HAND COUNTED paper ballots YES! Sourcecode anywhere-NO!

Cheap, easy and reliable. What's wrong with this picture?

You might want to consider the thread http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x456594

Particularly from post #38 on down. We are shown a hand-counted ballot from Scotland:



There is ONE vote per ballot to be tallied in Scotland.

Then here is a ballot in California with 53 separate items to be voted on.



In just one county in U.S. there are 3 million registered voters, and in Nov. 2006, 53 items on the ballot. 156,470,999 votes to count.

Say only 2 million voters show up in that one county, ONE HUNDRED MILLION separate races/issues must be counted.

We are given an approximate number of 30 counters per 30,000-50,000 votes in Scotland.

At that rate, we would need 60,000+ vote counters in ONE California county.

I would be interested to know from the All Hand Count proponents how any one county elections office can manage 60,000 one-day workers/volunteers... screen and train them, supervise/oversee them, etc.

Looking to see more than empty posts merely ranting for "all hand counted ballots" with a bunch of exclamation marks. I would like to see some practical explanations of logistics... any kind of practical knowledge of how such a thing has been or could be done.

How many voting items were on your ballot? How many voters in your county? Have you spoken to the head of your elections office to figure out the logistics of hand counting? How many counters? Supervisors? Have you talked to your county about their standards and regulations about election workers, whether paid or volunteer? Does your county have problems, as so many do, getting poll workers as it is?

"cheap, easy and reliable"... how do you know that? How "cheap" is it yo bring this number of people in for a day of training and a day of counting? Are they volunteers or paid? Have you asked your county? How "easy" is it to recruit, screen, train, track, oversee and supervise these people? "reliable"? All systems have some fault or risk.... You might note in the other thread that hand counts don't guarantee an honest and accurate count, either. Nor do ballots by mail.

Looking forward to answers (rather than just the Rah-Rah cheerleading) relating to the actual practical and logistical problems/questions.

There are a couple of keys to being able to efficiently and reliably hand count.

The races must be organized on the ballots so that the national and statewide issues are printed separately from other races and issues. (For example, president and senator might be together, some statewide races and issues together, others grouped differently.) Use different colored paper to avoid confusion and maybe have different ballot boxes for each ballot type.

Volunteers from all parties show up as the polls close to do the counting. They are divided into teams with at least one person from each party. Each team is assigned to count a group of ballots. When they finish their tally (with everyone in agreement), the same ballots are counted by a different team, possibly using a different counting method. One team might have called out the result for each ballot with two people keeping tallies, while the second might sort the ballots by winning candidates and count the piles a couple of different ways. If the independent counts by teams don't agree, another count is done.

This works best when the precincts don't get too large. I remember that we could finish a precinct where 1-2 thousand people had voted in 2-3 hours. We would do the national and statewide races first and phone those results in to the county where they were combined with other precincts. A typical election might have 3 US races, 15-20 statewide races (NC elects counsel of state and judges), 10-20 county offices (commissioners, school board, city council, clerk of count, judges), plus a couple of initiatives, bonds, or ammendments. A total of 30-40 separate races to be counted, more or less.


The cost is held down because most people counting and observing were volunteers. Anyone could participate, at least as an observer/witness. Teams seemed to self-organize quickly and produce reliable counts. No one wanted to get it wrong. We received brief instructions before we started (maybe 10 minutes), but the majority of people had done this many times, so the newcomers were distributed among the teams.

No, we didn't have final results on all races until after midnight, but we didn't have races where no one would ever really know who had received the most votes.

And the first shot at determining a voter's intention was done in front of everyone, usually as a concensus. The ballots were returned to the ballot boxes and sealed. They were taken to the county seat and retained in case there was a recount. All the tally and summary sheets were also retained.

Updated to today, maybe rather than phoning in the preliminary counts we could use a web site, but with the precinct tally sheets providing the canvas numbers.


Satisfied? Anything you don't understand about this? My experience was just as a volunteer, it was in North Carolina in the 1970's into the 1980's. There probably many others on DU who have done a variation of this and were one of the supervisors.

but I am not convinced that what you are describing is actually secure. You say:



If all that is true, there is no problem. But I see no way of guaranteeing that it is true, any more than there is any guarantee that the people in charge of programming e-voting machines are honest. Anyone could participate. Teams seemed to self organize quickly and produce reliable counts.

But how do you ensure that honest people actually do participate? Or that teams actually do produce reliable results?

We do pretty well ensure those things in the UK, but we do it by dint of holding the counts in big halls or school gyms, and having a great many people looking on. I do not see how you could ensure that this happened in every US precinct, and it would, as we know, take only a small proportion of corrupt precincts to swing an election.

In other words, hand-counts will work if people are honest. But so will any form of count. It's checking the accuracy and honesty that is the essence of a fair election, not hand-counting per-se. What I really like about your description is the care with which the ballots were apparently guarded. A secure chain of custody of the ballots seems to me to be an absolutely key element, however they are actually counted.

ITs been some time, but I called each of the 3 counties in NC
that had hand counted paper ballots.

Hyde, Tyrell, and Graham.

The election officials were proud of their accuracy,
but advised me that they finished counting the 2004
Presidential election about halfway through the next
day.

It was not easy.

With so many items on the ballot,
with some people choosing straight ticket,
some not choosing straight ticket
some crossing parties while choosing straight ticket
some crossing over in multi seat contests thereby invalidating their other multi seat choices

Anyway, in 2004, it took till mid way through the next day to get finished.

In NC, you can vote "straight party" on a ballot (for example for governor,lt. gov, attorney general, ...) and then override this by explicitly crossing parties in some of the races. In multi-seat contests, I believe you would have to explicitly vote for candidates of the "straight party" as well as any in the other party. Whatever the state rule is, it would also apply to the other ballot types.

I can't remember ever having a multi-seat contest that was included with other contests and which allowed a straight party vote. It is messy enough that when tallying, if it is straight party with overrides you have to tally the individual contests separately as if the marks were there.

2004 in NC took nearly one year to complete the count of statewide offices and seat the last officeholder (Sec. of Education).

In 2006, the last votes on scanned or DREs took a couple of days, compounded by things like a precinct being redistricted into two different state house districts and the wrong ballots being given to a lot of voters.

The county procedures for counting the ballots have traditionally not been uniform and I can't speak for those counties without doing some research.

for compilers verified by trial counts before and after election

Personally, I think that a good audit regime makes optical scan very viable -- and if you can't get a good audit regime, you can't get good 100% hand counts, either. It isn't as if elections were fraud-free prior to the advent of electronic voting.

I don't know whether open source code really solves any security problem.

I understand, sort of, why some people think nothing less than 100% hand counts will do. It's just one of those things.

You said:



If a multi-seat contest is partisan, than straight ticket vote does count for it.
Take a look at the contest for "County School Board" on this ballot.
http://www.forsyth.cc/Documents/2006General_Election_Ballot.pdf

Your straight ticket counts in any partisan multi seat contests UNLESS you cross over.

Here are the straight ticket voting instructions on
North Carolina ballots:



You also said:



It didn't take a year to complete the count, the count was never completed
for State Commissioner of Agriculture. Britt Cobb conceded.

The loss of over 4,400 votes on a paperless machine in Carteret county
helped us greatly in getting our verified voting law passed.

We had already warned the SBOE and the media that such a loss could
occur if they continued to use paperless machines.

In 2006, the election results were in on election night, and yes
we did have audits, and later we had recounts across the state, and the
contests were settled. Not a year later. (Not that I am not willing to wait)


Countries that HCPB -

-don't vote for president, their parliament does
-have single item ballots, not 50+ item ballots
-have several elections a year]

So, HCPB does not just require eliminating the machines, it
requires completely changing all of the electoral system.

We had Chuck Herrin present testimony, which included his recommendation
for HCPB, to our lawmakers, and some of the lawmakers present had HCPB back decades ago,
so they know what it entails.

To get HCPB nationally?? requires persuading the lawmakers and election officials
that they want to change elections drastically,completely, and not just the
mechanism of counting.

After several court challenges, it was finally settled by a vote of the General Assembly.

It was the last ballot contest in the nation to be settled and the winner to be seated.

I am completely aware of the NC GS, and read the entire election chapter again before replying to you.

I was not disagreeing with how a partisan multiseat contest would be counted. What I stated was that I couldn't remember actually having hand counted a partisan multi-seat that was also included as part of a straight ticket ballot. Those contests would have been for things like county commisioners and they were on separate ballot sheets from the statewide races where one could vote a straight ticket.

The Superintendent of Schools' contest was not thwarted by
the voting machines, but by a lawsuit over Provisional
Ballots. That is another topic altogether.

That is why the state legislature was tasked with
making the decision.

The longest contest in the state, and maybe the country -
was the County Commissioners race in Guilford County NC,
Wade/Parker contest.



WHAT HAPPENED? N.C. Supreme Court threw out 11,310 ballots cast outside the voters' home precincts.

WHAT'S NEXT? Case goes back to Wake County court, then counties count the improper ballots and them from November vote totals.

WHAT'S AT STAKE? Election of state superintendent of public instruction remains unsettled and questions raised about several local races statewide

Since the court challenges were about whether certain ballots were to be counted or not, then the "count" was not finalized until after all challenges. Atkinson was not the Secretary of PI until after the General Assembly vote.

And the point of that comment was just to reinforce that no matter what counting methods were in use, not finishing the counts of some contests for 18-24 hours is not really a problem; they already can and do take weeks, even months, to be finally resolved.

UNDERSTAND THAT! WE know what is going on, Soon the MEN will take control of this secret vote counting BULLSHIT, And it will be over.

We have had enough of their goddamn @^$42$^&&^ Games!!!

The women did their part. Its time for the men to stop acting like boys, and call a duck a duck...........a twelve year old knows better than to count the votes in secret.

such lengthy and complicated ballots.

how many races (roughly, on average) do you vote for in each election?

where we are allowed to amend our constitution fairly easily so the ballot can get up to four pages long. But most elections are a page or two.

But our ballots are a single page containing one race. The only reason we need a page for that one race is that we often have lots of silly candidates. But when the votes are counted, all that needs to happen is for the ballots to be sorted into votes for each candidate. The silly candidates will have very small piles.

unc70 has described how ballots with multiple races on them can be counted by hand. But it is very much more complicated than the way we do it in the UK, and takes a few hours to count a few thousand votes ballots. In the UK it takes a few hours to count tens of thousands of ballots, which means we can count in far larger units (constituencies). This ensures good oversight.

It is the oversight protection I am not seeing in the HCPB systems proposed for the US. I think you would still need rigorous random audits. And if you had rigorous random audits, I'd actually be more confident of the accuracy of scanned counts than hand counts, I think. But that would depend on the chain of custody of the ballots. But the validity of all audits will depend on that.

as long as you can run the ballots through two different machines and compare the totals. If they match, you're good to go. If not then bring in the hand counters to resolve the discrepency. Voting is the only accounting system I know of that doesn't count the numbers twice.

My voting place in Nov. had hand-marked paper ballots, placed into the opti-scanner by the voter at the polling place. Tallies then posted at the end of the day.

The ballots should then be scanned again at the main elections office (with public witnesses) on a different machine, and if tallies don't match, hand counted.

Yessiree, I think that does make sense.

It's not that complicated to add 1+1 over and over and come up with the same total every time.

One of the things I do in my job is score questionnaires, which is essentially what your ballots are. Admittedly I'm not very good at it, but I can easily come up with a different total every time.

It's not that I can't add one and one, but that it is very easily for the eye to slip to the wrong part of the page.

But that's not to knock hand-counting. I think it's a good idea, and I think manual random audits should be mandatory. But if machines are working as they should, I'd expect most discrepancies between manual count and machine count to turn out to be due to error in the hand-count. However, I'd be surprised if the hand-count was ever out by more than a few votes, whereas it wouldn't surprise me at all if the machine was out by orders of magnitude more.

The real point is that there need to be independent random checks on the accuracy of any count. It makes sense to me, given your complex ballots, to use optical scanners as the primary counting system, running them through two different scanners if you like. But then I think you need to conduct random checks on the scanners, using a completely different method, preferably a manual count. And if a single machine is found to count outside its specified accuracy, I'd call for a complete manual recount of the entire race.

they say you are only allowed to count the ballots once, which is so ridiculous, it's criminal.

What a daft law.

was all about. Gore wanted to 'recount' the votes. They had only been counted once and when it was apparent there were some major anomalies, he wanted them counted 'again'. The Repugs fought tooth and nail to prevent that from happening.

...that a legitimate hand-count may be off by a few, but machine count may be off by large numbers. DoYouEverWonder's comment regarding op-scanning ballots on two machines with random hand counts is a logical idea for a double-check.

I still worry gravely about hand-count as a main system in this country, particularly if it is instituted across-the-board immediately. Partly because of our lack of experience in launching such a project, and in the actual logistics of procedure itself. Many areas have had a difficult time just recruiting basic poll workers (many who have traditionally served are aging/passing). What recourse is there if we simply do not have enough willing citizens?

An experimental sample in various areas across the country might show us how it can work and what problems there are. This can certainly be done in conjunction with op-scan. But to suddenly go all hand-count seems dangerous to me.... we simply do not know if it will be a fiasco or not.

Can't help it... I feel very leary of one party putting on a big push to get their own followers into some precincts. We have seen certain elections boards where "dems" are actually life-long republicans who simply register as dems, and the "bipartisan" claim is a sham.

Some areas are extremely concentrated with one party. Don't want to say too much about my neighborhood poll-workers, but I am sure not sorry they aren't handling my ballot. I don't even like that the poll register book lists my party, as I am treated like Typhoid Mary, while my (GOP) neighbors are warmly welcomed. How carefully would they treat my ballot if we don't know that the process is overseen scrupulously? So I worry about the 'local precinct' angle.

I find the posts of hand-count experiences very interesting and worthy. But I remain of the idea that op-scan (with DoYouEverWonder's note of scanning on two different machines) along with rigorous audits and sizeable random recounts to be a safe and practical course.

Very interesting discussion. Most of it (with obvious exception), anyway.

Counting votes is just about the simplest thing a computer can be used for.

Regarding security: the ability to open a lock should depend on having the key - not on knowing how the lock works. Open source software such as Linux is not known for its weak security, even though anyone can easily find out how the locks work.

So there's no good reason to not make the whole thing, including the hardware, open source.

They both have stated unequivocally that we need an exit strategy from electronic voting and counting. They maintain that at minimum we would need a 10% audit of every precinct and even then it wouldn't be 100% accurate. They both strongly advocate paper ballots, hand counted. Read O'Dell's articles at opednews.com.

Our election system was set up in small precincts DELIBERATELY so as to allow our votes to be counted by hand.

you are saying that Bruce and Jonathan "disagree 100%" with.

I wrote:



Which is simply true, as is this:



As for this:



can you tell me why Bruce and Jonathan might disagree that rigorous random audits would still be needed if you had HCPB? Or with the statement that the validity of all audits would depend on the chain of custody of the ballots? Or did you respond to the wrong post?

I tend to agree with Bruce and Jonathan that hand-counted paper ballots would be an ideal solution, although I don't expect that they would disagree with me that even then, rigorous audits and secure chains of custody are essential. I don't actually think that their audit protocol is the best one for monitoring machine counts, however, and I actually think that if machine-based audits were mandatory (entire machines audited), together with mandatory hand recounts of entire races were a single machine to fail, you would very rapidly see the end of DREs. Whether that would eventually lead to complete manual counts, I don't know, but at the very least it would be the end of machine counts that cannot be manually checked, ie. it would mandate paper ballots.

I think you would still need rigorous random audits. And if you had rigorous random audits, I'd actually be more confident of the accuracy of scanned counts than hand counts, I think.

You are allowing for scanned counts. They want hand counts.

I'm not "allowing for" anything. I'm just saying that I suspect that an optical scanner in good working order may be less error prone than a group of hand counters. But that is not an argument for or against hand-counting. It's an argument for auditing, whichever you use.

The auditing issue is a particular part of O'Dell's discussion, though. He feels that even 10% of all precincts isn't enough, which leads him to the conclusion of hand counting period. I don't recall what exactly he said about audits in that scenario, however. Seems to me that if you hand count you don't need audits since audits are by definition hand counts.

hand counts can be in error, and they can be corrupt. If you don't have random checks, you have an invitation for corruption, and you can't check things like the efficiency of your counting protocol. The important things about an audit is that it should be independent and (conducted by different people), random (so you don't try to second guess where the problems might be, and miss something) and unpredictable (so people can't cheat).

The thing about the percentage is that a percentage isn't what matters - it's the sample size. A small percentage would be good if it came to a large number of precincts, whereas a large percentage could be useless if it came to a small number. The problem I find with Bruce's audit system is that it doesn't audit complete precincts or machines, so there is no actual check on any one counting unit. Even with hand counts you'd need to audit precincts at random, and re-count all the votes in that precinct. It would be like doing random checks on people's tax returns. Bruce's wants to audit all precincts, but only a sample of ballots from each. I can see where he's coming from, but I don't think it's the right approach, myself.

What I recall reading is that he recommends dumping machines altogether, because even a 10% audit of all precincts wouldn't be enough. Therefore, he wants to move to hand counting only.

But I don't know what he recommends, if anything, about auditing in a hand-counting scenario, do you? It sounds like you believe he is recommending random percentage audits rather than auditing entire precincts. Is that what he says?

the EDA audit protocol was to hand-count a proportion of ballots in every precinct, rather than every ballot in a proportion of precincts.

I don't know what Bruce's views are about audits if all ballots were hand-counted. As his expertise is computer security, he may not have thought about that, but I don't know.

My own view is that the first step is to mandate paper ballots (i.e. make it unlawful for a vote to be cast without a voter-verified paper record that has the status of a ballot), and to make a good audit protocol mandatory (and I favor a manual recount of every ballot on in a random sample of either precincts, or machines, the sample size being appropriate to the race being audited), and to have failure of such an audit trigger an entire manual recount of that race.

But I'd want to keep the audit protocol, even in the case of hand-counts. The trouble is that if people want to steal a hand-counted election, there other ways to do it than miscounting - spoiling, losing, or adding ballots - that wouldn't necessarily show up on an audit. Making elections secure is difficult, and hand-counts certainly aren't a panacea.

He reiterates his position from earlier papers on opednews.com re: 10% auditing and goes on to argue completely against optical scanners--so what auditing position does he take on hand counting? Hand counting is hand counting. There is no auditing. And, on edit, I add this: what is an audit other than a hand-count? Is there then an audit of the hand-count audit? I guess I don't get how auditing the hand counting is going to help and it seems to me that would really invite arguments against hand counting if you are going to make election workers recount the count.

"As a consequence, Jonathan Simon and I have shown that to even consider using optical scan devices, you need a secure hand count audit of a least 10% of the ballots in a congressional election. And potentially more to protect elections with fewer than 150,000 voters. You need to perform this audit whether it's Diebold's Jeffrey Dean, or Avi Rubin, or Alan Turing himself come down from Heaven, who writes the tabulation code. Because you simply don't, can't and never will be able to know - with sufficient certainty to possibly throw away the American Republic - what each of those thousands of optical scan devices are actually doing unless you check their output. By hand. But ignore the software for a moment.

What. Is. The. Point. of. Optically. Scanning. Ballots?

Cost? Voting is national security. Design, buy, count and secure a decent paper ballot for every election in the US you care to protect for less than the tab for one week's unprovoked military conflict. Speed? Come off it. Canadian federal election results are known with certainty by midnight. Canadians are worthy people, but they possess no magical powers that enable them to count to 500 or so, in public, with repeatable accuracy and to all parties' satisfaction in a reasonable amount of time.

Ability to infer voter intent? Let me get this straight: it's somehow a good idea to substitute a self-correcting collection of multiple human brains (each one with the processing power equivalent to thousands of conventional computers) with a device dumber than a cockroach; in practice, so limited in its abilities that we have to severely dumb down our ballots for the poor little things to even have a snowball's chance of interpreting voter intent. ("Be careful to fully fill in the oval. Don't go over the edge, or it doesn't count. You must draw a dark line precisely joining the other two lines next to the candidate's name..."). Surely some of the brilliant user interface designers in the IT industry could come up with a paper ballot that would be designed for people and not for machines; for accuracy in recording voter intent and suited to public counting of votes by hand... rather than foisting ballots on people that look like 19th Century newspapers.

In my opinion, one of the most important ethical obligations of a computer professional is to inform the public when automation is an inappropriate solution. Since op scan tallies always have to be verified by hand counting, why invite the machines to the party in the first place? What is the problem that "open source" voting software solves? What aspect of "open source" optical scan tallying supersedes the civic benefits of restoring trust in elections by entrusting election administration to the citizens themselves?

Can we swallow our collective technical pride for once as a profession, and just say "no" to such an utterly inappropriate use of technology?

for Bruce's position.

But there is an answer to his question:

The point of optically scanning ballots is that some of your ballots are extremely complex. In the UK, Scotland is changing to a Single Transferable Vote system for its regional parliament. To make this possible, they are bringing in optical scanning, simply because handcounting STV ballots would be too difficult.

I'm not saying it's impossible, just that the alternative isn't obviously better. That's why I think audits are the way forward (even as a step towards hand-counts), although, as I've said, I don't think the EDA audit protocol is the best one.

I believe that is what they do in Canada, for instance. Perhaps there is a blended approach that would work: Employ scanning with aggressive random hand counted audits in lesser races but otherwise employ hand-counting exclusively. How would you approach auditing in exclusively hand-counted races?

Here is what Fritakis reports:

http://www.freepress.org/departments/display/19/2006/1923

(SNIP)

There have been two recent efforts to promote an HCPB system in the United States, and a third will take place later in 2006. In 2004, voting rights activists Sharona Merel, Kaen Renick, Ellen Theisen, and Kathleen Wynne proposed federal legislation for federal offices. In 2005, four voting rights activists (this writer and three members of CASE Ohio – John Burik, Phil Fry, and Dorri Steinhoff) began work on a protocol for HCPB. Some of this writing has been modified and is included in this paper in the specifics for HCPB. In November 2006, voting rights activist Joanne Karasak plans to promote a state constitutional amendment for HCPB in Ohio. There are 18 states where such constitutional amendments are possible.

The key elements of an HCPB system are as follows: (1) Electronic voting machines are not involved in this process in any way whatsoever. (2) Nothing used in an HCPB system is purchased from companies or vendors who have ties to partisan political groups or parties. (3) Each voter hand marks a sturdy paper ballot with a black felt pen provided at the precinct. (4) The counting process happens at each precinct immediately after the polls close. (5) Each ballot is hand counted by registered voters from that precinct in full view of other registered voters from that precinct. (6) The counting process is filmed. (7) A chain of custody of the ballots and ballot boxes is specified. (8) Ballot boxes are observed and filmed as they are opened and closed and move from place to place.

Three categories of registered voters are included in this process: the official counters, the official observers of the counters, and the public watchers of the counters and observers. The hand marked, paper ballots are hand counted in full view of the public in each precinct by a specified number of registered voters in that precinct – e.g., four, six or eight voters. Half of the counters will consist of one person from each party on the ballot, chosen by the party itself; the other half of the counters will consist of registered voters, chosen by lottery. The hand counting is observed by the same number of registered voters (e.g., four, six or eight), and chosen in the same way as the counters. Counting is filmed by a video projection unit; a process will be set up to determine how the videotaping unit will be selected. The videotaping will be broadcast over closed-circuit TV and streamed over the Internet while the counting is happening. All watchers may also videotape and/or take photographs.

...

Even with all these safeguards in place, the chance for fraud still exists. Therefore, immediately after the first count, there will be a 100% hand counted audit of the vote, carried out in the same way as the first hand count, but in the audit, the observers will be the counters and the counters will be the observers.

Although I think it is very similar to the UK system, in which there are very few races, usually one. In the UK occasionally there is more than one race, in which case there is a separate ballot for the other race.

I do think this is not well understood in the US. In the UK our elections are for ONE race. If there is more than one, it is described as another election taking place on the same day. Most officials are not elected, and we don't have an elected judiciary either.

I lived in Canada for 7 years, and my understanding is that the system is essentially the same.

I added some info from freepress.org

if it could be implemented, that would be great.

And the one advantage it would have over the UK system is that the counting units would be relatively small (about 1000 ballots, fewer in many places).

Against that is the fact that each ballot may contain many votes, so I imagine it would be not less time consuming (especially with the 100% recount) than a UK constituency count (1000 votes for many races, as against around 30,000 - 40,000 votes for one race).

But more importantly, you have to multiply this scenario by several thousand precincts for each state. In the UK there are only a few hundred constituencies (and therefore counts) in the entire country. And our population is about a quarter of yours.

Sure, we have overseers and TV cameras at each count - they are public spectacles - but there are very many fewer per head of population.

I'd love to see the Fitrakis plan work. But it's going to involve an appreciable proportion of the population, if it's going to be done properly. And if it isn't done properly, it will be corruptible.

National and provincial elections in Canada use hand-marked, hand-counted paper ballots, and Canadians expect (and get) precise counts of voters, votes, and ballots with no discrepancies.

Though on wiki I read that some provinces do use e-voting. There is a healthy election reform movement up north, apparently, as well.

that Bruce is one of my heroes, a man of utter integrity. I think this is one of the best pro HCPB pieces I have read.

What I'd like to see is actually workable proposals for HCPB, in the US context.

I don't know him but judging by what he has written and done, I can tell that he indeed is a man of utmost integrity. I have encouraged my SOS to be in close contact with him (we have a new, Democratic SOS in my state who is very pro-reform) as he does also live in our state.

There may well be answers, but I am not seeing them.

When we have more than one race (and there is rarely more than one race, and very rarely more than two) we have a separate ballot paper for each race, each on different coloured paper. The first thing that happens is that the ballots are sorted into different races.

Then the ballots for each race are counted, separately. Votes for each candidate are stacked in piles of 50. This is done in front of bipartisan scrutineers, members of the public, candidates and TV cameras. Mistakes are made, but it would be difficult to get away with deliberate mistakes. The whole thing is done in a festival atmosphere, and culminates in the announcement of the result. All candidates have a vested interest in an accurate count. If the count is close, the whole thing is redone, although of course it is a bit quicker second time round.

A precinct level count would be nothing like this. Instead of a few hundred constituencies being counted in the whole nation (and in terms of population, the US is only four times bigger), you would have several thousand in each state. How easy would it be to get hundreds of people to witness each count, as happens in the UK? How likely, in fact, would it be that only a few people turned up to watch? And how easy would it be to check that those few people weren't actually conspiring to corrupt the count?

Sure, hand-counted paper ballots work for us. But it's not the hand-counting itself that ensures an accurate count. It's the fact that it is done quickly and publicly, and checked at every stage. Even the checking is checked. I hope that if we move to optical scanners (which I think is inevitable if we want a more representative electoral system than our current first-past-the-post) that all those checks will remain in place. I see no reason why they shouldn't. But you'd have to run checks on the scanners just as we know run checks on the piles of votes.

If want a system as uncorruptible as ours you have to do more than simply post pictures of hand counting in Scotland. You have to demonstrate how that system could possibly be applied to the US.

And our system isn't even uncorruptible. It's easily corruptible. But election thieves in the UK have to use a different method - they steal ballots. Which are then counted with a high degree of accuracy.


Here's a nice series of images that may help you see just how simple our system is (Cambridge this time):

http://www.lucas-smith.co.uk/photos/elections/


It's a simple system because we have simple ballots.

My experiences at the precinct were like a small version of what you described, but with only 30 or 40 people. It was all happening in a large room with tables for the counting, viewable by everyone. There were representatives of the media, at minimum a stringer with the local newspaper or radio station who also contributed to AP wire service election reporting.

Since those involved were mostly from the precinct itself, many knew each other well and there was often good-natured banter among supporters of various candidates, particularly for local races.

My earlier post might have implied that this was a disorganized situatation (or self organized). The precinct had its own "election board" (I can't remember the actual term we used.) under the county election board. This group, with representatives of both (then) political parties, were legally responsible for administrating the count. They trained and supervised the rest of us. While anyone could observe the count, I believe there might have been a restriction that the volunteers be voters from the precinct. We each signed the tally sheets for the counts we personally participated in. We were able to count all the races (30-40) for 1-2 thousand voters in a couple of hours. Once we got started, it went rather quickly.

The self organization of teams I mentioned was in regard to which counting sub-task one did (e.g. reading the results for multi-winner races, recording tallies, observing). Some people are just better counting or more careful recording or better at math. Most people avoided taking a task where they would be uncertain of their competency. For example, observing the ballots as someone else read the candidates receiving votes (while two others tallied) was an important task which almost anyone could feel competent to do.

I believe that sheriff's deputies were used to transport the re-sealed ballot boxes to the county seat where they were impounded pending any recount.

Keeping the ballots separated by precinct makes it easier to detect and prevent fraud. It gives one many more ways to cross check internally this election and against previous elections and precinct demographics. Recounts (other than those during the precinct counting) were done at the county level by the election board, again in plain view. Recounts were by precinct; the ballots were never physically combined.

There were abuses, there were cases of stuffing ballot boxes, people being blocked from observing, "ringers" put in place as representatives of one party or the other, and lots more. The process became more open and reliable as communication, roads, and publicity made it harder to avoid scutiny.

If I were updating what we did 30 years ago, I would now have video cameras record all the counts and allow any observer to do the same.

I currently vote using optically scanned ballots. After the 2004 election which was a real mess and which, unlike most places, had unrecoverable errors with DREs that hurt Republicans, with a lot of hard work by a lot of people, NC election law was changed to require a voter-verified paper trail from DREs (or hand-count or scanned paper ballots), random audits of elections, and a number of other improvements. While a great improvement, I would still prefer hand count because I deal with computer security and trust my neighbors more than I do any computerized system (see my journal for some rants about how vulnerable all computer systems are.) Try to see "Hacking Democracy" for an example of hacking the opscan systems.

The biggest problems I remember were from larger urban precincts. It becomes increasingly difficult to count without an error as the number of items increases. Then, if there was a problem with a count mismatch, another count would have to be done making it take still longer. You can minimize these problems by having multiple ballot boxes within a large precinct and counting them separately. For a large precinct, you have to be a lot better organized to avoid problems and delays.

although that might be the problem. I didn't get the impression you were disorganised, just that it would be hard to guarantee the efficiency and probity you describe in tens of thousands of precincts nationwide. But perhaps I underestimate you guys! I just think precincts are a bit small to remain uniformly uncorrupt.

Anyway, good to have a substantive post on how it is actually done! I'd love to see it work.

I agree wholeheartedly with this statement:



That is what I keep telling the computer scientists.

We don't want fancier gizmos, we want a ballot that any
person could recount.

Thats paper.

And I trust bi-partisan teams to audit, count, or recount an election
better than some machine.

Because of the political realities, we had to fight the battle we could win
in NC, because to lose meant ALL paperless machines across the state.

Publicly hand Count at the precinct or polling station level, not the county level.
Post results on precinct door. Turn in tally to the county.
:)

How do you check they've done it right?

--that while complex programs involving machines are inherently unreliable, complex administrative procedures involving people just have to be reliable! :sarcasm:

we are letting our kids DOWN. You may be willing to do that, but I am NOT, willing to do that.

What worries me about forcing all ballots to be hand counted in America is that the ballots might need to be transfered from polling place to counting place, which takes time. And during that time people can get a good idea of what they need to win and start removing or adding counterfeit ballots to the bunch.

Message removed by moderator. Click here to review the message board rules.

Perhaps you would like to explain how you would conduct an all hand-counted election in America and not expose the election to easy opportunities for fraud.

Election fraud is all too easy. Hand counting is slow. The more slowly you count the ballots, the less oversight you will have. The more counting centers you have, the less oversight you will have. The more complex the ballot, the less obvious miscounting would be to those overseeing the count.

Tell us exactly how you propose to do it. Because you are right - I don't actually understand the GAME you say you KNOW.

I understand how it is done in the UK, and it works. But I cannot see, unless you explain, how you would apply our system to yours. The reason it works in the UK is because of the way our constitution works. The same system did not work well in Ukraine. Elections throughout history, and throughout the world, have been rigged on hand-counted paper ballots.

Why would it work in the US?

and you begin counting the ballots, with observers from both parties, get the cameras rolling and you keep counting until the loser is satisfied.

If we were computer experts and we teamed up to start manipulating elections which counties would be the easiest target? The Counties that Hand Counted Paper Ballots or the Counties that have E-Counting machines? Just asking.

Elections officials have trouble finding a sufficient number of poll workers as it is, and these workers are generally paid.

Where are all the counters going to be found?

WATCH US!!!

e-voting is "faith-based".

Look, I'm not against hand-counting. I think it's a really good way of counting votes, and it works brilliantly in the UK. But you haven't even begun to address the question I asked, so let me phrase it more clearly:

In the UK we have a few hundred constituencies, counting about 30,000 - 40,000 ballots each, for a single race. Rarely, there is more than one race, in which case the ballots for the other race are counted separately.

Because we count in constituencies (units of 30,000 - 40,000 voters), and because the count is normally for a single race, it is fairly easy to engage skilled counters (often bank tellers) for the night, and to ensure public supervision.

So my question is:

How do you ensure that well-trained counters are supervised, and their tallies checked, at tens of thousands of American precincts for elections in which there are multiple races to be counted? Remember that although we are a small country, your population is only four times larger.

Saying "we will get it done" is like Peter King saying "we'll take care of the counting".

The question in both cases is: "how?"

and we teamed up to start manipulating elections which counties would be the easiest target? The Counties that Hand Counted Paper Ballots or the Counties that have E-Counting machines? Just asking.

then e-voting would be the easiest target. If your expertise was in polling, you might use push polls or robocalls to discourage voters for the other party. If you were Karl Rove, you might try burglary - which might include paper ballots. If you were a general all-purpose ratbag you might go for voter suppression.

The point is that elections on hand-counted paper ballots can be corrupted all too easily, so if you want to use that method, you have to figure out how to secure it and check it. There is nothing intrinsically secure about hand-counted paper ballots.

could you imagine, Rove would need pickup trucks secretly hidden all over the country to pick up all that paper, and the guy at the counting place would be scratching his head, saying "I know there was 18,000 paper ballots here two minutes ago"

:)

in Ukraine.

:)

But with the click of a mouse Rove can steal hundreds of thousands, and the machine heads can all scratch their heads - like they are doing now - wondering what in the world happened.

Really..... counting paper ballots is as simple as making two piles on each race. The winner- if the margin is large- would be readily evident.

If close, start the hand count.

Three people running? Three piles.

53 items on the ballot? No problem. 106 piles, one race at a time.

Heck, if we wanted to get real advanced, we could have a machine separate the ballots according to each race... have the machine make the piles for us. How lazy can you get?

But we start with federal races...ala HR6200. And leave Rove et al, holding nothing but his balls.

I'm with ya Kster... we're gonna do this!

them 18,000 ballots would have been undetectable, but being as there where people watching we can get to the bottom of it, RIGHT after we get all them paper ballots and count them up, oh wait there are no ballots and if there were ballots we would have to sue in order to count them ballots, then the SCOTUS would stop us from counting them ballots, and if we wanted to hire computer experts to check and see if the actual vote counting machine was counting the ballots correct, we would have to drag our asses back into court to fight the manufactures to give us permission to take a peak into them machines. AND THE BAND ON........

COUNT THE Goddamn ballots from the get-go....... Keep everyone honest!!

The Politicians vote counting gig is up, They just hate to let it go, But they have NO CHOICE, ITS OVER. ONLY A MATTER OF TIME!







ITS OVER, ONLY A MATTER OF TIME!

there is no pure straight ticket voting in many states, so your picture doesn't
work.

Further, countries that still have HCPB only have 1 ONE item on the ballot.

TO convince any lawmakers or officials to do HCPB, you have to address all of
the issues.

US - 50 + items on the ballot

Germany, UK, CA - 1 item on the ballot.


Shall we switch to voting for members of parliament here?

Hold elections 4 times a year?

You can't just order something done, you have to make it
doable. That requires sitting down and providing answers to all of the questions.

First, we have to make paperless voting illegal, or everyone is just
blowing hot air.

has switched to op-scan paper ballots, Do you think, Providing it is done within a reasonable time frame, with observers from both parties and cameras trained on them ballots, that concerned citizens should be able to hand count ALL of them op-scan ballots, just to make sure that the machine counted correct? In other words, two seperate mandatory and automatic vote counts, machine for speed, haNd count for our sanity or accuracy. Would you have a problem with that?

I believe that if we don't get off the stick now and
ban paperless voting, (which it took 3 years to get that much
support for HR 550) then we are screwed in 2008.

If we don't get a ban of paperless voting, then there will be
nothing at all to count.

We will have paperless verification systems in place.

Probably ordering that HR 550 require all ballots to also be
hand counted would kill the bill,and we would have nothing.

However, pass HR 550, and hopefully pass Kucinich's bill, and you
would get a full HC of the presidential contest, without
having to change the entire system.

That is the good thing about Kucinich's bill, it accomplishes
an excellent goal within the system we already have.

If HR 550 doesn't pass, then Kucinich's bill probably never will.

If it took 3 years to get 220 co sponsors for 550, think how long
it will take to get K's bill moving along.

Eventually, perhaps the public will be more enlightened, and let
us start holding elections 3-4 times a year,
and allow us to draft poll workers etc..
and decrease dependence on machines.

But we dont have the luxury or security to get everything we want exactly
as we want.

I don't want to see a President Jeb Bush in 2008.

two seperate mandatory and automatic vote counts, machine for speed, hand count for our sanity or accuracy. Would you have a problem with that? Does that sound reasonable? I'll deal with the lawmakers later. :)

Message removed by moderator. Click here to review the message board rules.

They go on and on, probably because they don't realize that the silly shit of secret vote counting is over, I figure, I could just put my 12 year old in front of the computer to debate them, because thats how GODDAMN silly their arguments are for keeping these secret vote courting machines in place.

Whose? Where?

"machine-heads" "domestic or otherwise" are, but speaking for myself, I don't trust hand-counts without checks anymore than I trust machine counts without checks.

That's why I'm in favor of a bill that mandates both manually countable paper ballots and random audits, whether or not the ballots are counted by hand or machine on the initial pass.

I don't think that makes me a machine head. It makes me someone who wants to see accurate and transparent elections.

Because you are a computer expert, you are very aware of how a computer expert could steal an election.

But there are other ways of stealing elections, including those conducted using HCPB, witness Ukraine November 2004.

That is why ensuring transparent and uncorrupt elections is not as simple as advocating HCPB. HCPB can be transparent (it is in the UK) but also may not be (it wasn't in Ukraine). If you could ensure the kind of rigorous protocol in every precinct that you yourself described, I'd be for precinct-level handcounts in the US. But given the level of corruption I see in US elections (including voter suppression tactics) I think it is vital to ensure that any counting method is checked and double checked, and that the chain of custody of the ballots is ensured at all times, including during the count itself (by rigorous bipartisan or non-partisan oversight). I'm not yet convinced that this can happen with precinct-level HCs, and I think optical scanners may have an important role to play in ensuring the integrity of the count. But there should always be random checks using a completely independent method.

I know that if you wanted to rig an election, you'd exploit electronic voting, because that's your area of expertise. But think: how would you do it if there were only HCs? How would you exploit the HC system you describe? Could it be done? How would you prevent it?

Since the US votes on Tuesday, a workday, it restricts who is able to serve in those positions. The counters (as opposed to the election judges), show up when the polls close, are sworn in, receive instructions, and get to work. That is why so many of the poll workers are retirees.

I would have considerable difficulty being a poll worker most years, but was a counter on a number of times back when we still hand counted. To return to hand-count everywhere in the US all at once would probably be a mess because relatively few people have experience with it now.

Because hand counts are done at the precinct immediately after the poll closing, the worst consequence of not having enough counters in a precinct would be that some down-ballot contests would not be counted until the next day. Under such a situation (or with something like severe weather), you keep the ballot boxes sealed and impounded until they can be counted. While it isn't satisfying to have to wait, it is most important to preserve the integrity of the ballot, every bit as important as trial by jury.

A post here explains how long counties took for hand counting in 2004 http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=462218&mesg_id=462220



It is hard to imagine that the people in the counting teams were generally working people, coming to count after a workday, staying up all night until noon the next workday.

I read it every morning instead of the newspaper. Doesn't everybody?

I dream in source code these days, why not let it determine who governs me?

****************

HR 6200

"Strike back with a vengeance" place the words sourcecode or secret vote counting where applicable :)

http://www.youtube.com/watch?v=Wnexu_eGyYs

Shout it out to any and all in YOUR state legislature!