting system.

[Zaak]

Look, the paper trail isn't the important part.

I disagree. The thing about paper is that people can read it as well as machines can. There's no way for a voter to trust a voting machine unless he can read the record it's making.

TTFN

Re:Iowa has the best voting system.

[barawn]

Only if he can read.

What you're assuming is that everyone can read - that is, education is now at a level that "reading" is a default. (This isn't really true, especially in some of the lesser educated areas, where you really do want to make sure voting is safe).

I'm not saying that assuming people can read is too much. What I'm saying is that trust needs to start somewhere. How do you know the paper isn't printed with a special ink that disappears after a few hours, and another name reappears?

It's not the record that's important. It's the auditability: an outside source can come in, and read the vote record, somehow. If you have to give them the format that the vote is in, fine. That's not too much to ask. They can then check that the two totals match each other.

Could someone still defraud the election? Of course. But someone could do that with paper ballots, too - you just replace the paper ballots after the voting is done. Obviously, there are safeguards to prevent that. In this case, you prevent defrauding an electronic voting system by publicly testing the same machines before the election starts.

Let's add to that..

[mindstrm]

Let's check the CIA world factbook...
US:
Government type:
Constitution-based federal republic; strong democratic tradition

Re:Let's add to that..

[karstux]

Yeah, and of course the CIA, being the honest and independent source of information that it is, would tell you that this "democratic tradition" is a bit shaky right now...

Difficult for other states to not follow suit?

[Shoten]

That's ridiculous. It'll be easy for other states to not follow suit; what will be difficult will be for the companies who make these machines to avoid producing them with this as an option. This, as a result, will make it easier for states to follow California's example, if they are so inclined. But sticking to the status quo of electonic voting has not become more difficult yet.

and it's not enough

[twitter]

The software needs to be open if not free. Paper reciepts are a great first step, but the system can still be manipulated within paper counting accuracy. For the sytem to be an improvement, it should do better and it can. Open software can be verified for the soundness of it's methods and rigourously tested by interested parties. The results of that kind of testing would be secure and accurate voting that's really better than paper.

Closed source junk, on the other hand, is imposible to test and verify. Windoze based machines are a great example of how bad it can get. No one knows what goes on in DLL hell, no amount of testing can find all the flaws and backdoors and the inability to copy the software makes third party tesing limited if not imposible.

Still, California is to be commended for taking this step. Closed source machines with a paper trail can only be manipulated so far. It's unlikely that upsets there will be fradulent. I can forgive them for needing a year to get it set up and working. They might be bright enough to not use the mystery vote machines before they are fixed with a paper trail. Let's hope they and Dibold take the next step and use software that people can realy trust.

Re:and it's not enough

[matchlight]

Closed source junk, on the other hand, is imposible to test and verify.

I have a hard time accepting this often used stance against the choice of closed vs. open source in voting. Whether or not the public has access to the source does not necessarily assure the product is:
a. bug free, since most source, open or closed has bugs. That's life.
b. is tamper proof, given that the code is eventually compiled and there's nothing keeping a person from altering the code before compile.

And although I am a proponent for BSD or Linux vs. any WinX OS, bad code is bad code regardless of the OS it resides.

I personally think there is a better way than using open sourced code for electronic voting systems. I believe that the code should be restricted but reviewed by partisan and non-partisan groups to ensure fairness and robustness because I don't see the value of letting just anyone see the code. For those few bright people who would be intelligent enough and industrious enough, there should be an application process to become a member of the review team. This allows accountability and also security in the event that something illegal or incorrect happens but also keeps it difficult for those who would try to illegally manipulate the system.

But if I had to choose, I would rather an open source system used that a completely closed unreviewable system.

Re:and it's not enough

[anthony_dipierro]

I have a hard time accepting this often used stance against the choice of closed vs. open source in voting.

I have a better reason. It's a waste of money for the government not to insist on an open source solution. Voting systems are going to be around for a long time. We shouldn't tie ourselves to a single company. Doing so is wasting the taxpayer's money.

Re:2005? 2006?

[Almost-Retired]

but if you force this on them too soon, you will get a backlash because the counties will have to pull the money from other parts of their budget.. AND that would piss voters off.

I'm sorry for those counties that have to do that. But since they made the decision to buy those POS in the first place, then in retrospect, I don't feel sorry for them.

Leaving a situation in place that allows even one election to be stolen by the likes of Diebold et all, is not a situation any voter should tolerate, ever. And if it means you don't get your street plowed till the next day after a snowstorm, so beit. The place to point your anger at in that case is the county commissioner who saved a buck by buying it from so-and-so without setting out a set of specifications that precluded such hanky-panky.

And, while I'm a firm believer in the one man, one vote rule, I'm a bit ambivalent about the isolation of the voters ident from the vote. Personally, I'd like to have the ability to go back after the election, and verify that the candidates I voted for did indeed get my actual vote.

To that end, a 2 kilobit pgp(or gpg) style public key for you as a receipt and a private key attached to the vote seems like a hell of a good idea. I could check my vote with my public key, but to check somebody elses would take a considerable computational effort.

Now all we have to do is convince TPTB that democracy, to be protected, must be well protected even from their prying eyes.

Also, does anyone know why the preview to post isn't working about 50% of the time?, it draws the column headings on the left at full screen width, and says its done. Like this is the second time in a row I've got to post blind...

Cheers, Gene

Your system might already be electronic.

[twitter]

Of course, the circle has to be completely filled in. But the again, if you can't fill in a circle then you probably shouldn't be voting. Counting the votes is relatively fast. We usually know within 2 hours of the polls closing who has won. Why do we even NEED an electronic system?

Sounds like a scantron system to me. A machine is counting your votes already. A machine might also be adding the county results up for you too.

Re:Your system might already be electronic.

[John Hasler]

Nothing wrong with that, as long as the paper ballots are there to be recounted if necessary.

there goes anonymity

[TheSHAD0W]

The problem with this little scheme is that the printer generates a linear log of votes, and this might be used to figure out who voted for whom. There goes your anonymity. People might be afraid of retribution for voting the wrong way.

I recommend using blinded signature techniques to solve the problem. "Poll watchers" will network their computers to the voting machine, and when someone votes, their machines will sign the voter's choices through a blinding mechanism that will validate the vote. The vote will then be released to the poll watchers' machines mixed with "chaff".

The chaff would be generated prior to the vote; a large number of votes would be created, tabulated and signed blindly. Each vote broadcast on the network would be mixed with ten or so randomly chosen chaff votes. At the end of voting, the unused chaff votes would be tabulated again, the number of chaff votes cast would be calculated and subtracted from the total, giving the true number of votes cast.

Re:there goes anonymity

[squarooticus]

What is to stop the voting machine from modifying the ballot choices right as the voter casts them? Your system provides no integrity guarantees at all: we need a system in which the voter actually looks at a piece of paper with his choices and validates by eyeball that the votes are correct, and in which this very same piece of paper that the voter validated is used in random manual recounts. Nothing else provides any confidence in integrity.

Re:there goes anonymity

[Have Blue]

Yes, why don't we replaced a centralized, self-contained system with one that involves thousands of anonymous Internet users. That's much more secure and accountable.

Re:there goes anonymity

[TheSHAD0W]

And what stops our current voting machines from modifying the ballot choices? Have you ever inspected the mechanism?

Get a plain vanilla computer with no hard drive, boot it off a CD containing open-source voting software. Worried the CD might be tainted? Make up a dozen of 'em, have one chosen randomly and let the poll watchers validate the others contained the correct binaries. This'd be a lot more easily verifiable than anything we have now.

Re:there goes anonymity

[Ancil]

Refresh my memory: what was wrong with just giving people a piece of pasteboard, a sharpie pen, and some privacy? That's how I voted when I lived in California, and it seemed to work well enough. Feed the ballots through a vote-validator, and you will even eliminate spoiled ballots.

You want poll watchers to show up with laptops and sort "chaff" off a 100baseT network? WTF is chaff? What if someone's cheating? 99% of the electorate won't even understand how you know they're cheating.

A box which is shown to be empty when polling starts, into which ballots are placed -- that's easy to understand and almost impossible to tinker with.

These quantum-encrypted, fusion-powered voting machines are accomplishing what, exactly?

Re:there goes anonymity

[John Hasler]

Clever, and perhaps even provably secure.

Unfortunately, such schemes cannot even be explained (let alone proven) to the average poll watcher, and she should not trust a system she cannot understand.

Make a touch screen machine that prints individual
paper scantron ballots. The rest of the system can remain unchanged.

Re:there goes anonymity

[Politburo]

No, if the system is correctly implemented, the reciepts will go into a box (separately), not be kept as a long journal reciept.

Re:there goes anonymity

[Ancil]

Thanks, but it was a rhetorical question. "WTF is chaff" is the response of the entire U.S. electorate, upon hearing that some geek somewhere thinks a voting machine might be flawed. "Does chaff have something to do with hanging chads??"

Back to my original point: voting machines are a non-solution to a problem that never existed. The post I responded to seemed to think that requiring poll watchers to show up with networked notebook computers running god-knows-what software, would solve all these problems.

See Rivest's paper on chaffing and winnowing

How about an alternative: use a vote-counting system which doesn't actually require a PhD to understand. My cleaning lady doesn't understand information theory or steganography, but she would still like some assurance that her vote is being counted. A locked box with paper ballots works better, and no one needs it explained to them.

I agree, but the Constitution stops it...

[MyNameIsFred]

I think paper ballots probably are the best. The process is more transparent. Although fraud can be committed with paper ballots.

...Everyone gets a ballot with a standard design, from Victoria to Halifax...

However, there are some differences between the American and Canadian electoral systems. Please remember, the US Constitution explicitedly puts the responsibility for conducting elections in the hands of the states, for example Section 4, Clause 1 on the election of Senators and Representatives. Furthermore, as witnessed in the last election, we use an Electoral College to pick the President. The selection of the Electoral College members is decided by the individual states. So the Federal government cannot mandate a uniform ballot. (Your statement also ignores the fact that most, if not all, localities use the national elections as opportunities to decide local issues that require some customization of the ballot.)

To do what you propose, while it has merit, would require a Constitutional amendment. One that is not likely to be passed because the states would have to give up some of their power.

Re:I agree, but the Constitution stops it...

[finkployd]

While true, nothing is stopping the states from (at least trying to) getting together and deciding on a standard format. Even if some things are added on at the end (referendums and such) template for voting should be consistent.

Finkployd

Re:I agree, but the Constitution stops it...

[will_die]

"I think paper ballots probably are the best. The process is more transparent. Although fraud can be committed with paper ballots."

All this talk about electronic ballots came about because of paper ballots.
Remember it was the electon in florida where in a handfull of conties the people who were in charge of thoses areas were to lazy or incapable of empting so trays where chads from a paper ballot went into.

What's the difference?

[Ulven]

So what's the difference between reading a candidate's name on paper, and reading it on the screen?

Or do the machines speak names out loud?

Re:What's the difference?

[Jeremi]

So what's the difference between reading a candidate's name on paper, and reading it on the screen?

Not much. But there is a big difference when it comes to writing on the paper... you can do "illegal" things on the paper that would invalidate your ballot. A computer can ensure that your ballot will be valid.

Or do the machines speak names out loud?

Yes. Many of them can, if the person isn't able to read the text (for whatever reason).

1 to 150?

[Espen]

With one machine for every 150 voters you've got to wonder what the point of machine voting is.

Re:1 to 150?

[Politburo]

Voters or people of Voting Age? There is a huge difference, as turnout is ~35-50% currently.

*sigh*

[Guppy06]

Yes, I didn't RTFA.

Re:Repealed in 2005

[Wardish]

Nutz!

Unclench your sweaty hand from the crack pipe, deposit it carefully on the table in front of you and then slowly back away.

Re:The real question is...

[bitflip]

So, just make it a checkbox on the receipt: "This does not match what I voted"

If you get a few of those, there's probably not much too worry about. If you get a lot of those, then you do. It doesn't give any insight into what is wrong, but all that's needed is an indication that _something_ is wrong - then count the paper backups.

Very interesting...

[dentar]

that the deadline is 7 months AFTER the 2004 election.. Aaaahnold!!!

Check out VoteHere

[ca1v1n]

If you want to see a really clever electronic voting system, check out VoteHere. They use paper receipts that basically records a hash of your vote, so your receipt cannot prove to anyone who was not looking over your shoulder when you cast the ballot what that vote was, but still allows you to prove that your vote has/has not been changed after the polls close. As VoteHere points out, authoritative paper receipts really just turn the machine into a very expensive pencil, when they offer the potential to do so much more.

By the way, I have no ties to VoteHere, I've just been studying electronic voting a lot lately.

For more info, see http://www.verifiedvoting.org/

Of course, this system has weaknesses, as will any system which enforces both authenticity and anonymity, but even if it cannot be protected against all attacks, it at least lets you know when an attack is happening, which is a huge step up from most paper and even electronic systems.

Re:Check out VoteHere

[John Hasler]

> As VoteHere points out, authoritative paper
> receipts really just turn the machine into a very
> expensive pencil, when they offer the potential to
> do so much more.

A pencil, expensive or otherwise, is all that is needed.

> it at least lets you know when an attack is
> happening, which is a huge step up from most
> paper and even electronic systems.

All that you need to do to detect an attack on a paper system is stand there and watch it. Molly Pollwatcher can do that. With computerized systems she has to trust the experts.

Re:Check out VoteHere

[TRACK-YOUR-POSITION]

As VoteHere points out, authoritative paper receipts really just turn the machine into a very expensive pencil, when they offer the potential to do so much more.

No, it's better than just a pencil, because pure paper voting is more vulnerable to small-scale election fraud--i.e. just replacing the box of ballots when no is looking.

So with paper we have to trust the pollworkers, with machines we have to trust corporations. With both machines and paper, we just have to trust that pollworkers and corporations are not in a giant partisan conspiracy.

Re:Check out VoteHere

[ca1v1n]

The key is *authoritative* paper receipts. If, as most of these plans assume, the paper ballot is trusted over the electronic ballot in case anything is ever called into question, then all you have to do to force fallback to the paper count is demand a recount. That brings us right back to centuries-old vote fraud schemes we've been suffering from for ages, and also obviates the very expensive and complicated computer equipment.

Re:Check out VoteHere

[ca1v1n]

There are all sorts of attacks that Molly Pollwatcher will never catch. You also have to make sure that the ballot doesn't get changed/lost after submission. VoteHere provides exactly that capability. If your vote isn't present or is different in the counting set after the polls close, you can prove it. Paper doesn't offer that capability, because it would violate anonymity. Thanks to cryptographic techniques, we can ensure integrity without breaking the secrecy of your vote. Since any system that requires voters to shuffle ElGamal pairs by hand is obviously overcomplicated, a computer is the only way we can accomplish this.

Re:Check out VoteHere

[John Hasler]

> There are all sorts of attacks that Molly
> Pollwatcher will never catch.

Name some (hint: she and her predecessors have been monitoring elections for centuries).

> You also have to make sure that the ballot doesn't
> get changed/lost after submission.

You figure the poll watchers all go home as soon as the polls close?

Re:Check out VoteHere

[TRACK-YOUR-POSITION]

Theoretically true, but in the real world if there is a large difference the machines and the paper, then we know that something mischievious happened. So we might declare the paper ballot winner the true winner--but now we also know to investigate both the manufactuer of the machine and the pollworkers at that particular site until we manage to account for the difference.

People needs to vote, and look at the receipt

[camperslo]

Some incentive would be helpful, both for getting out the vote, and for getting people to double-check the receipt. Perhaps if they had to read it to get a code allowing them to get a discount on auto registration? Maybe pay the old rate, instead of the triple one that was temporarily in effect but overturned by Arnold. This would bring out more people, Arnold could still say he'd axed the tax, and the state would get some added revenue from those that didn't vote.

I'm disturbed that changes, adequate or not, won't take effect well before the November 2004 election. Perhaps we should all be urging people to bypass the first-level machines by voting absentee?
If only the spam people got suggested that instead of trying to sell a bigger penis...

NETWORK, not internet.

[TheSHAD0W]

I never suggested voting over the internet; this would be a set of networked machines, with no attachment to the general net.

The Constitution doesn't stop much of anything

[jjo]

The modern US constitution has really precious little protection against the federal government usurping powers that nominally belong to the states.

While the several states are indeed responsible for conducting federal elections. Congress could essentially mandate a uniform ballot standard by appropriating substantial aid to support the states' election functions, but only for those states that adhered to the federal standard. In that way, each state could theoretically conduct elections in its own way, but as a practical matter every state would have to implement the federal standard. (This is exactly the method used to compel the states to pass traffic laws that conform to what Congress wants.)

Re:The Constitution doesn't stop much of anything

[Jameth]

Yet somehow Louisiana still has a drinking age of 18 and several states have no speed limits.

Re:The Constitution doesn't stop much of anything

[Peaceful_Patriot]

Maybe this is due to a little thing called the 10th Ammendment to the Constitution:

Amendment X
The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.

Re:The Constitution doesn't stop much of anything

[jjo]

That's all wrong. Every state has now passed a speed limit. Also, the official drinking age in Lousiana is 21. (The de-facto drinking age may be another matter.)

It's certain that states like Wyoming wouldn't have imposed a speed limit, and Lousiana wouldn't have raised the drinking age, without heavy pressure from Congress.

Re:The Constitution doesn't stop much of anything

[Jameth]

I was not aware Louisiana had finally raised their drinking age. Likewise, I thought Montana still had no speed limit. Sorry for the mistake.

Re:The Constitution doesn't stop much of anything

[Kenneth]

I was not aware Louisiana had finally raised their drinking age. Likewise, I thought Montana still had no speed limit. Sorry for the mistake.


When the federal government allowed states to set their own speed limits on interstate highways, Montana quickly removed the speed limit. The problem was that suddenly the accident statistics jumpped by a great deal. Germany can get away with no speed limit because they have very strict training, and auto maintainece standards, as well as laws that make it possible to drive very fast.

For example, in Germany, you MUST drive in the right lane, and are only allowed into the left lane to pass. If you happen to be in the left lane, or not in the far right lane, and someone comes up behind you, you MUST move over, which means you MUST be watching what's behind you.

In the U.S. however, Drivers training is about 6 weeks, you can drive pretty much anything, there is little to no lane dicipline, and I actually know people that insist that the rear view mirror is a BAD thing.

Anyway, the result was that Montana quickly rose to #1 on the highway statistics, and they basically had to put the speed limit back.

Ironically enough, according to some people the real reason was that Europeans, particularly Germans came to Montana so they could go flat out for an extended period of time, something that increasing population has made impossible, even on the autobann.

In fact, speed limits are posted near population centers, and even where there is no legal speed limit, there is often a practical speed limit, because of the traffic.

Please don't use thermal paper

[eples]

It is important to keep a written record of elections - printing out the ballots is a good idea. But please - PLEASE - do not use thermal paper.

The print fades, it is not good as a lasting record.

Re:Please don't use thermal paper

[Politburo]

It really doesn't need to be a lasting record. Most election laws say that the election cannot be challenged 14-30 days after the election.

Finally!!!

[mod_parent_down]

...A governor who understands technology!

Re:Finally!!!

[SmashPDX]

He promised an Action Oriented State. By printing the voting machines will have more Action and therefore he supports them. It is much more interesting than an Action-Less machine which doesn't have moving thingies to show the People of California all their Action.

Sorry couldn't resist. :)

Still Room for Fraud

[Prince Vegeta SSJ4]

So there is a paper reciept, who says whats in the computer AND on paper isn't an error. How do you know that the machine didn't generate a false vote with a corresponding false reciept?

Re:Still Room for Fraud

[phr2]

I wish they didn't say "receipt". A receipt is a piece of paper that memorializes a done deal, and that you take home with you. The thing these machines are required to print out is a ballot. It gets dropped in a ballot box and is the authoritative record of how you voted, if the electronic count is suspect in some way.

How do you know it says the right thing? Well, uh, you look at it before you drop it in the ballot box. That's why it's called a "voter-verifiable" paper audit trail. If Alice is running against Bob and you want to vote for her, the machine gives you a piece of paper and you make sure it says "I vote for Alice" and not one that says "I vote for Bob".

Re:Still Room for Fraud

[Simple-Simmian]

The paper record is voter verified. But if you had read any article on the Cal Sec of State mandate for paper records you would have noticed it.
To many people who haven't RTFA commenting.

It would be easier and cheaper to go to a scantron type postal ballot and be done with these expensive and unproven machines. At least you can do a visual recount with them with ease. Perhaps more people would vote if they didn't have to leave home.

Paper Ballots are best.

Re:Couldn't voters insist on using the old machine

[koreth]

Vote absentee. Then you know there's a paper record, since you punched the card yourself. (Plus you have all the time you need to check for hanging chad.)

I've voted in every election in the last fifteen years and have yet to wait in line at a polling place.

You might have a point, if that's how you did it..

[raehl]

But that's not how you do it.

You have a set of voting machines that print ballots voters can read. THESE MACHINES DO NOT COUNT VOTES - they just print ballots. THEY ALSO DO NOT HAVE PRINTER ROLLS - like most home printers, you stick a blank ballot in the top, it prints on that ballot and it comes out the bottom.

If the machine doesn't print out the ballot you want, you just go back to the election official, deposit the ballot you don't like in a shredder or box of "discarded" ballots and get a new one, then go back to the machine and have it printed.

Then, once you have a ballot that is correct, you deposit in the ballot box for counting by another machine later, or perhaps just stick it in another counting machine then which has a box for keeping the ballots it counts.

Nobody sees your vote, no hanging chads, no "out of paper" problems.

There are no electronic votes.

[raehl]

1) Go to table - prove you are registered to vote, receive ballot.

2) Place ballot on input of voting machine.

3) Vote.

4) Machine prints your ballot, but DOES NOT COUNT YOUR VOTE.

5) Inspect ballot. If correct, deposit in ballot box for machine counting later, or feed into separate machine on-site which is attached to a ballot box which keeps your ballot. If incorrect, return to election official, who destroys ballot and gives you another one, or voids your ballot, deposits in a voided-ballot box, and gives you another one, go back to voting machine and repeat.

6) In the event of contested election results, take ballots and recount them using means of your choice.

Re:There are no electronic votes.

[frdmfghtr]

This sounds exceptionally familiar to the voting process I have used in the past, except that I filled in the ballot using a black marker for reading by an optical scanner. You replaced the magic marker with the electronic marker.

1) Go to table - prove you are registered to vote, receive ballot.

Is this a completely blank ballot or is it pre-printed with the different political races and candidates?

2) Place ballot on input of voting machine.

How do you ensure against a misfeed? If the ballot is pre-printed, a misaligned ballot could put marks in the wrong places.

I like the idea of the machine marking the ballot; a standard means of printing and marking the ballots would help the electronic counting process.

In fact, this brings about an interestig means of cross-checking the results. The ballot is printed with the votes with multiple machine-readable forms: marks in particular locations to indicate choices in a human-readable format, like the Scan-Tron sheets of old where you fill in the bubbles with a #2 pencil, plus a machine-readable barcode.

When the ballot is counted, BOTH types of marks are read. When the ballot is read, both types or marking must match. If there is a discrepancy, then the ballot is not electronically counted and counted by hand.

Precinct audits can be done randomly or when there is a certain percentage of discrepant ballots.

It says something when with all the technology available to us in out so-called "high tech" society, we can't accurately count pieces of paper without lawsuits and litigation taking center stage.

Yes, but also be trivial to detect on audit...

[raehl]

When you feed in 100 ballots, 50 of which say 'Candidate A' in the margin, and 50 of which say 'Candidate B', and the results come out 60-40.

Re:2005? 2006?

[Peaceful_Patriot]

Yeah, it's not soon enough and not perfect yet. But I am truly heartened that the "vocal contingent that opposed a paper trail" did not prevail over the "minority of computer and voting experts who supported the requirement."

This puts voting machine manufacturers on notice that we know there are flaws in the system and we will be watching closely.

Hurray for California Secretary of State Kevin Shelley, who obviously was paying attention to the facts over company PR.

Re:REcount will result in errors

[bussdriver]

I'm sure they already know the % that will correctly verify the vote. And the % that even checks it.
In fact, there have been studies were people were asked to recheck their answers, and you'd be surprised how the human mind messes with you...The error rate on that was high enough to get a few points, let alone those who do not check at all.

How big is the print on the paper? Is it in english? Its it black on white? or grey on offwhite? Does it list the whole ballot or just those you voted for?
All of these questions will greatly effect the accuracy of verification.

ALSO, I will bet my new G5 on the fact no recount will completely match the computer total.

Lots of things.

[raehl]

What if you can't read small print?

What if you have arthritis?

What if you're allergic to pencils? (Ok, ok...)

It's a matter of degrees. If electronic voting is done right, you increase the accuracy of elections.

Why is this market insightful?

[raehl]

-1 Paranoid or -1 Bad Solution would be more appropriate.

There's already a linear log of votes - votes at the bottom of the ballot box were turned in first. And that doesn't change - THE VOTING MACHINE DOES NOT COUNT VOTES! It just produces paper ballots with greater accuracy than previous methods. That's it. It's the paper ballots that count.

No, idiot, no potential...

[raehl]

Because it's the RECEIPTS that count, NOT THE VOTING MACHINE!

If the votes in teh machien don't match the votes on the receipts at the end of the day, you scrap what the machine says and go with the receipts. That's the WHOLE POINT!

Re:No, idiot, no potential...

[corebreech]

So tell me, Mr. Oh-so-intelligent-person, how do you know that the votes in the machine don't match the votes on the receipts if you don't count the receipts in the first place?

Read the article again, and this time, try to understand what is being said. The receipts are only used in the event of an "irregularity", ergo, no irregularity, no recount, and the tally maintained by the machine stands.

What you're describing is where the receipts are always counted. This isn't what California is proposing. And as I've commented elsewhere here, that is too bad. I'd prefer that they did this.

Re:No, idiot, no potential...

[raehl]

"Irregularity" = loser 'requests' recount, even if it's a small sample recount.

Don't buy machines that can "run out of paper"

[raehl]

Machines shouldn't have a roll of paper in them at all. They should have a feed port like most home printers that accept one sheet of paper at a time. You're given the sheet by an election official when you sign your name, you place the sheet on the input port of the voting machine, make your votes, and then the machine prints out a ballot. Deposit ballot in ballot box.

Now, the machine could run out of ink/toner, but hopefully a fresh set of either can be installed prior to the election.

Re:Don't buy machines that can "run out of paper"

[ElectricRook]

Now, the machine could run out of ink/toner

I have novel idea, we could use hollorith cards, they don't have the "out of toner" issue, they're fairly light resistant, and not completely destroyed by small amounts of water. The card can be read either by humans or machines. The technology is very stable, and has been in use for more than 120 years. The voters could mark their own ballots, even at home as a "absentee voter". Therefore home voters, and at the poll voters could use the same ballot. There's already a large number of reader machines distrubiuted across the country.

Re:Don't buy machines that can "run out of paper"

[matrix29]

Machines shouldn't have a roll of paper in them at all. They should have a feed port like most home printers that accept one sheet of paper at a time. You're given the sheet by an election official when you sign your name, you place the sheet on the input port of the voting machine, make your votes, and then the machine prints out a ballot. Deposit ballot in ballot box.

Now, the machine could run out of ink/toner, but hopefully a fresh set of either can be installed prior to the election.

The obvious hack would be to fake up ballots that get fed in by double-sheeting it or by tampering the software to spit out the occasional phony ballot from imaginary voters when the machine is idle. Remember that the cheapest ballot fraud trick is to bribe the electoral workers in secret. Adding in hidden software virus style would allow fake imaginary voter ballots to be added to the output list to skew the votes in the unwanted candidate's favor. Another trick would be to use a virus to intercept the ballot output, read it, null output the real vote, and fake up a ballot with the desired vote. With all software, the hardware has to be secure in proving that the output matches the actual ballot.

The actual vote counting software should be an unchallenging exercise for even a first year BASIC programmer. The verification codes should be included in the output to checksum the actual ballot with random variances tossed in to confuse any potential hidden viral intercept software that could be loaded later to tamper the actual votes.

The machines themselves should be very open and basic, but the ballot saving and output should be prime cryptography software so that each outputted ballot is unique and entirely verifiable along with each electronic ballot being verifiable. Now here is the kicker. All of the electronic ballot results should be sent to every city in the United States for independent verification as each ballot is cast to prevent delaying the results and to provide a tangible record of any real-time ballot tampering. Granted the feeds could be faked, but if the vote verification software is also open source and available to everyone then every Internet using voter could immediately react and correct any instances of ballot fraud.

What mess?

[John Jorsett]

The only "mess" we've had came about because one group didn't like the outcome of the election and decided to create a pseudo "crisis" on purpose so that the election could be decided by the courts. It was never about a failure of our systems of voting. This rush to electronic voting is an example of why it's a bad idea to let one incident stampede us into ill-advised "action" without reflection.

Re:What mess?

[Malcontent]

Yes and tens of thousands of jewish retirees actually meant to vote for Buchanan. Right!

Re:What mess?

[jjo]

Uh, the Buchanan problem was due to a ballot designed by DEMOCRATIC election supervisors. Besides that, the flawed ballot design was not the issue in the big court battle: the counting method (or lack of one) was.

No guarantee

[FreekyGeek]

This is an extremely positive step, and should greatly help prevent *accidental* mistakes by the machine. Keep one thing in mind, though: the vote the machine records and the one it prints out don't have to be the same.

If someone maliciously tampered with the actual code on the machine, it could print out a vote for Moron A while recording a vote for Moron B. Assuming they kept the tampering within reason (i.e., not a big enough discrepancy to make people suspect something and call for a count of the paper votes), a conspiracy to tamper with the voting wouldn't necessarily be prevented by paper reciepts.

In an environment where elections are often won and lost by only a few percent, a cleverly-altered machine would only have to change a few votes out of every hundred to make a big impact, certinly enough to swing very close elections.

Re:No guarantee

[Todd Knarr]

Remember that the receipt is actually a printed ballot, put into a ballot box at the polling place just like the current ballots. If the machine printed out one thing but recorded another, then during the inevitable recount (in CA a recount is automatic if the margin is less than a certain amount) they'll find a discrepancy between the results of the recount and the results reported by the machine. You start seeing that in several recounts, especially if it changes the outcome of the election, and there'll be enough of an outcry that an investigation will have to be started.

Give it a rest.

[Ungrounded Lightning]

I wonder if diebold CEO is still promising (and meaning it) to deliver W..
Oh, wait.
The printer was delayed until AFTER the next major election.

Give it a rest.

EVERY elected executive-branch office in California is held by a Democrat except the new gubernator - who is a flaming liberal on all issues except partly on fiscal AND married into the Kennedy clan and advised by them.

That includes the Secretary of State who promulgated this decision.

Yes we'd ALL love to have this done in time for '04. But CA is in debt up to its eyeballs and you KNOW the election machine companies will charge extra for a rush job.

It's going to be tough enough deciding how to handle the inevitable cases where the pissed-off voter comes to the official with the ballot stub and says "HEY! This machine didn't vote it right!" without leaving the barn door open for tampering with the electronic count.

CA won't go Bush anyhow.

[Ungrounded Lightning]

Way to deflect the issue, kids. "yeah yeah, we have to be accountable... but in two years". Too bad they're going to have a little thing like "presidential election" first before all that comes about, huh?

If there's a single state in the US that doesn't vote to re-elect Bush it will be California - hacked machines or not.

The (Democratic) Secretary of State just wants to get them fixed before The Gubernator is up for reelection. B-)

Interestingly, some people oppose this

[John Jorsett]

Here's an article in the Los Angeles Times about groups that oppose this. They claim, among other things, that it'll "make it harder for some people to vote." I love free speech, but sometimes the people exercising it just piss me the hell off.

Re:Everyone Demands a democracy, they don't get it

[crimethinker]

If a few hundred thousand americans showed up on the doorstep of the whitehouse, weapons and ballots raised, and demanded that their voting system be put back into order, and demanded that they don't fuck with them, the congressman would do it or order mass genocide.

My money's on the "mass genocide" option. Handgun possession is completely illegal in D.C. (bye-bye 2nd amendment since 1977), and any rifle or shotgun must not only be unloaded, but at least partially disassembled. The crowd would last all of 2 minutes 30 seconds before the cops started machine-gunning them down.

And of course, it's all in the name of the "war on terror" because the only reason to possess a firearm (thanks, Herr Schumer, Fraulein Clinton, und Fraulein Feinstein!) is for terrorism. Al Qaeda operatives will of course come to the US to buy machine guns, where it takes 6 months and $10K, rather than pay $100 cash at a street market in Kandahar. Fucking Nazis. (Sorry, Godwin.)

-paul

Relax

[djupedal]

It hasn't happened yet...just keep watching. Have another Oreo.

Re:Relax

[Simple-Simmian]

Bush won't win California.
Last election he didn't even try.
So why would anyone try and "steal" it for him.
That is the question I keep asking myself when I read this stuff.
Why would you try and steal an election in a state where you hardly tried to run?
Where you chance of winnning is near zero
The answer is you wouldn't
But thats too obvious to the tin foil hat crowd.
My fear is that in a democrat dominated state like California that the democrats would steal election like the Daley machine in Chicago does all the time.
I live in California so it's a issue I worry about.

I am all for a postal voting using scantron type paper ballots.
Forget these expensive machines.

Paper ballots are better even if these machines will have to have a voter verified recipt/paper trail by 2005 for new machines and 2007 for existing machines.
It's throwing good money after bad for a problem that didn't exist in California.

The paper ballot.

[Ungrounded Lightning]

But in the event of discrepancy, which one is considered correct? The machine record, or the paper record? Considering one makes the other.

This has already been discussed and published.

The paper ballot (which, presumably, at least some of the voters checked before depositing in the box) is the official ballot.

The electronic count from the machines is simply another counting process, and is declared official only if it is not challenged.

Further, a sample of precincts (randomly selected AFTER the election to prevent selective UN-gimmicking of the machines) are recounted, challenge or no, to check for bugs and detect (and deter) large-scale tampering.

Chile?

[GQuon]

Pinochet isn't running the place anymore, is he? At least they don't have pregnant chads down there.

NOTE TO MOD: ANSWER TO ANONYMOUS COWARD

[GQuon]

Note to Moderators: I was answering an Anonymous Coward.
I wasn't slagging Chile.

And two more.

[Ungrounded Lightning]

Reasons to use an electronic voting terminal:

1. [Error-checking to avoid spoiled ballots]

2. [Assistance to people with sensory or literacy handicaps]

3. [Eases implementation of preferential balloting]

4. Hand counting human-marked ballots is slow, error-prone, and susceptable to selective interpretation and/or cheating.

5. Electronic counting of hand-marked ballots (i.e. optical fill-in-the-oval readers) are error-prone. Election officials "helping out" by filling in light marks opens an opportunity to cheat and destroys the original record, while falling back to a manual count with ALL the problems of 4.

You SAW the screaming with a close election with punched cards in a close election. Human-interpreted X-in-box ballots have ALWAYS had this very same problem - magnified by the manual counting of ALL the ballots, rather than selected precincts. That's one of the things mechanical/electronic vote counting is intended to solve.

The combination of using a machine to collect the vote and mark the ballot (and prevent accidental invalid voting) with a paper ballot audit trail and human recounts to audit the machine's counting looks like the best solution yet.

Would it matter?

[roystgnr]

Your vote gets tossed into the same count as all the rest of them. It's not like an election fraud would require changing *every* vote cast, just the few percent or a few tens of percent of votes necessary to change who wins. Perhaps a discrepancy between the machine votes and the handcounted votes would be noticed, but perhaps it would be explained away with "well, sure all those Luddites voted for my opponent, but the voters without tinfoil hats clearly elected me!"

Re:no you miss the point

[bussdriver]

You're missing the point. Here is a quick top 10:

1) a receipt would be something you keep. they are talking audit trails.

2) who is to say the paper matches the computer? ONLY recounts would show anything going on. If candidate X loses by 10% do they ask for a recount?

3) How many "bugs" will produce incorrect printouts that will go unverified? (at least enough to win another 10+%)

4) Transaction ID system can be compromised.

5) A voter ID system is harder to compromise, but is illegal in that you are not anon.

6) A complex hack would involve dumping printouts you don't like; which is EZ since its automated!

7) A more compex hack would result in 2 printouts.

8) A paper trail will not effect the next 1 or 2 elections. So they can come up with other tricks.

9) verification could make butterfly ballots look like child's play. (BTW, nobody mentions how EZ that is to hack, and how it was done in miami..)

9.5) Why not print off a report later? who could tell the difference???....

10) The point behind breaking areas up, is that you can remove corrupted area's results. It also is supposed to make it more difficult to cheat. Machines negate this, esp. when there is only 1-2 kinds of them used.

The motivation to WASTE our money on something so cheaply and securly done, is either foolishness or something else...

Re:no you miss the point

[bussdriver]

clarification

#1) I mean by "they", I mean most the posts I read, and other concerned people. I have heard very little about take-home receipts; but then I'm around smarter people than that I guess...

Take-home crap is worthless; and obvious enough that it should not appease most people with a brain.

Yes, I read the article, like an hour ago. The posts are more fun to read anyhow.

Almost... But here's the fix.

[Ungrounded Lightning]

There is no reason a spoiled ballot from a computer should be treated any differently than a spoiled ballot from a punchcard machine, or you accidentally messing up an optical scan ballot.

Actually there is ONE reason: You have to clear the "spoiled ballot"'s count from the machine, to prevent corruption of the electonic count.

Sample procedure:
- Good ballots go into the "good ballot" box, sight-unseen.
- Spoiled ballots go in the "spoiled ballot" box, sight-unseen.
- Official sets the authorization token to "revoting spoiled ballot" to tell the machine to dump the previous vote and record the new one.

Variants:
1) Same machine is not reused by another voter until the previous voter declares his vote "good" or comes back to revote. Machine holds the tentative vote in temporary memory, recording it as valid only when the next voter votes or the election ends.
2) Machine stores each ballot, spoiled or not, separately. Re-authorization identifies the stored ballot that is to be marked "spoiled". Voter votes at same machine.
3) The smartcard voter-authorization token holds a copy of the ballot vote, which persists if the card is marked "revoting spoiled ballot". Voter can vote at any machine. (Note that this can produce negative vote totals on individual machines (if the valid and spoiled counts aren't tallied separatly), which is OK but might be confusing to some observers.)

Note that the voting machines can produce both a "valid" and a "spoiled" count for comparing against manual recounts.

ordinary paper?

[js7a]

If the machine could have the paper replaced like normal reciept printers, you would be clamoring about the security of the paper record.

No, I wouldn't.

It would be nearly impossible to meaninfully compromise the paper records by tampering with the supply, at least as far as I can tell.

What could be done with ordinary reciept paper in advance of it being printed upon? Any legible tampering would be pretty pbvious after it had been over-printed.

Re:ordinary paper?

[js7a]

The obvious defense against disappearing ink would be to use thermal reciept printers, which produces marks lasting for years if the paper isn't exposed to sunlight.

But still, how could disappearing ink be used to skew the vote? It could eliminate the paper trail, but surely for one time only, and when the tampering became obvious, forensic science could almost certainly detect the remnants of any disappearing ink that could be used in a reciept printer.

You misunderstand the system Also modest proposal

[Ungrounded Lightning]

There is a very big problem with this type of system: it enables vote selling (someone gives 10$ to every person that produces a receipt proving they voted for a specific candidate)

You misunderstand the "receipt".

They don't keep it. They put it in the ballot box for potential recounts. It IS the official ballot - the count in the machines is just a convenience.

- - - - -

The point about vote selling, however, is significant.

One thing I'd have liked to do, a few elections back, was to get a raw record of the ballot-reader output from the individual precincts with electronically-tabulated ballots (punched cards, OCR marked cards, etc.)

Putting the raw record up on the internet would allow:
- anyone to write their own software to check the counting software.
- individuals (or neighborhood groups) to check that a ballot marked THEIR way (or the correct number of ballots marked their way) was included in the count.
- Statistical analysis to hunt for signs of election tampering (i.e. runs of identical ballots, precincts with counts wildly divergent from polling expectations, systematic spoiling of ballots otherwise voted for one party or candidate, etc.)

Unfortunately this might fall under the bans on exposure of individual ballots that were passed to hinder vote-buying schemes.

Vote buying laws would have to be changed.

[Ungrounded Lightning]

If you want to see a really clever electronic voting system, check out VoteHere. They use paper receipts that basically records a hash of your vote, so your receipt cannot prove to anyone who was not looking over your shoulder when you cast the ballot what that vote was, but still allows you to prove that your vote has/has not been changed after the polls close.

If a take-home paper reciept can be used to verify TO YOU that your vote was counted, it can also be used to verify TO SOMEONE ELSE that you voted in a particular way.

This enables vote-buying schemes: Show up at the party headquarters, show your reciept, collect $10, a chicken dinner, or what-have-you.

Such schemes are banned by law. And the laws also prohibit you from taking home a copy of your ballot to aid in prevention of the schemes.

Re:Vote buying laws would have to be changed.

[ca1v1n]

RTFP/RTFM. You can't do vote-buying with the vote-here scheme, because the receipt doesn't say who you voted for. The receipt just shows a cryptographic hash of your vote that you can use to confirm that the vote didn't change. You can verify that the hash corresponds to the vote you intend in the voting booth, but once you leave the poll, there's no information left to prove who you voted for. The ballot remains anonymous, and your vote is secret.

The vote-buying schemes you describe are *exactly* what VoteHere's system is designed to avoid.

Re:Vote buying laws would have to be changed.

[Ungrounded Lightning]

You can verify that the hash corresponds to the vote you intend in the voting booth, but once you leave the poll, there's no information left to prove who you voted for

If you can verify that votes for A, B, and C encrypt to hash H, so can anybody else. Claim you voted for A, B, and C and got hash H, and Mr. Bagman can verify this and give you your $10

Re:Vote buying laws would have to be changed.

[ca1v1n]

The hash is different per ballot, and the extra information that distinguishes the ballots (and thus differentiates the hashes) is not tied in any way to voter identity. Check the documentation. They've got a rather complete threat/countermeasure matrix that explains why their system is secure enough to back up their claim that it's better than paper.

Random audits

[phr2]

Besides requiring a manual ballot count in the event of a close election, the HAVA bill also requires a manual count for a random small sample (I think 0.5%) of [i]all[/i] computerized elections. That should make deliberate fraud pretty scary.

Let's not distort history.

[Ungrounded Lightning]

In the 1970's millions came on capitol hill to do their work. Nowadays, if you get a couple hundred people together to protest anything you get charged with unlawful assembly and get teargassed.

They did that in the '60s and '70s, too. (Trust me - I was there.)

It just didn't USUALLY end up in the media - until Mayor Daily made the mistake of doing it to the reporters, too.

He did this JUST as the TV networks (with major studio/distribution centers in Chicago) deployed the first minicams (backpacks of electronics plus giant shoulder-mounted cameras). By the time the billyclub hit the lens the image was already on millions of TV screens. B-)

- - - -

Republics have ALWAYS been about finding out how the civil war would come out so you don't have to actually fight it. To do that the election has to be believablely accurate enough to predict the outcome of the war. That way the losers aren't tempted to reverse it by violence. (That's why each group of voters was given the franchise [for real] just AFTER they'd proven capable of organizing violence.)

The election doesn't HAVE to be dead-on accurate to serve this purpose. If a close election comes out wrong the losers still won't be tempted, because staging the war would bring out a lot of opposition from people who don't like reversing elections by force.

But it DOES have avoid the appearance of massive corruption. Or you get things like the cleanout of San Francisco's government back in the gold rush days, or the "Battle of Athens" just after WW II.

Fortunately the people of the US have generally resorted to force only to RETURN their governments to honest elections (or establish them in the first place), rather than replace it with a particular power-group's totalitarianism.

Re:Let's not distort history.

[TyrranzzX]

I agree. I don't like totalarian regimes, or dictatorships. I like democracy, in it's many forms. I'd prefer to sit here infront of my cosy computer, with my cosy lab and routers, eatin' warm mashed potatoes n' steak than getting angry over this shit.

But, thanks to modern day tech, we've got cameras inside of ligthers and portable video cams, and yes baby, the internet and p2p networks as our distrobution medium.

What I do hope happens is that this crap gets cleaned up by 2010 at the absolute latest. As long as the ability to vote is intact, and the govverment isn't taken over by assholes it can still be salvaged and repaired over time without violent conflict. I understand it takes time to redo the systems and explain to old people how electronic voting works between their latte's and 5 o-clock hair dressings. If I knew how to program I wouldn't be sitting here bitching, I'd be working on a stripped down linux kernel with a voting package that can work on x86 and proceeding to put up and get a webpage with the stuff on it slashdotted.

Disinterested voters SHOULDN'T vote.

[Ungrounded Lightning]

The majority of voters don't even bother to vote - should they all be made to?

Since the REAL purpose of elections in a Republic are to find out how the civil war would come out, so you don't have to fight it, people with no interest in the outcome of a particular issue SHOULDN'T vote on it. Requiring them to vote (or even making it TOO EASY to vote) pollutes the result and destabilizes the government.

(Of course it's up to THEM how interested they are.)

They DID something like that.

[Ungrounded Lightning]

This is a very good step being taken by California, but I think they need to go one step further and mandate a recount for every election, regardless of whether it is seen to have irregularities or not.

I believe they also mandated a manual recount of a random sample of the precincts, to detect large-scale machine tampering.

Of course the loser - or perhaps even the Secretary of State - will demand a recount of any precinct where the results are significantly different from unofficial polling results.

Diptard

[Simple-Simmian]

The Sec of State is elected in Califorina so Arnold is never going to appoint anyone unless the elected one keels over.
I doubt California is going to get showered with anything but shit. I know this because I live here.
Washington takes money out of California and we get about 70 cents of every dollar back with added unfunded mandates, that total way more than that.
Arnold is already a breath of fresh air.
Appointing a enviormentalist to be in charge of the enviroment what a concept. He is going to be a great governor. One like this state hasn't seen in decades. A governor that will govern in the peoples interest not some political parties.

No need to steal it

[djupedal]

Arnie just sold CA to Bush...wait for the announcement of Federal funds going west.

Re:Almost... But here's the fix.

[argent]

That's a bit overcomplex. Obviously the transaction on the computer side can't be considered complete until the voter has a chance to verify the printout. If they indicate it's spoiled, their vote is already voided before any third party is notified.

Rome before Caesar (unashamedly offtopic)

[townmouse]

Before the Social War, only those born in the city of Rome or its environs were eligible to vote - and given the speed of transport in those days, ordinary people could not spare the time and expense of travelling to Rome just to cast their vote. None of the other cities had any representation in central politics. Even after other Italians became citizens, Rome still had the bulk of the authority - and by this time the most effective route to power was to march an army into Rome, regardless of the opinions of the senate and people.

It's worth noting too that the tribe-based voting system was highly skewed in favour of the patricians and other hereditary nobles. Only when the ruling classes were about evenly divided did the mass of the population have any significance. An exception was the election of the tribune of the Plebs, an ombudsman with significant power, but not one of the most important officials.

The Roman republic was highly centralised by the standards of its time, aguably more so than the other states of comparable size (China, Parthia and Mauryan India). Where decisions were taken locally, they were made by Roman appointees. Generally speaking, governors in the Republic did not need or try to gain the support of the local population; they just pocketed as much money for themselves as possible. The aggrieved populace could petition Rome for redress, so it was important to ensure you extorted enough money in fines to be able to pay your own fine if you were caught. There were however one or two special cases (such as Genua IIRC) where the local elites were allowed a say in their own region's affairs.

Re:The real question is...

[four2five]

In light of the recent mistakes the bigger issue isn't for people to make sure that their vote was registered to the right candidate but to allow them to trace the behaviour of the system. It is important that the right vote is registered but I would think that systems that can't even count right would be a bigger issue. Also, the post to which I responded mayd the comment that this system wouldn't even matter if they voter didn't check their vote and that is wrong. The very fact that this provides some type of documentation over nothing, which there is now, is an important step in the right direction of holding these companies accountable. I'd much rather see an open source voting system with the source available to all, but I don't know if that will ever happen.

While we're at it.

[Ungrounded Lightning]

As long as the ability to vote is intact, and the govverment isn't taken over by assholes it can still be salvaged and repaired over time without violent conflict.

Unfortunately, sometimes it DOES get taken over by assholes who just ignore the voters. Especially if they think the voters can't do anything about it.

If it comes to that you either have to sit there and take it or resort to violence. (Which is what the Second Amendment is about.) To serve as a credible threat, the vote must be backed up by the POTENTIAL for a civil war.

Fortunately, with a well-armed and determined population you usually only have to resort to a LITTLE BIT of violence - along with a lot of talking and saber-rattling. Once they're convinced they've lost they usually go quietly and things return to normal.

Re:tamper proof

[matchlight]

anyone can run an md5 sum on a voting machine

This is not true. The system would not be public accessible even if the code was open source. This would be for security purposes since it's not good, as any sys admin will support, to allow open access to a machine with sensitive material. MD5 is an excellent verification tool but in this instance only a restricted group of individuals would be allowed to run it, and who is to say that they are not also the ones tampering with the application.

Although I love the idea of open source in many applications, to say that closed source is obsolete is not factually supported.
Additionally, given the wide percentage of users of closed source versus open source, the arguement against closed source seems a little premature. Call me ignorant and dishonest, but I'm just looking at the facts of consumer usage to date.

accesible to those who need it.

[twitter]

Twitter: anyone can run an md5 sum on a voting machine

matchlight: This is not true. The system would not be public accessible even if the code was open source.

Ah, but election officials can and they can invite anyone they need. This is a vast improvement over their current unverifiable faith in Dibold.

Call me ignorant and dishonest, but I'm just looking at the facts of consumer usage to date.

Why should I call you names? Did you say that people should use closed source junk or did you simply not that many people do? I looks like you said that closed source software was economically viable and therfore not obsolete becuse many people use it. There's a world of difference between that and saying that's the way things should be. Time will prove that current market share is no proof against technicall inferiority.

Re:2005? 2006?

[leerpm]

I was replying to the parent who said:

Way to deflect the issue, kids. "yeah yeah, we have to be accountable... but in two years". Too bad they're going to have a little thing like "presidential election" first before all that comes about, huh?

Before electronic voting I thought...

[SmashPDX]

... that punch cards had problems. Dang, bring them back. As long as you ignore Florida and leave it to the states that can actually figure out how to vote everything works fine. And hey, you don't have to generate e-waste. :)