I voted today.

[Bill Bored]

On a NY lever machine! And it felt SOOOO GOOD!

I flicked the little lever next to the candidate's name, and the X appeared in the box. I felt quite a bit of resistance. Then I voted the second race, and it felt kind of weak. So I took back both my votes, and reversed the order, voting for the second race first, and then the first race. And what do you know? This time the second race had a nice solid feel and the first one felt weaker. Whatever was going on inside that machine, it was very consistent! The first race you vote for, trips some kind of widget that makes the second one easier to to vote for. At least that's how it works if both votes are in the same party.

Now I know these machines aren't perfect, but they don't switch votes in the middle of casting a ballot unless I say so! And if there is a question about what's going on inside, there's nothing proprietary in there. And it only takes half a minute to vote on them! They can handle 800 voters per machine on Election Day!

And I dare anyone to try and hack 1,000 of 'em or more at one time like you can with DREs from a GEMS server that also switch votes in the middle of casting the ballot, can't be examined internally, and take f$^%$ing forever to vote on!

Yeah I voted today. And it felt SOOOO GOOD!

For those of you who have never experienced a mechanical lever machine, you just don't know what you're missing. It's a thing of beauty. HAVA-compliant too of course!

I'm looking forward to a runoff, just so I can do it again in two weeks!

[Wilms]

1. Survey Says! Levers most accurate

http://www.notablesoftware.com/Press/Chronicle/Chronicle.htm#vote

[NYC]

3. Thank you.

I agree with Bill Bored. I love them. And when I'm finished, I always review very carefully. :)

[foo_bar]

6. my primal voting experience was a lever machine

My dad paused for a second before pressing down on Carter, then flicking every local race into the (D) position with one arpeggio motion. That's how baby Democrats are made.

NY politics is over my head. I read somewhere that I hate Bloomberg with the fury of a thousand suns, but he endorsed the guy that my neighbor's against, and she never picks up the garbage.

[Bill Bored]

9. Believe it or not, I think I know what you mean! nt

[Bill Bored]

5. That's amazing.

And that BIG lever at the bottom is pretty cool too!

There was a scene in "Hill Street Blues" when the S.W.A.T. guy said it really got the adrenals going to pull that one!

Touch Screens are for GIRLY MEN! :puke:

[NYC]

2. How was the turnout in your area?

Not so good in midtown. I hope more people are going now.

[Bill Bored]

4. Looked pretty good for an afternoon in Queens. nt

[garybeck]

7. there are stories about lever machine fraud too

I've heard stories about them being rigged with paperclips and things like that which prevent proper votes from being cast. don't want to burst your bubble, but it's important to acknowledge that any system can be tampered with if the wrong people have access to the machines, regardless of what type they are. that's why I think that VVPB and MMRA are the essential formula to fair elections, with any type of machine.

if the voter verifies the ballot, and there are random manual audits against the machine count, it is very difficult to rig an election.

the same principles are used in banking and other systems that require accuracy.

last week I got a call from my credit card company. they were doing a random check to make sure I had made all the recent purchases on my card.

[Bill Bored]

8. Random audits are limited though.

Concentrated fraud would likely not be detected.
But they're ok if the election isn't too close.

The leading candidate is about 200 votes short of avoiding a runoff in the NYC Dem Mayoral Primary so far tonight, out of 500,000 votes! There ain't no system that can't be rigged enough to swing this one! Good luck finding a 100-vote swing out of 6,000 precincts with a random audit!

I'll take lever machines over anything "programmable" any day. Computers are better at replicating mistakes than any technology known to man.

I can't argue with a system that puts in place the necessary safeguards, but we are so far from that it isn't even funny. I think the levers may offer the best solution until we throw private companies who are not beholden to the voters out of the election business.

I think Op Scans can be made safe but only if the powers that be have the will to do it.

[manwhoeatsrats]

10. Republicans can't vote outside of their party????

Okay, a friend of mine is a registered Republican, and did not vote in the last election because she did not want to vote for Bush. I asked why she did not vote for Carry, and she stated that here vote would not count because a registered Republican that votes a democrat will not be counted for voting outside of the political party. No matter what I say she is determined that this is correct. Is it? This is the first time I have heard anything like this. Can anyone provide some kind of a URL, or something that states this, or states that this is a missconception? LOL have any of you even heard about this?

[OnTheOtherHand]

12. oh, wow

I dunno whether you would find that on one of the urban legends websites.

She may be confusing general elections with primaries (New York just had its primary elections). Here in New York, I think you have to be registered with a party for a whole year before you can participate in its primary elections.

I can think of other things she _might_ mean, including quirks (or worse) in some of the DREs. Some machines are set up to make it easy to vote a straight party line -- although no machine is supposed to make it impossible not to!

In principle, in a general election, it doesn't matter and cannot matter which party (if any) you are registered in. People often vote across party lines.

[Bill Bored]

13. Welcome to DU. First of all, know this:

E-voting machines can be programmed off the shelf to do almost anything, including exactly what you are suggesting. But that would be illegal in every state that I know of except in primaries in which you have to select your party before you can vote. That doesn't mean it hasn't happened of course, but it's not the sort of thing anyone would want to advertise to the voters, so I'm surprised that your friend is actually saying this, unless of course she knows something we don't!

Does she work for the RNC or Diebold, or ES&S or Karl Rove?

And BTW, it's spelled K e r r y.

[garybeck]

11. audits

I don't think you can generalize about the effectiveness of audits. it depends on the sample quantity, and the tolerance level. statisticians can predict with very good accuracy how much of a sample will provide how much of an assurance that the results were correct. that's how TIA predicts millions-to-one that the election was stolen. exit polls are an example of auditing.

if there is an adequate sample and zero tolerance (meaning if the audit is off by just one vote, a larger sample must be hand counted), I have confidence in the system. of course this assumes that the hand count is using voter verified paper ballots, not some hidden audit log that isn't verified by the voter, or a tiny thing that has to be viewed with a magnifying glass (the diebold dre's).

even in a close race, i feel audits play a key role in ensuring the counting was done properly. I believe the closeness of the race is actually moot. The question is simply whether or not the audit is off from the original count.

I think your scenario of finding a 100 votes swing where there are 6,000 pecincts is missing the point. how do you know there is a 100 vote swing? that is just the official result. the point is that if there is fraud, the real result could be much more than 100 votes, or less. the difference in question is not the difference between the winner and the loser, it is the difference between the official result and the true result. Two completely different things. When the true result is unknown, it is missing the point to say that an audit couldn't catch fraud because the election was so close.

in solidarity,
gb

[Bill Bored]

14. Gary, Please see this paper by Kathy Dopp.

It's not about exit polls, but it's REALLY IMPORTANT!

<http://uscountvotes.org/ucvAnalysis/US/paper-audits/Paper_Audits.pdf>

The initial margin does matter because in order for the outcome of the election to be fraudulent (i.e., incorrect), you have to be able to show that the correct margin is less than zero, i.e., that the true outcome is reversed. In order for that to happen, you have to show that there could be enough corrupt machines in the jurisdiction to have swung enough votes to reduce the margin to -1 vote, thus reversing the outcome.

So, first, you need to know how many votes would swing the election (1/2 the vote margin + 1). Then you need to know the minimum number of corrupted machines that could make this happen. In the example of the NYC Primary above, it would be one machine out of 6,000, or 0.0167% of machines. Then you have to either rule this out by auditing, or prove that it exists. So how do you find 1 corrupted machine in 6,000 by auditing 3% of them? That means you'd only look at 180 machines! What about the other 5,820 of 'em?

See what the odds are of finding the 1 corrupted machine using Kathy's spreadsheet. Not too good I'll wager.

What Kathy has done is to show how much auditing is actually required to do this, and it's clear that if you have a higher margin to start with, and the election is fraudulent enough to have reversed the outcome, you would have more corrupted machines, and a higher probability of finding one with a random audit.

With a close election, it's a lot harder to find the small number of corrupted machines required to reverse the outcome.

See the above paper and try the associated spreadsheet to see some real world examples and get back to me.

Solid!