Chicago Voter Base Hacked

Mephisto · 24th October 2006, 05:57 PM

Fortunately, it was a non-partisan group who did the hacking just to show that it could be done. Hanging chads never looked so good.

Chicago Voter Database Hacked

By JAKE TAPPER and REBECCA ABRAHAMS

Oct. 23, 2006 — As if there weren't enough concerns about the integrity of the vote, a non-partisan civic organization today claimed it had hacked into the voter database for the 1.35 million voters in the city of Chicago.

Bob Wilson, an official with the Illinois Ballot Integrity Project — which bills itself as a not-for-profit civic organization dedicated to the correction of election system deficiencies — tells ABC News that last week his organization hacked the database, which contains detailed information about hundreds of thousands of Chicago voters, including their Social Security numbers, and dates of birth.

"It was a serious identity theft problem, but also a problem that could potentially create problems with the election," Wilson said.

A nefarious hacker could have changed every voter's status from active to inactive, which would have prevented them from voting, he said.

"Or we could've changed the information on what precinct you were in or what polling place you were supposed to go to," he said. "So there were ways that we could potentially change the entire online data base and disenfranchise voters throughout the entire city of Chicago."

"If we'd wanted to, we could've wiped the entire database out," Wilson claimed.

http://abcnews.go.com/Politics/story?id=2601085&page=1

Ace_of_Sevens · 24th October 2006, 06:14 PM

So all your base are belong to us?

Mephisto · 24th October 2006, 06:23 PM

Originally Posted by Ace_of_Sevens

So all your base are belong to us?

Wheezebucket · 24th October 2006, 06:34 PM

Bob Wilson. Not quite the 'Cereal Killer' or 'Crash Override' I was hoping for. I wish hacking was more like Hackers.

He totally nailed that Gibson.

Mephisto · 25th October 2006, 05:07 AM

Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale?

It wouldn't take much to skew the election results either way. If the hackers were smart, they wouldn't make the results so obviously "doctored." Perhaps changing the party preference of only a few people so as not to raise an eyebrow afterwards would do perfectly.

Of course, it would be really comical (and get the point across nicely) if the hackers make it possible that every other vote would be cast for Homer Simpson.

Ladewig · 25th October 2006, 05:39 AM

Originally Posted by Mephisto

Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale?

Well, for starters, they could make voting machines that couldn't be opened with hotel minibar keys.

http://www.wired.com/news/politics/e...=wn_politics_1

Zep · 25th October 2006, 05:53 AM

Originally Posted by Mephisto

Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale?

Why, yes!

We call it "a piece of paper" or "a ballot".

You use it like this...

Then you do this with it...

Then when everyone is done, lots of people count the votes, then check the count. Then they see who has won! And if they need to recheck, they get the pieces of paper out and count again.

Ain't technology grand?

Mephisto · 25th October 2006, 06:09 AM

Originally Posted by Zep

Why, yes!

We call it "a piece of paper" or "a ballot".

http://www.aph.gov.au/senate/pubs/od...es/chap4-1.jpg

You use it like this...

http://abc.net.au/elections/wa/2005/...tos/voting.jpg

Then you do this with it...

http://www.abc.net.au/elections/sa/2006/img/voting.jpg

Then when everyone is done, lots of people count the votes, then check the count. Then they see who has won! And if they need to recheck, they get the pieces of paper out and count again.

Ain't technology grand?

Then the best course is clearly - two steps backward. No hanging chads, no suspicious results, just counting what people actually marked on their ballots. Now, if we can keep track of all the ballots from all the different areas we'd be okay.

(edited to add) Maybe painting people's thumbs purple with an indelible ink to show they've already voted.

Hutch · 25th October 2006, 06:19 AM

Chicago voter base hacked?

You mean they moved all the dead people to the suburbs?

Mephisto · 25th October 2006, 06:38 AM

Originally Posted by Hutch

Chicago voter base hacked?

You mean they moved all the dead people to the suburbs?

Excellent point, Hutch, but the fact that those "voting irregularities" are (more or less) common knowledge could indicate that a method which leaves a paper trail (along with better voting practices, such as a picture ID voting card) is far more desireable than electronic voting methods alone.

Hutch · 25th October 2006, 06:46 AM

FWIW, I tend to agree with you, Mephisto, that as voting becomes more technological, the chances for abuse (and not just the dead voting and other shenanigans that have been part of the American body politic throughout our history) will grow. There were suspicions about Ohio in 2004; what will 2008 bring?

I think I prefer the 18th century custom of providing whiskey to potential voters outside the polling places...

Cleon · 25th October 2006, 06:49 AM

Originally Posted by Mephisto

Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale?

Anyone who tries to tell you that an electronic system can't be hacked is either lying or ignorant. (I don't mean "ignorant" in the negative sense--I mean ignorant as in "they really don't know any better.")

There is no such thing as a "hack-proof" system. There are things you can do to minimize security holes, steps you can take to try and make sure the system is as secure as humanly possible, but a system that "couldn't be hacked?" Nope.

Katana · 25th October 2006, 06:50 AM

Originally Posted by Hutch

I think I prefer the 18th century custom of providing whiskey to potential voters outside the polling places...

Lovely idea! Sign me up.

Cleon · 25th October 2006, 06:50 AM

Any minute now corplinx will be along to remind us that any questioning of Diebold or other voting machine security is "woo."

Whyatica · 25th October 2006, 06:56 AM

I would love to see the hackers change 51% of the votes to Robin Williams (see his latest movie). Does Chicago use the same Diebold voting machines?

Luke T. · 25th October 2006, 07:11 AM

There is a difference between a voter database and a voting machine. The hackers broke into the voter database.

And as for paper ballots, the error rate for them is higher than for electronic machines.

Geez, you'd think no one had ever heard of "stuffing the ballot box" prior to the invention of the computer chip.

Zep, how hard would it be for someone with a copying machine to reproduce the paper ballots in your post by the thousands?

Everything is hackable. But "hackable" does not equal "hacked".

Improvements need to be made. Definitely.

Cleon · 25th October 2006, 08:09 AM

Originally Posted by Luke T.

There is a difference between a voter database and a voting machine. The hackers broke into the voter database.

In this specific case, that's true. However, hackers have broken into voting machines as well--often using very simple methods. It's not "this case" or "that case" that's the problem, it's the compounding levels of security issues at all levels of the system. (By "system" I'm referring to everything involved in the electronic voting system, from the base DB, to tabulation, to the machines themselves, to the process involved.)

Quote:

And as for paper ballots, the error rate for them is higher than for electronic machines.

That may be true, but with paper ballots there remains a physical record you can check the results against. That's not the case with a computerized system.

There's also the issue of severity; a single security issue with a computerized system would, by its very nature, have far more impact than a voter who bungles a paper ballot. It would only take a single exploit to change the results of an entire election--potentially, with no trace that the results were falsified.

Quote:

Geez, you'd think no one had ever heard of "stuffing the ballot box" prior to the invention of the computer chip.

Zep, how hard would it be for someone with a copying machine to reproduce the paper ballots in your post by the thousands?

Not very. How hard would it be for him to actually carry out that sort of fraud? Very, without the collusion of the election workers.

First, paper ballots are usually specialized card stock, not your average 20# paper you find in your local Kinko's machine. Second, it's usually pretty simple to detect photocopies. Third, you'd have to stuff the box without being noticed (or with the collusion of) the multiple election staff onhand.

Most importantly, of course, is the fact that even if someone were to get away with that (and I have no doubt it can be done, it certainly has before), the skewed results would be localized to a single polling place. You would have to involve a conspiracy of a fair number of people at different polling locations to be able to make that happen on a scale large enough to influence an election. And if we're talking about a statewide election--such as for a senator, governor, or President--then forget about it. The logistics of a conspiracy on that scale would be difficult to impossible without someone getting wind of it.

Quote:

Everything is hackable. But "hackable" does not equal "hacked".

This is very true. However, at the heart of this issue, even beyond whether the systems are "hackable" or not, is the issue of accountability and verifiability (is that a word?). We can go back and verify paper ballots; we can verify that they're legitimate, we can verify the count was correct. The system may not be perfect, but there's some accountability there.

With an electronic system, we don't have that option. Because the systems are proprietary, we can't even go and evaluate the tabulation methodology. We have to have faith that the results are correct and accurate.

Luke T. · 25th October 2006, 08:42 AM

Originally Posted by Cleon

Most importantly, of course, is the fact that even if someone were to get away with that (and I have no doubt it can be done, it certainly has before), the skewed results would be localized to a single polling place. You would have to involve a conspiracy of a fair number of people at different polling locations to be able to make that happen on a scale large enough to influence an election. And if we're talking about a statewide election--such as for a senator, governor, or President--then forget about it. The logistics of a conspiracy on that scale would be difficult to impossible without someone getting wind of it.

And wouldn't that be true for electronic voting machines? You would have to have a fair number of people involved to make fraud happen on a large scale?

drkitten · 25th October 2006, 09:01 AM

Originally Posted by Luke T.

And wouldn't that be true for electronic voting machines? You would have to have a fair number of people involved to make fraud happen on a large scale?

No. Just hack the central tabulating faciility.

Or, for that matter, hack the source code for the electronic voting machines.

That's part of the problem -- single point of failure.

Cleon · 25th October 2006, 09:11 AM

Originally Posted by Luke T.

And wouldn't that be true for electronic voting machines? You would have to have a fair number of people involved to make fraud happen on a large scale?

Not at all, because it's all centralized. You set up a laptop at a local coffeeshop, hack a couple of central tabulating hubs or DBs, and you can affect votes not just all across the state, but potentially any other state that uses the same system.

Or you could, as drkitten said, get in early and hack the source code. At which point you don't actually have to do anything the day of the vote; it's already done for you.

varwoche · 25th October 2006, 09:33 AM

Originally Posted by Cleon

With an electronic system, we don't have that option. Because the systems are proprietary, we can't even go and evaluate the tabulation methodology. We have to have faith that the results are correct and accurate.

The proprietary aspect is SO bogus that it's a (not very funny) joke.

A paper receipt doesn't really solve the problem because the receipt is output by the same potentially compromised system. I finally heard a proposal that makes some sense, and that's a digital camera that photographs the voting screen and saves the images for auditing / recounts.

Cleon · 25th October 2006, 09:38 AM

Originally Posted by varwoche

The proprietary aspect is SO bogus that it's a (not very funny) joke.

Do you mean that it's a (not very funny) joke that the system is proprietary, or that it's a (not very funny) joke that I bring it up as an objection?

Quote:

A paper receipt doesn't really solve the problem because the receipt is output by the same potentially compromised system.

True to an extent, however, the voter could eyeball it to make sure that the receipt actually reflected the votes cast.

Quote:

I finally heard a proposal that makes some sense, and that's a digital camera that photographs the voting screen and saves the images for auditing / recounts.

Meh. Possibly, I suppose, but I see lots of room for failure--not to mention privacy issues (I really don't want my picture taken, even just the back of my head, with my voting choices on the screen).

varwoche · 25th October 2006, 09:42 AM

Originally Posted by Cleon

Do you mean that it's a (not very funny) joke that the system is proprietary, or that it's a (not very funny) joke that I bring it up as an objection?

Pardon me, the former.

geni · 25th October 2006, 09:51 AM

Originally Posted by drkitten

No. Just hack the central tabulating faciility.

Or, for that matter, hack the source code for the electronic voting machines.

Or go after the compiler so a review of the source code wont pick up any problems.

Beth · 25th October 2006, 10:29 AM

Originally Posted by varwoche

The proprietary aspect is SO bogus that it's a (not very funny) joke.

A paper receipt doesn't really solve the problem because the receipt is output by the same potentially compromised system. I finally heard a proposal that makes some sense, and that's a digital camera that photographs the voting screen and saves the images for auditing / recounts.

The best system I've heard of is a dual paper receipt system. Two copies are printed, one goes to the voter so he/she can verify that it is correct and verify that the two are the same. Once verified, the voter keeps one copy, the other goes into a sealed box so that, if and when a recount is done, the paper copies can be used to verify the voting maching tally.

drkitten · 25th October 2006, 10:43 AM

Originally Posted by Cleon

True to an extent, however, the voter could eyeball it to make sure that the receipt actually reflected the votes cast.

Won't work. The voter has no way of knowing that the electronic vote that was recorded in the database was the same as the one printed on the receipt.

For that matter, he has no guarantee that the vote machine doesn't happily make up non-existent votes in its spare time to keep itself amused.

Cleon · 25th October 2006, 11:31 AM

Originally Posted by drkitten

Won't work. The voter has no way of knowing that the electronic vote that was recorded in the database was the same as the one printed on the receipt.

For that matter, he has no guarantee that the vote machine doesn't happily make up non-existent votes in its spare time to keep itself amused.

Point taken.

Though I did just get a rather amusing mental image of a voting machine, sitting by itself in some obscure precinct, deciding to keep away the boredom by playing "Pong" with the votes.

NoZed Avenger · 25th October 2006, 02:02 PM

Voting shenanigans in Chicago?

Absurd.

Rob Lister · 25th October 2006, 02:25 PM

Originally Posted by Beth

The best system I've heard of is a dual paper receipt system. Two copies are printed, one goes to the voter so he/she can verify that it is correct and verify that the two are the same. Once verified, the voter keeps one copy, the other goes into a sealed box so that, if and when a recount is done, the paper copies can be used to verify the voting maching tally.

Yea, I'd love to see that system implemented.

If you listen to some people, you've got people too stupid to read the ballot and mark the correct slot. How are they going to compare two sets of receipts? They'd have to be double smart for that!

Me? I'm of the opinion that if you're too stupid to vote using any system currently in use, you shouldn't be voting.

There will always be fraud attempted and fraud achieved (chad dimple anyone?). My bet is there is going to be less fraud with the electronic system.

drkitten · 25th October 2006, 02:29 PM

Originally Posted by Rob Lister

There will always be fraud attempted and fraud achieved (chad dimple anyone?). My bet is there is going to be less fraud with the electronic system.

I'm afraid that we disagree. My bet is that there is going to be more, simply because it's much easier and the potential stakes are so much higher.

If I can, for example, intimidate you out of voting, I've shifted the vote count by one.

If I can program the computer to drop your vote -- and the vote of everyone else who voted as you did -- on the floor, then I have guaranteed myself the election.

With the old system, I needed to persuade 51% of the electorate.

With the new (post 2000) system, I need to persuade five of nine Supreme Court Justices.

With the incoming system, I only need to persuade one programmer.

Rob Lister · 25th October 2006, 02:33 PM

If, if, if.

Line them up and you've got a lot of Ifs.

I don't care either way.

No matter the system.

aerosolben · 25th October 2006, 02:38 PM

Originally Posted by Cleon

Not at all, because it's all centralized. You set up a laptop at a local coffeeshop, hack a couple of central tabulating hubs or DBs, and you can affect votes not just all across the state, but potentially any other state that uses the same system.

I would have thought that the machines are individually auditable. Even if you can't get the details, they would at least be able to report what they sent to the central facility.

If not, the design goes from extremely stupid to astoundingly stupid. If you aren't going to secure the machine, at least avoid making wide-spread tampering somewhat difficult.

Originally Posted by drkitten

Won't work. The voter has no way of knowing that the electronic vote that was recorded in the database was the same as the one printed on the receipt.

But this is an impossible problem to solve. There is no way for the voter to confirm this - some system along the line could always be lying.

However, it does allow you to audit, which is the real goal. A computerized system could report the immediate results, and a hand count could be done as well. Any non-trivial discrepencies would indicate some failure in the system. A (well-designed) computer system could even help guard against fraud in the human side of the equation by producing an independant count.

Obviously, it still isn't foolproof, but it is more secure.

drkitten · 25th October 2006, 02:49 PM

Originally Posted by aerosolben

But this is an impossible problem to solve. There is no way for the voter to confirm this - some system along the line could always be lying.

Actually, Beth's duplicate paper reciepts does a good job. The voter confirms that the reciept as printed matches her vote, and then puts a (validated) copy of the vote in a traditional ballot box.

We can do spot-checks of accuracy by confirming that the counts in the ballot boxes are identical to the counts as reported by the machines, which is an easier task than counting all the ballot boxes. Any substantial difference becomes grounds for questioning the validity of the machines and we then use the ballot boxes as traditional paper boxes or have some other resolution method. Hopefully such a procedure will not be necessary very often.

For that matter, there are a number of cryptographically secure methods of confirming end-to-end security of votes. For example, one could "publish" a list of votes cast (and their totals), but with identifying information hidden via encryption. Anyone could then confirm that the vote totals were in fact accurate, while each individual voter could re-encrypt her own information and identify her vote in the published list to confirm that it was counted accurately.

aerosolben · 25th October 2006, 03:20 PM

Originally Posted by drkitten

Actually, Beth's duplicate paper reciepts does a good job. The voter confirms that the reciept as printed matches her vote, and then puts a (validated) copy of the vote in a traditional ballot box.

I was referring to the electronic count - the voter cannot validate that the system has recorded the vote they expect. What you mention is essentially just paper balloting if you use it as the official count. If it backs the electronic count, then it is auditing as I mentioned, and I of course agree with your points.

Quote:

For that matter, there are a number of cryptographically secure methods of confirming end-to-end security of votes. For example, one could "publish" a list of votes cast (and their totals), but with identifying information hidden via encryption. Anyone could then confirm that the vote totals were in fact accurate, while each individual voter could re-encrypt her own information and identify her vote in the published list to confirm that it was counted accurately.

I could make a snide comment about perfect security and cryptography, but I'll refrain.

Instead, I'll say that first, building secure voting systems is not a trivial problem. While it is trivial to beat what we have currently, it is difficult to account for all vulnerabilities even as an academic exercise - a large part of which is avoiding extending trust to any part of the system. Second, the voter isn't really doing the things you describe - the voting machine is, and the voter really has no idea what's going on. Most academic approaches will ignore this aspect of the problem, which is why I focused on it.

I still think it's practically doable (especially if focused on major vulnerabilities and made some limiting assumptions about the problem environment).

I think we're largely on the same page here.

Luke T. · 25th October 2006, 03:31 PM

This is the age of the ATM and computers. We should move in the direction of electronic voting, not regress to paper ballots.

Work out the bugs. Progress. Welcome to the 21st century.

And if anyone knows how to hack an ATM, PM me.

drkitten · 25th October 2006, 03:32 PM

Originally Posted by Luke T.

This is the age of the ATM and computers. We should move in the direction of electronic voting, not regress to paper ballots.

Work out the bugs.

First step in that is to identify what the bugs are, and to fix them.

We don't need to deploy the technology until it actually works, you know.

HerNibs · 25th October 2006, 05:20 PM

Hacking the central tabulation system would only be one step towards tipping an election.

Many tabulation systems are stand alone, not on a network, manual updates and only connected to the outside world when receiving results. Many results are contained on memory cards that the election judges, one from each party, carry into the election center for upload. If a tabulation system is hacked, then the hacker must make sure that a second set of results are not being sent or have been sent.

When a person votes, they do not just walk in and vote – they check in with election judges/workers. Here they sign into a poll book, show ID in some states and it is verified that they have not early voted or sent in an absentee ballot.

If a person is not on the rolls, they then vote provisionally. Provisional ballots are not counted until it is verified that the person is eligible to vote in that election and that they have not voted anywhere else.

When results are tabulated, the number of votes is supposed to match the number of signatures in the poll books that are from the same voting site. Generally these numbers are only one or two off. If these numbers do not match, the difference must be shown to the representatives that will certify the results. These representatives are from the different parties.

If some one can overcome all of these, then an election could be rigged. I am not saying it is impossible, but darn hard.

Even if a dishonest machine tech can tamper with a tabulation system, if nothing else matches (turn out, signatures, absentee and early votes), the whole thing would be suspect.

I like the idea of the two printed verification methods. Gives the election official the ability to get a good, second source recount.

Regardless of what type of voting machine or method is used, the hard part would be to get thousands of election workers together into the conspiracy.

I believe most of the problem is either carelessness or incompetence on the both sides, election officials and the public.

When I was in elections, we frequently had people who did some very strange stuff.

Luke T. · 25th October 2006, 05:26 PM

Originally Posted by drkitten

First step in that is to identify what the bugs are, and to fix them.

We don't need to deploy the technology until it actually works, you know.

Is that the way it works? Technology is not deployed until it actually works? What's a customer service desk for then?

I remember ATMs had bugs at the beginning. Giving out more cash than was asked for. Bonus!

Some bugs aren't caught until the stuff is deployed. It works great in the lab, but then all kinds of problems crop up in the real world application. Every piece of military gear I ever worked with that was new was like that.

And after the 2000 elections, there was a screaming demand for electronic voting machines. Now! We want them now!

I suppose the people in charge of the voter database in the OP that was hacked should have stuck with paper voter rolls?

Everything is hackable.

varwoche · 25th October 2006, 08:56 PM

Originally Posted by Luke T.

Everything is hackable.

Which is why auditability -- votes and source code -- is critical.

Zep · 26th October 2006, 04:27 AM

Originally Posted by Luke T.

And as for paper ballots, the error rate for them is higher than for electronic machines.

It's certainly possible that they MIGHT be, but it does depend on the processes of using them.

Originally Posted by Luke T.

Geez, you'd think no one had ever heard of "stuffing the ballot box" prior to the invention of the computer chip.

Yep! Absolutely undeniable!

Originally Posted by Luke T.

Zep, how hard would it be for someone with a copying machine to reproduce the paper ballots in your post by the thousands?

The way we operate the voting practices here, it is exceedingly UNlikely that such a practice would succeed. Just a quick run-down on what actually happens when you vote here:

0) All major elections are conducted by the Australian Electoral Commission, a government body whose job it is to ensure that all elections are absolutely squeaky clean and conducted correct to the last letter of the law.
1) When you vote, you approach the electoral official in the polling station, who physically marks your name off the roll of voters.
2) They then initial some blank ballot papers to identify them, and they are passed to you. The number of ballot papers is known before the election.
3) You retire to your little voting booth and mark your ballot paper. Spoiled papers must be returned to the official, who keeps them aside in a special bin, and issues you a new initialed ballot.
4) After marking, you fold the ballot and put it in the voting box, under the watchful eye of another electoral official. You cannot leave the station with a ballot on you, or without voting.
5) The first pass of the vote counting process, usually under the eyes of a half-a-zillion scrutineers anyway, is to ensure that the count of ballot papers reconciles - unused, spoiled, voted, etc. Then the counting of votes proceeds by hand. The results are usually checked twice, more if there are discrepancies, then reported to a central tally room. Only then do computers come into it (and a competent spreadsheet program would probably be sufficient!

)
6) The ballots are then all officially packaged, securely sealed, and archived safely, in case of recounts, challenges, etc.

There are actually a number of places that this process CAN be rorted (which has been known to have happened), but the paper voting system itself is still proven and approaching foolproof. I also agree there are possibly more efficient ways of doing the same thing while retaining the reliability of this method.

Originally Posted by Luke T.

Everything is hackable. But "hackable" does not equal "hacked".

Improvements need to be made. Definitely.

Totally agree, even for paper ballots.

24th October 2006, 05:57 PM	#1
Mephisto Philosopher Join Date: Apr 2005 Location: in a state of disbelief Posts: 6,065	Chicago Voter Base Hacked Fortunately, it was a non-partisan group who did the hacking just to show that it could be done. Hanging chads never looked so good. Chicago Voter Database Hacked By JAKE TAPPER and REBECCA ABRAHAMS Oct. 23, 2006 — As if there weren't enough concerns about the integrity of the vote, a non-partisan civic organization today claimed it had hacked into the voter database for the 1.35 million voters in the city of Chicago. Bob Wilson, an official with the Illinois Ballot Integrity Project — which bills itself as a not-for-profit civic organization dedicated to the correction of election system deficiencies — tells ABC News that last week his organization hacked the database, which contains detailed information about hundreds of thousands of Chicago voters, including their Social Security numbers, and dates of birth. "It was a serious identity theft problem, but also a problem that could potentially create problems with the election," Wilson said. A nefarious hacker could have changed every voter's status from active to inactive, which would have prevented them from voting, he said. "Or we could've changed the information on what precinct you were in or what polling place you were supposed to go to," he said. "So there were ways that we could potentially change the entire online data base and disenfranchise voters throughout the entire city of Chicago." "If we'd wanted to, we could've wiped the entire database out," Wilson claimed. http://abcnews.go.com/Politics/story?id=2601085&page=1
	__________________ "What difference does it make to the dead, the orphans and the homeless, whether the mad destruction is wrought under the name of totalitarianism or the holy name of liberty or democracy?" Mahatma Gandhi

24th October 2006, 06:14 PM	#2
Ace_of_Sevens Thinker Join Date: Mar 2005 Posts: 225	So all your base are belong to us?
Ace_of_Sevens Thinker Join Date: Mar 2005 Posts: 225

24th October 2006, 06:23 PM	#3
Mephisto Philosopher Join Date: Apr 2005 Location: in a state of disbelief Posts: 6,065	Originally Posted by Ace_of_Sevens So all your base are belong to us?
	__________________ "What difference does it make to the dead, the orphans and the homeless, whether the mad destruction is wrought under the name of totalitarianism or the holy name of liberty or democracy?" Mahatma Gandhi

24th October 2006, 06:34 PM	#4
Wheezebucket Master Cylinder Join Date: Jan 2006 Location: Mr. Belvedere's House (in Seattle) Posts: 1,493	Bob Wilson. Not quite the 'Cereal Killer' or 'Crash Override' I was hoping for. I wish hacking was more like Hackers. He totally nailed that Gibson.
	__________________ The Evangelists myspace.com/evangelistscomic "My nipples start to bleed on very long runs." - Ian "We called her the gorgon. Not because we liked her." - Nikki

25th October 2006, 05:07 AM	#5
Mephisto Philosopher Join Date: Apr 2005 Location: in a state of disbelief Posts: 6,065	Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale? It wouldn't take much to skew the election results either way. If the hackers were smart, they wouldn't make the results so obviously "doctored." Perhaps changing the party preference of only a few people so as not to raise an eyebrow afterwards would do perfectly. Of course, it would be really comical (and get the point across nicely) if the hackers make it possible that every other vote would be cast for Homer Simpson.
	__________________ "What difference does it make to the dead, the orphans and the homeless, whether the mad destruction is wrought under the name of totalitarianism or the holy name of liberty or democracy?" Mahatma Gandhi

25th October 2006, 05:39 AM	#6
Ladewig Hipster alien Join Date: Dec 2001 Location: not measurable Posts: 16,792	Originally Posted by Mephisto Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale? Well, for starters, they could make voting machines that couldn't be opened with hotel minibar keys. http://www.wired.com/news/politics/e...=wn_politics_1
	__________________ Is the JREF message board training wheels for people who hope to one day troll other message boards? It is not that hard to get us to believe you. We are not the major leagues or even the minor leagues. We are Pee-Wee baseball. If you love striking out 10-year-olds, then you'll love trolling our board.

25th October 2006, 05:53 AM	#7
Zep Banned Join Date: Sep 2002 Posts: 26,985	Originally Posted by Mephisto Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale? Why, yes! We call it "a piece of paper" or "a ballot". You use it like this... Then you do this with it... Then when everyone is done, lots of people count the votes, then check the count. Then they see who has won! And if they need to recheck, they get the pieces of paper out and count again. Ain't technology grand?
Zep Banned Join Date: Sep 2002 Posts: 26,985	Last edited by Zep; 25th October 2006 at 05:55 AM.

25th October 2006, 06:09 AM	#8
Mephisto Philosopher Join Date: Apr 2005 Location: in a state of disbelief Posts: 6,065	Originally Posted by Zep Why, yes! We call it "a piece of paper" or "a ballot". http://www.aph.gov.au/senate/pubs/od...es/chap4-1.jpg You use it like this... http://abc.net.au/elections/wa/2005/...tos/voting.jpg Then you do this with it... http://www.abc.net.au/elections/sa/2006/img/voting.jpg Then when everyone is done, lots of people count the votes, then check the count. Then they see who has won! And if they need to recheck, they get the pieces of paper out and count again. Ain't technology grand? Then the best course is clearly - two steps backward. No hanging chads, no suspicious results, just counting what people actually marked on their ballots. Now, if we can keep track of all the ballots from all the different areas we'd be okay. (edited to add) Maybe painting people's thumbs purple with an indelible ink to show they've already voted.
	__________________ "What difference does it make to the dead, the orphans and the homeless, whether the mad destruction is wrought under the name of totalitarianism or the holy name of liberty or democracy?" Mahatma Gandhi

25th October 2006, 06:19 AM	#9
Hutch A broken man on a Halifax pier, the last of Barrett's Privateers Join Date: Dec 2003 Location: About 7 Miles from the Saturn 5B Posts: 6,533	Chicago voter base hacked? You mean they moved all the dead people to the suburbs?
	__________________ If sheer righteous fury could accomplish anything worthwhile, Wolverines would have inherited the Galaxy long ago." -Web DuHavel, David Weber's "Honorverse" Series

25th October 2006, 06:38 AM	#10
Mephisto Philosopher Join Date: Apr 2005 Location: in a state of disbelief Posts: 6,065	Originally Posted by Hutch Chicago voter base hacked? You mean they moved all the dead people to the suburbs? Excellent point, Hutch, but the fact that those "voting irregularities" are (more or less) common knowledge could indicate that a method which leaves a paper trail (along with better voting practices, such as a picture ID voting card) is far more desireable than electronic voting methods alone.
	__________________ "What difference does it make to the dead, the orphans and the homeless, whether the mad destruction is wrought under the name of totalitarianism or the holy name of liberty or democracy?" Mahatma Gandhi

25th October 2006, 06:46 AM	#11
Hutch A broken man on a Halifax pier, the last of Barrett's Privateers Join Date: Dec 2003 Location: About 7 Miles from the Saturn 5B Posts: 6,533	FWIW, I tend to agree with you, Mephisto, that as voting becomes more technological, the chances for abuse (and not just the dead voting and other shenanigans that have been part of the American body politic throughout our history) will grow. There were suspicions about Ohio in 2004; what will 2008 bring? I think I prefer the 18th century custom of providing whiskey to potential voters outside the polling places...
	__________________ If sheer righteous fury could accomplish anything worthwhile, Wolverines would have inherited the Galaxy long ago." -Web DuHavel, David Weber's "Honorverse" Series

25th October 2006, 06:49 AM	#12
Cleon King of the Pod People Moderator Join Date: Aug 2001 Location: Atlanta, GA Posts: 20,506	Originally Posted by Mephisto Maybe this is a question for the science forum, but is there a electronic voting machine that couldn't be hacked, and could it be implemented on a such a wide scale? Anyone who tries to tell you that an electronic system can't be hacked is either lying or ignorant. (I don't mean "ignorant" in the negative sense--I mean ignorant as in "they really don't know any better.") There is no such thing as a "hack-proof" system. There are things you can do to minimize security holes, steps you can take to try and make sure the system is as secure as humanly possible, but a system that "couldn't be hacked?" Nope.
	__________________ "People like me are what stand between us and Auschwitz." - Newt Gingrich

25th October 2006, 06:50 AM	#13
Katana Illuminator Join Date: May 2006 Posts: 4,894	Originally Posted by Hutch I think I prefer the 18th century custom of providing whiskey to potential voters outside the polling places... Lovely idea! Sign me up.
Katana Illuminator Join Date: May 2006 Posts: 4,894	__________________ "I think Katana is as big of a perv as the rest of us." - Dragonrock "The rationality was there, and clear and concise. The condescention was hinted at and was like french onion dip on the perfect potato chip. Tasted like woo smackdown." - Fowlsound (aka Ducky, darnit) "Katana is one quick shut-yo-mouth!" - JonnyFive StopSylviaBrowne

25th October 2006, 06:56 AM	#15
Whyatica Chatroom Über-god Join Date: Apr 2004 Location: Orange County, CA Posts: 412	I would love to see the hackers change 51% of the votes to Robin Williams (see his latest movie). Does Chicago use the same Diebold voting machines?
	__________________ Come join the SkepticsRock Chat now! "Do you understand what the chain of command is? It's the chain I'm going to beat you with until you understand who's in ruttin' command here!" Jayne, Firefly

25th October 2006, 07:11 AM	#16
Luke T. Guest Join Date: May 2003 Posts: 14,759	There is a difference between a voter database and a voting machine. The hackers broke into the voter database. And as for paper ballots, the error rate for them is higher than for electronic machines. Geez, you'd think no one had ever heard of "stuffing the ballot box" prior to the invention of the computer chip. Zep, how hard would it be for someone with a copying machine to reproduce the paper ballots in your post by the thousands? Everything is hackable. But "hackable" does not equal "hacked". Improvements need to be made. Definitely.
Luke T. Guest Join Date: May 2003 Posts: 14,759

25th October 2006, 08:09 AM	#17
Cleon King of the Pod People Moderator Join Date: Aug 2001 Location: Atlanta, GA Posts: 20,506	Originally Posted by Luke T. There is a difference between a voter database and a voting machine. The hackers broke into the voter database. In this specific case, that's true. However, hackers have broken into voting machines as well--often using very simple methods. It's not "this case" or "that case" that's the problem, it's the compounding levels of security issues at all levels of the system. (By "system" I'm referring to everything involved in the electronic voting system, from the base DB, to tabulation, to the machines themselves, to the process involved.) Quote: And as for paper ballots, the error rate for them is higher than for electronic machines. That may be true, but with paper ballots there remains a physical record you can check the results against. That's not the case with a computerized system. There's also the issue of severity; a single security issue with a computerized system would, by its very nature, have far more impact than a voter who bungles a paper ballot. It would only take a single exploit to change the results of an entire election--potentially, with no trace that the results were falsified. Quote: Geez, you'd think no one had ever heard of "stuffing the ballot box" prior to the invention of the computer chip. Zep, how hard would it be for someone with a copying machine to reproduce the paper ballots in your post by the thousands? Not very. How hard would it be for him to actually carry out that sort of fraud? Very, without the collusion of the election workers. First, paper ballots are usually specialized card stock, not your average 20# paper you find in your local Kinko's machine. Second, it's usually pretty simple to detect photocopies. Third, you'd have to stuff the box without being noticed (or with the collusion of) the multiple election staff onhand. Most importantly, of course, is the fact that even if someone were to get away with that (and I have no doubt it can be done, it certainly has before), the skewed results would be localized to a single polling place. You would have to involve a conspiracy of a fair number of people at different polling locations to be able to make that happen on a scale large enough to influence an election. And if we're talking about a statewide election--such as for a senator, governor, or President--then forget about it. The logistics of a conspiracy on that scale would be difficult to impossible without someone getting wind of it. Quote: Everything is hackable. But "hackable" does not equal "hacked". This is very true. However, at the heart of this issue, even beyond whether the systems are "hackable" or not, is the issue of accountability and verifiability (is that a word?). We can go back and verify paper ballots; we can verify that they're legitimate, we can verify the count was correct. The system may not be perfect, but there's some accountability there. With an electronic system, we don't have that option. Because the systems are proprietary, we can't even go and evaluate the tabulation methodology. We have to have faith that the results are correct and accurate.
	__________________ "People like me are what stand between us and Auschwitz." - Newt Gingrich

25th October 2006, 08:42 AM	#18
Luke T. Guest Join Date: May 2003 Posts: 14,759	Originally Posted by Cleon Most importantly, of course, is the fact that even if someone were to get away with that (and I have no doubt it can be done, it certainly has before), the skewed results would be localized to a single polling place. You would have to involve a conspiracy of a fair number of people at different polling locations to be able to make that happen on a scale large enough to influence an election. And if we're talking about a statewide election--such as for a senator, governor, or President--then forget about it. The logistics of a conspiracy on that scale would be difficult to impossible without someone getting wind of it. And wouldn't that be true for electronic voting machines? You would have to have a fair number of people involved to make fraud happen on a large scale?
Luke T. Guest Join Date: May 2003 Posts: 14,759

25th October 2006, 09:01 AM	#19
drkitten Penultimate Amazing Join Date: Mar 2004 Location: Wits' End Posts: 21,647	Originally Posted by Luke T. And wouldn't that be true for electronic voting machines? You would have to have a fair number of people involved to make fraud happen on a large scale? No. Just hack the central tabulating faciility. Or, for that matter, hack the source code for the electronic voting machines. That's part of the problem -- single point of failure.

25th October 2006, 09:11 AM	#20
Cleon King of the Pod People Moderator Join Date: Aug 2001 Location: Atlanta, GA Posts: 20,506	Originally Posted by Luke T. And wouldn't that be true for electronic voting machines? You would have to have a fair number of people involved to make fraud happen on a large scale? Not at all, because it's all centralized. You set up a laptop at a local coffeeshop, hack a couple of central tabulating hubs or DBs, and you can affect votes not just all across the state, but potentially any other state that uses the same system. Or you could, as drkitten said, get in early and hack the source code. At which point you don't actually have to do anything the day of the vote; it's already done for you.
	__________________ "People like me are what stand between us and Auschwitz." - Newt Gingrich

25th October 2006, 09:33 AM	#21
varwoche Philosopher Join Date: Feb 2004 Location: Puget Sound Posts: 7,228	Originally Posted by Cleon With an electronic system, we don't have that option. Because the systems are proprietary, we can't even go and evaluate the tabulation methodology. We have to have faith that the results are correct and accurate. The proprietary aspect is SO bogus that it's a (not very funny) joke. A paper receipt doesn't really solve the problem because the receipt is output by the same potentially compromised system. I finally heard a proposal that makes some sense, and that's a digital camera that photographs the voting screen and saves the images for auditing / recounts.
	__________________ To survive election season on a skeptics forum, one must understand Hymie-the-Robot (and/or Fat Jack)

25th October 2006, 09:38 AM	#22
Cleon King of the Pod People Moderator Join Date: Aug 2001 Location: Atlanta, GA Posts: 20,506	Originally Posted by varwoche The proprietary aspect is SO bogus that it's a (not very funny) joke. Do you mean that it's a (not very funny) joke that the system is proprietary, or that it's a (not very funny) joke that I bring it up as an objection? Quote: A paper receipt doesn't really solve the problem because the receipt is output by the same potentially compromised system. True to an extent, however, the voter could eyeball it to make sure that the receipt actually reflected the votes cast. Quote: I finally heard a proposal that makes some sense, and that's a digital camera that photographs the voting screen and saves the images for auditing / recounts. Meh. Possibly, I suppose, but I see lots of room for failure--not to mention privacy issues (I really don't want my picture taken, even just the back of my head, with my voting choices on the screen).
	__________________ "People like me are what stand between us and Auschwitz." - Newt Gingrich

25th October 2006, 09:42 AM	#23
varwoche Philosopher Join Date: Feb 2004 Location: Puget Sound Posts: 7,228	Originally Posted by Cleon Do you mean that it's a (not very funny) joke that the system is proprietary, or that it's a (not very funny) joke that I bring it up as an objection? Pardon me, the former.
	__________________ To survive election season on a skeptics forum, one must understand Hymie-the-Robot (and/or Fat Jack)

25th October 2006, 09:51 AM	#24
geni Anti-homeopathy illuminati member Join Date: Oct 2003 Location: UK Posts: 26,561	Originally Posted by drkitten No. Just hack the central tabulating faciility. Or, for that matter, hack the source code for the electronic voting machines. Or go after the compiler so a review of the source code wont pick up any problems.

25th October 2006, 10:29 AM	#25
Beth Philosopher Join Date: Dec 2004 Location: Flatland Posts: 5,307	Originally Posted by varwoche The proprietary aspect is SO bogus that it's a (not very funny) joke. A paper receipt doesn't really solve the problem because the receipt is output by the same potentially compromised system. I finally heard a proposal that makes some sense, and that's a digital camera that photographs the voting screen and saves the images for auditing / recounts. The best system I've heard of is a dual paper receipt system. Two copies are printed, one goes to the voter so he/she can verify that it is correct and verify that the two are the same. Once verified, the voter keeps one copy, the other goes into a sealed box so that, if and when a recount is done, the paper copies can be used to verify the voting maching tally.
	__________________ Beth "You are not the stuff of which you are made." Richard Dawkins, July 2005, 10:45 http://www.ted.com/talks/richard_daw..._universe.html

25th October 2006, 10:43 AM	#26
drkitten Penultimate Amazing Join Date: Mar 2004 Location: Wits' End Posts: 21,647	Originally Posted by Cleon True to an extent, however, the voter could eyeball it to make sure that the receipt actually reflected the votes cast. Won't work. The voter has no way of knowing that the electronic vote that was recorded in the database was the same as the one printed on the receipt. For that matter, he has no guarantee that the vote machine doesn't happily make up non-existent votes in its spare time to keep itself amused.

25th October 2006, 11:31 AM	#27
Cleon King of the Pod People Moderator Join Date: Aug 2001 Location: Atlanta, GA Posts: 20,506	Originally Posted by drkitten Won't work. The voter has no way of knowing that the electronic vote that was recorded in the database was the same as the one printed on the receipt. For that matter, he has no guarantee that the vote machine doesn't happily make up non-existent votes in its spare time to keep itself amused. Point taken. Though I did just get a rather amusing mental image of a voting machine, sitting by itself in some obscure precinct, deciding to keep away the boredom by playing "Pong" with the votes.
	__________________ "People like me are what stand between us and Auschwitz." - Newt Gingrich

25th October 2006, 02:02 PM	#28
NoZed Avenger Bitter Whiner Join Date: Apr 2002 Posts: 11,313	Voting shenanigans in Chicago? Absurd.
	__________________ [This Space Available. PM for Rates.]

25th October 2006, 02:25 PM	#29
Rob Lister Suspended Join Date: Apr 2004 Location: Virginia Beach, VA Posts: 8,523	Originally Posted by Beth The best system I've heard of is a dual paper receipt system. Two copies are printed, one goes to the voter so he/she can verify that it is correct and verify that the two are the same. Once verified, the voter keeps one copy, the other goes into a sealed box so that, if and when a recount is done, the paper copies can be used to verify the voting maching tally. Yea, I'd love to see that system implemented. If you listen to some people, you've got people too stupid to read the ballot and mark the correct slot. How are they going to compare two sets of receipts? They'd have to be double smart for that! Me? I'm of the opinion that if you're too stupid to vote using any system currently in use, you shouldn't be voting. There will always be fraud attempted and fraud achieved (chad dimple anyone?). My bet is there is going to be less fraud with the electronic system.

25th October 2006, 02:29 PM	#30
drkitten Penultimate Amazing Join Date: Mar 2004 Location: Wits' End Posts: 21,647	Originally Posted by Rob Lister There will always be fraud attempted and fraud achieved (chad dimple anyone?). My bet is there is going to be less fraud with the electronic system. I'm afraid that we disagree. My bet is that there is going to be more, simply because it's much easier and the potential stakes are so much higher. If I can, for example, intimidate you out of voting, I've shifted the vote count by one. If I can program the computer to drop your vote -- and the vote of everyone else who voted as you did -- on the floor, then I have guaranteed myself the election. With the old system, I needed to persuade 51% of the electorate. With the new (post 2000) system, I need to persuade five of nine Supreme Court Justices. With the incoming system, I only need to persuade one programmer.

25th October 2006, 02:33 PM	#31
Rob Lister Suspended Join Date: Apr 2004 Location: Virginia Beach, VA Posts: 8,523	If, if, if. Line them up and you've got a lot of Ifs. I don't care either way. No matter the system.

25th October 2006, 02:38 PM	#32
aerosolben Evil Genius Join Date: Aug 2001 Posts: 2,270	Originally Posted by Cleon Not at all, because it's all centralized. You set up a laptop at a local coffeeshop, hack a couple of central tabulating hubs or DBs, and you can affect votes not just all across the state, but potentially any other state that uses the same system. I would have thought that the machines are individually auditable. Even if you can't get the details, they would at least be able to report what they sent to the central facility. If not, the design goes from extremely stupid to astoundingly stupid. If you aren't going to secure the machine, at least avoid making wide-spread tampering somewhat difficult. Originally Posted by drkitten Won't work. The voter has no way of knowing that the electronic vote that was recorded in the database was the same as the one printed on the receipt. But this is an impossible problem to solve. There is no way for the voter to confirm this - some system along the line could always be lying. However, it does allow you to audit, which is the real goal. A computerized system could report the immediate results, and a hand count could be done as well. Any non-trivial discrepencies would indicate some failure in the system. A (well-designed) computer system could even help guard against fraud in the human side of the equation by producing an independant count. Obviously, it still isn't foolproof, but it is more secure.
aerosolben Evil Genius Join Date: Aug 2001 Posts: 2,270	__________________ You can tell a lot about a fellow's character by his way of eating jellybeans. - Ronald Reagan

25th October 2006, 02:49 PM	#33
drkitten Penultimate Amazing Join Date: Mar 2004 Location: Wits' End Posts: 21,647	Originally Posted by aerosolben But this is an impossible problem to solve. There is no way for the voter to confirm this - some system along the line could always be lying. Actually, Beth's duplicate paper reciepts does a good job. The voter confirms that the reciept as printed matches her vote, and then puts a (validated) copy of the vote in a traditional ballot box. We can do spot-checks of accuracy by confirming that the counts in the ballot boxes are identical to the counts as reported by the machines, which is an easier task than counting all the ballot boxes. Any substantial difference becomes grounds for questioning the validity of the machines and we then use the ballot boxes as traditional paper boxes or have some other resolution method. Hopefully such a procedure will not be necessary very often. For that matter, there are a number of cryptographically secure methods of confirming end-to-end security of votes. For example, one could "publish" a list of votes cast (and their totals), but with identifying information hidden via encryption. Anyone could then confirm that the vote totals were in fact accurate, while each individual voter could re-encrypt her own information and identify her vote in the published list to confirm that it was counted accurately.

25th October 2006, 03:20 PM	#34
aerosolben Evil Genius Join Date: Aug 2001 Posts: 2,270	Originally Posted by drkitten Actually, Beth's duplicate paper reciepts does a good job. The voter confirms that the reciept as printed matches her vote, and then puts a (validated) copy of the vote in a traditional ballot box. I was referring to the electronic count - the voter cannot validate that the system has recorded the vote they expect. What you mention is essentially just paper balloting if you use it as the official count. If it backs the electronic count, then it is auditing as I mentioned, and I of course agree with your points. Quote: For that matter, there are a number of cryptographically secure methods of confirming end-to-end security of votes. For example, one could "publish" a list of votes cast (and their totals), but with identifying information hidden via encryption. Anyone could then confirm that the vote totals were in fact accurate, while each individual voter could re-encrypt her own information and identify her vote in the published list to confirm that it was counted accurately. I could make a snide comment about perfect security and cryptography, but I'll refrain. Instead, I'll say that first, building secure voting systems is not a trivial problem. While it is trivial to beat what we have currently, it is difficult to account for all vulnerabilities even as an academic exercise - a large part of which is avoiding extending trust to any part of the system. Second, the voter isn't really doing the things you describe - the voting machine is, and the voter really has no idea what's going on. Most academic approaches will ignore this aspect of the problem, which is why I focused on it. I still think it's practically doable (especially if focused on major vulnerabilities and made some limiting assumptions about the problem environment). I think we're largely on the same page here.
aerosolben Evil Genius Join Date: Aug 2001 Posts: 2,270	__________________ You can tell a lot about a fellow's character by his way of eating jellybeans. - Ronald Reagan

25th October 2006, 03:31 PM	#35
Luke T. Guest Join Date: May 2003 Posts: 14,759	This is the age of the ATM and computers. We should move in the direction of electronic voting, not regress to paper ballots. Work out the bugs. Progress. Welcome to the 21st century. And if anyone knows how to hack an ATM, PM me.
Luke T. Guest Join Date: May 2003 Posts: 14,759

25th October 2006, 03:32 PM	#36
drkitten Penultimate Amazing Join Date: Mar 2004 Location: Wits' End Posts: 21,647	Originally Posted by Luke T. This is the age of the ATM and computers. We should move in the direction of electronic voting, not regress to paper ballots. Work out the bugs. First step in that is to identify what the bugs are, and to fix them. We don't need to deploy the technology until it actually works, you know.