Its Impossible To Hand Count The Ballots?

[kster]

These people don't think so. Click on link, video on left side.

Watch the videos (requires Windows Media Player)

Lyndeborough (~18 min.)

State Recount (~10 min.)

Wilton (~10 min.)

http://www.democracyfornewhampshire.com/

[bleever]

1. What's so hard? We got ballots, we got hands, and we got time.

Every year, catalogued by hand, millions of:


Library books,

Urine samples,

Bank deposits,

Cash transactions at grocery stores.


To quote The Six Million Dollar Man:

"We have the technology."

[Kelvin Mace]

43. Huh?

To quote The Six Million Dollar Man:

"We have the technology."

Yeah, and WITH technology it can be done. Without it, it can't be done accurately, fairly and timely.

[bleever]

63. The technology is hand counting, in this case.

I'm no Luddite. But I do want to hear more about why accuracy and fairness are sacrificed more under a hand-count system than an electronic system. The timeliness, I don't care about so much, and in fact I think it's an impediment to vote counting, just as it is in childbirth. It's done when it's done.

[Kelvin Mace]

86. The complexity of the ballot

and the shear volume of ballots make accurate hand counting impossible. I have gone on over the reasons probably twenty times in the last month.

Folks are always cavalier about time, but the American people are not about to wait weeks for every election to be decided.

[eridani]

2. Nice to know that small towns where everybody knows everybody else--

--can count 1000 or 2000 ballots by hand. Any suggestions for a single metropolitan county with high address turnover of local populations who have to count 1.5 MILLION ballots? Oh, and we have 7000+ different ballot formats as well.

[electropop]

5. India does it with over a BILLION people. It''s not that hard.

Please keep in mind that the number of available poll workers is simply proportional to the population. Hand counts are much faster than electronic ones, because of all the "glitches" (real and otherwise) introduced by vapor-voting machines.

[WillYourVoteBCounted]

19. 2004 India had all-electronic vote

The Bombay Ballot
What the U.S. can learn from India's electronic voting machines.
By Eric Weiner
Posted Wednesday, Sept. 29, 2004, at 8:17 AM ET

...While we in the United States agonize over touch screens and paper trails, India managed to quietly hold an all-electronic vote. In May, 380 million Indians cast their votes on more than 1 million machines. It was the world's largest experiment in electronic voting to date and, while far from perfect, is widely considered a success. How can an impoverished nation like India, where cows roam the streets of the capital and most people's idea of high-tech is a flush toilet, succeed where we have not?
http://www.slate.com/id/2107388/

More about India's voting machines here
http://en.wikipedia.org/wiki/Indian_voting_machines

[electropop]

30. Well, OK. How about Canada or the UK?

I believe I've seen them doing it on TV recently. Point is, I don't care if it's 10 people or 300 million, there are always enough people to count ballots, if the society chooses to have a fair election. It surprises me that people fall for the tired old right-wing excuses that "fair elections are too expensive," or "we have too many people for paper ballots," or "hand counting takes too long." These are all just silly, even at a glance.

Answers, in order:

What is the price of losing our democracy? Also, electronic machines are FAR more expensive than paper systems.

If you use a fixed percentage of the population (say, 0.1%) for poll workers and ballot counters, at what population would the number of ballots exceed their ability to count them? Answer: There is no limit. If you had a trillion voters, you would need a billion poll workers. So what? It's still a tiny cost to pay 0.1% of the population to count 1,000 votes, which would be much less than a day's work.

Considering all the delays introduced by real and artificial "glitches" in vapor-voting systems, and the inevitable rising tide of legal challenges, and the efforts, such as in 2000, to block vote counting no matter how long it takes to block it, does vapor voting actually save any time? When you took a test in school, did you get extra credit for finishing early, even if all your answers were wrong?

[Febble]

31. Well, read my posts

(I'm a Brit, who spent seven years living in Canada).

One reason is that we vote in constituencies (ridings in Canada) for members of parliament or assemblies, and have very few races (rarely more than one, even more rarely more than two, on one election day). In the UK referendums are very rare, and often held separately.

Count is held at constituency level. I don't know how you'd make it work for American elections.

[electropop]

32. We have precincts.

We already have poll workers at the precincts. Most of them can count.

A typical precinct might have anywhere from a few dozen to a couple of thousand voters, depending on the locale. The number of workers at a typical precinct would be sufficient to count the votes there, and in fact in olden times, that was part of the job.

Similar to your constituencies, we have Congressional Districts, with about 500,000 people (not voters) each.

[Febble]

35. I think the ideal unit for counting

is probably somewhere in between. Ours are about 30,000 -40,000 voters, i.e. a lot bigger than a precinct, but a lot smaller than a congressional district.

i.e. big enough to get public oversight (including TV cameras) and skilled people, but small enough to be transparent.

[LuckyChoice]

105. better to count at the polling place

Transporting to a central counting location introduces too many additional vulnerabilities. (E.g., transport providers, counting computers, changing things around when loading or unloading, disregarding broken seals)

Counting at the polling place is simpler, faster and more secure. Put the equipment away, pull out big tables, and get the counting done asap. Post the results immediately--it makes it harder to introduce changes elsewhere. It can be done.

Relying on cameras to police the counting is a bad idea IMHO.

[lonestarnot]

131. Presidential elections should be held with election of a President

as the only objective. Then the ballots could be handcounted, 1 precinct at a time.

[Kelvin Mace]

22. India's ballots are simpler

and their election are VERY crooked in a lot of places. Accuracy has never been a hallmark of Indian elections.

[OnTheOtherHand]

9. I've decided to adopt a "show me" approach

Maybe all the "naysayer" rhetoric has taken a toll, but I no longer try to explain to people what makes it so hard to implement HCPB in a metro county. I just invite them to do it.

Impossible? certainly not inherently so. So, where is it happening? I'm done arguing about abstract feasibility. My personal opinion, or yours, is not the limiting factor here.

(Don't get me wrong -- I'm not telling you to stop making good points. I'm just reporting.)

[diva77]

12. with this logic that would mean that large counties shouldn't have juries

for jury duty either. There are not 7000+ ballot formats in a single precinct. If you view the problem as precinct by precinct it's about figuring out how to count the results for 1000 (and most likely far fewer) cast ballots.

[Kelvin Mace]

16. We have 147 ballots styles

in our county with around 220,000 votes cast.

[Febble]

18. Yeah.

I don't think the UK or Canada is a very good comparison.

Which came first: fancy ballots or fancy counting machines?

[Kelvin Mace]

27. I think they started getting complicated

in the 70's when folks like California started seriously into voter referedums where anyone who can raise x number of signatures can get an issue on the ballot. This is how the recall election was done in California a few years ago. Lever machines started earlier in the century, with OpScans coming along in the 60's.

[OnTheOtherHand]

28. multifarious electoral geographies have a lot to do with it

Jim Fishkin, in the intro to his book Voice of the People, reported that he couldn't even figure out exactly how many elected officials represented him, because no one had an authoritative list of all the special districts that he might belong to.

But he was able to identify over three dozen distinct positions (over 100 people) who represented him, at geographies including state, Congressional District, state senate district, state representative district, county, district, precinct, city, and school district.

Those overlapping geographies contribute hugely to the complexity of the ballot (both length and variety of "ballot styles").

[electropop]

33. So what?

We have multiple ballot types, whether implemented as electrons or as ink. Each precinct has one ballot type. The ballots for a precinct can be counted. I fail to see how the number of ballot types is something that is more easily handled electronically than on paper. Are you saying it's harder to ship 50 types of ballots to 50 precincts in a county, than it is to ship 50 configurations of vapor machines? Which is easier for the poll worker to verify: looking at the top sheet of a box of ballots, or a machine with no visible distinguishing characteristics?

As for printing ballots, we've done it for centuries. We have the technology. In fact, with the aid of computers, printing and correctly shipping innumerable unique document is trivially easy. Ask the direct mail industry.

[OnTheOtherHand]

44. please refer to post #9 n/t

[Kelvin Mace]

36. Agreed

You have city, county, state and federal races, various referendum and bond issues, at large offices, plus some races where you are allowed to vote for multiple candidates in the same race.

[JimDandy]

66. "Voice of the People" was published in 1997.

Statewide voter registration databases, which were mandated by HAVA to be up and running this last Jan., now can do this automatically. When a voter's address is input, a list of every special district corresponding to that address can be generated, along with a list of officials representing each district. In addtition, maps of each of those overlapping districts can be generated.

Fishkin should have no problem obtaining that info now.

One caveat. In my state this past election, every time I tried to access those online maps on election day, the web page was down. Turns out web pages with those maps were down most of the day, not only on my county's website, but also on the State Election Office's site. Electronic election technology - where is it when you NEED it?

[OnTheOtherHand]

78. that's an interesting point

Fishkin's main point was complexity, not the unavailability of info -- although he did get a bit carried away on the latter. But your point underscores how useful "high tech" is for dealing with at least many election-related conundrums. (Obviously that isn't an argument for paperless DREs that leak like sieves. And as you point out, high tech can fail when you need it.)

[JimDandy]

65. Surely there aren't 147 ballot syles

in each polling place in your county? In the last election, many polls in the largest county in my state had only ONE ballot style. The MOST we had at any one polling place was ten. Very doable for hand-counting paper ballots at each polling station! Especially if each paper ballot syle is clearly differentiated by number, or color, or some other means.

[eridani]

71. There are 7000+ ballot formats in King County, WA

which is why Andy Stephenson, who had firsthand knowledge of how elections actually work here, favored audited opscans.

My state legislative district is split between THREE different congressional districts, and FIVE different King County Council districts. We vote for individuals (not party slates) in all these different combinations and permutations. Not to mention school boards, water and sewer districts and referendums.

[Kelvin Mace]

110. We keep trying to explain reality to them

and they keep clinging to the fantasy.

[Febble]

13. We count hand-ballots

in both rural and metropolitan areas. Rural areas are more of a problem, not less, because the polling places are further from the count.

The counting is done by constituency (about 30,000-40,000) voters, usually at a local school gym or school hall. Ballot boxes are transported by care from polling station to count (signed and sealed in the presence of witnesses) at the polling place, unsealed in the presence of witnesses at the count.

I think the unit is one key to our success - bigger than a polling station, smaller than a county or metropolitan area.

Another second key is that we only have one race per ballot, and rarely more than two races (most often a single race). Where there is more than one race, ballots are on different coloured paper.

And a third key is that constituencies are an actual democratic unit - we vote for constituency candidates, so the candidates are present at the count. The winner will represent that constituency.

I think there are more, but those things certainly help our system work.

[kster]

14. Did you say tranported, maybe I can interest your country in to buying

some........

Clear paper ballot counter, transporter and storage box
Clear Automatic Paper Ballot Counter / ballot transporter / and ballot storage box, all in one?

Clear compartmentalized ballot boxes. Run an election for pennies




http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=425015

[mod mom]

15. Count at the precinct level so no hanky panky in transport occurs and

have the cameras filming. That way if there is any question of tainting we can document it.

[Kelvin Mace]

23. I can steal an election right in front of a camera

Just ask any magician how hard it is to fool a camera.

[mod mom]

45. So you advocate using machines?

I am suggesting you allow counters from all parties with a camera overseeing the procedure.

Anyone who still thinks using the machines with so many reports stating they are insecure would seem to me might have some financial reason to do so. The cost of the machine, the operating and maintenance costs, and replace fee are ridiculously high. Why not use a safer, less costly method like most of the other democracies of the world?

[Kelvin Mace]

47. Oh, have you been paying attention?

I don't mean to be sarcastic, but I am all over this board stating my position. I support the use of OpScan with requirements for disclosing code and random audits of the machines post-election.

In NC this was codified in s.223, which is probably the toughest law in the nation, and one which Diebold walked out on rather than comply with.

As a person with a long and established record opposing paperless voting, I find the insinuation that I have financial dealing with these scum a bit offensive.

[mod mom]

54. Why the need for Op-Scan? Why use any machine that can provide errors?

especially with low percentage of audits and high costs (not to mention funnelling $ to GOP cronies) Why not get rid of machines altogether? Random audits are a joke-we witnessed that in Ohio. I had no idea of your "long and established record". I didn't say you but those who support machines in general-Why advocate them when we can handle elections w/o them and their high cost (and with confidence)?

[Kelvin Mace]

58. Properly performed audits are quite good

at ferretting out fraud and error. I have stated many times why HCPB cannot be done with 100,000,000 ballots. The cost of OpScan is a third the cost of TS, and maintenance costs are much lower.

[mod mom]

76. 1. You're still suggesting funnelling $ to GOP Corporations with higher

(maybe not as high as TS-but still considerably higher than HCPB) costs associated with it. If this is done at the precinct level the magnitude is not as overwhelming as you suggest. Precincts have around 1,000 (The 20+precincts I audited from the '04 election (read high turnout) had around 280-400 ballots cast including absentee) registered voters, so it's non as monumental when broken into smaller units. Other large democracies can do it as well as the US prior to the 1980's.

I have yet to here your reason for your fondness of opscan. So it's lower than touchscreen, it's still higher than HCPB and subject to "error".

[Kelvin Mace]

81. Actually, properly

calibrated OpScan is more accurate with large number of ballots than handcounting. It is also faster.

I don't know where you live, but we had an average of 1,500 ballots cast in my precincts, of which there were 147.

Please name a single large democracy that has a ballot as complicated as the US does, doesn't use a standardized ballot, yet counts them by hand.

[mod mom]

83. Standardized ballots can be used with HCPB. What occured pre-machine?

Did anarchy rule in your state? I guess not.

[Kelvin Mace]

85. Optical scan and early TS

systems have been used in our state for over two decades. Smaller counties have counted by hand, or used lever machines.

Pre-machine?

The first lever machines were used in 1892. In fact, if you check the 1892 election you will find that a bit less than 12 million people voted.

Today you are counting nine times as many ballots as from the "pre-machine" days.

[mod mom]

88. 2001 CalTech/MIT study of HCPB have the lowest incidence of spoiled,

uncounted and unmarked ballots.

Source: http://www.votersunite.org/takeaction/handcounted.asp

BTW whether 12 million or 200 million doesn't matter if you are handling it at the precinct level.

[OnTheOtherHand]

92. Kimball 2005 finds lower residual rates for precinct-count op-scan

in 2000, 2002, and 2004
http://www.umsl.edu/~kimballd/rtables.pdf
in fact, an entire percentage point lower in each year.

The results of the two studies diverge somewhat for 2000 -- I'm not sure yet why. For people who are actually interested in empirical research, it's important to know that quantitative research on technology and residual rates can yield different results based on arcane differences in method. As fewer and fewer votes are cast on hand-counted paper, the residual rates become less robust -- more susceptible to influence by a few counties which may vary in many ways apart from technology.

I'm not prepared to conclude that precinct-based op-scan 'really' leads to lower residual rates than HCPB. But it isn't serious just to cite the five-year-old study as if it is the final word -- any more than it would be serious just to cite Kimball's tables as if they settled the issue.

[Kelvin Mace]

93. Where to begin

First, what does the report actually say:

The central finding of this investigation is that manually counted paper ballots have the lowest average incidence of spoiled, uncounted, and unmarked ballots, followed closely by lever machines and optically scanned ballots.

How closely? On average 0.2%, but note below that OpScan reliability is improving (in 2000 it beats HCPB by 0.1%) while HCPB fluctuates.

Now, look at table 1.

Only 1.3% of the voting population used HCPB, 27.5% use OpScan. Who is using HCPB?

A somewhat different distribution of voting technology across counties holds, owing to the very different population sizes of counties. Punch cards and electronic devices tend to be used in more populous counties, and paper ballots tend to be used in counties with smaller populations.

Third, voting equipment usage has a strongly regional flavor. The Eastern and Southeastern United States are notable, even today, for their reliance on lever machines. Midwestern states have a penchant for paper. And the West and Southwest rely heavily on punch cards.

Now why do you think small counties use HCPB while big counties use machines?

If you look at Table 3 which compares residual votes between machine types, you will see that by 2000 the difference between OpScan and HCPB was 1.2 and 1.3 respectively, thus in the 2000 election, OpScan was 0.1% better than HCPB.

Now look at the results from 1988 until 2000. HCPB swings wildly: 2.2/1.4/2.1/1.3, whereas OpScan shows a constant decline: 2.5/2.4/1.5/1.2 (I would love to see results from 2004)

So, it seems that OpScan is getting more accurate, while HC fluctuates.

Some other comments of interest:

Counties tend to adopt newer technologies that are analogous to the technology they move away from. Optical scanning has been most readily adopted in areas that previously used paper, especially in the Midwest.

Another take on the average reliability of equipment is the percent of all ballots cast for which no presidential vote was registered. This is displayed in the fourth column of numbers: this is the weighted average of the county residual vote, in which we weight by total ballots cast in the county. All of the figures shrink toward zero but the same general pattern holds. In fact, optical scanning seems to do particularly well by this measure. Only 1.6 percent of all ballots cast with optical scanners showed an over vote or no vote over the years 1988 to 2000. Approximately, 1.8 percent of voters cast an over vote or no vote using paper ballots or lever machines.

Second consider optical scanning. The difference in the residual vote rate between scanners and levers is trivial once we hold constant where equipment is used, how many people voted, the year, other statewide candidates on the ballot, and technological changes. In both analysis, the difference between optically scanned ballots and lever machines is quite small and statistically insignificant. Levers and paper and scanned ballots appear to offer similar rates of reliability, at least as it is measured using the residual vote.

[OnTheOtherHand]

95. wish granted

"I would love to see results from 2004"

I'll repeat the link: http://www.umsl.edu/~kimballd/rtables.pdf -- again with the caveat that not all the results are entirely comparable.

[Kelvin Mace]

97. Well, remind me to dance at your wedding

Interesting!

All technologies show a spike in 2002. Now, if I had to deduce why, I would say that it had to do with a lot of counties introducing newer technologies in 2002 (people converting from lever to OpScan, or paper to OpScan), except I don't know why paper showed a jump as well.

Except for 2002, OpScan continued to decline in errors, as did HCPB. Of course I would point out that the number of counties using HCPB declined while OpScan increased.

Stunning to me is that OpScan used for the precinct level tabulation (instead of central tabulation was SIGNIFICANTLY more accurate than HCPB (average .97% versus 1.97 for HCPB)

This is a quick reading, as I have to dash out of the office and will read the report closer when I come back.

[OnTheOtherHand]

98. too late, but our 16th anniversary is coming up

Oh, shoot, I meant to warn you about 2002, but I forgot. These residual rates are all top-of-ticket; I think Kimball used gubernatorial (pardon, too many windows open already -- I can't stand to open one more). So, higher dropoff rate for governor than for president -- not too shocking.

PCOS may truly be state-of-the-art, but I am reading Ansolabehere and Stewart 2005, and they argue that HCPB to some extent performs better than it looks (not that it looks bad), because small counties tend to have higher residual rates regardless of technology. The theoretical reason why PCOS might have lower residual rates (let's not say "more accurate") is that it spits out overvotes and can even be configured to warn on undervotes.

[Kelvin Mace]

99. Which is why we are pressing our BoE

to program the machines to reject undervotes, not just overvotes as it does now. Still, that is an impressive jump. I wish I had had these figures handy when I was battling our county election director.

The report seems to bear out my assertion that OpScan is continuing to improve.

Also of interest was Table 2 which broke things down by hardware. The LOWEST residual rate was for the 17 counties in Nevada using Sequoia DRE with a VVPT, only 0.3%.

It would be interesting to see if we could figure out what the average spike is in off-year election, then factor it out of the number in order to get an equalized trend line.

[eridani]

118. Paper beats opscan for counting single races, though

Opscan starts to get significantly better when you have three or more races to count.

[OnTheOtherHand]

123. could you say a bit more about this?

It makes sense generally, because hand counts will require multiple passes, but I'm not sure I have any good sources for direct comparisons.

Honestly I don't care that much, but after the first 100 posts, I sort of want to have had an opportunity to learn something.

[Wiley50]

129. Where to Begin?

I thought that thread finally died?


Sorry David, I couldn't resist

[JimDandy]

117. He doesn't seem to get that

it "doesn't matter if you are handling it at the precinct level."

[JimDandy]

68. Yes...

with black box voting anyone can steal an election right in front of a camera, the voters and a horde of election officials and we'd never know it. I don't need to ask a magician how easy it is for someone to be fooled -- I simply read your posts.

[Kelvin Mace]

100. You can steal elections right in front of a camera

with paper too.

I don't need to ask a magician how easy it is for someone to be fooled -- I simply read your posts.

Oh, gee, what a retort. Think of that one by yourself?

[Febble]

17. Transport really isn't a problem

And we don't need compartmented boxes, because we only have one race per ballot. I'm a little concerned about ballot secrecy with your boxes, pretty though they are.

With regard to Mod Mom's point about hanky panky, I'm still concerned about precinct-level counting. The more counting places you have, the easier, I would think, to corrupt the count, and the more difficult it will be to have good oversight. With 30,000 to 40,000 ballots being counted in one place, you can have TV cameras, and members of the public all gathered. It is quite a ceremony.

It's true that the weak link is the transport but as I say, it isn't really a problem. The ballot boxes are numbered in huge print, and sealed in the presence of witnesses. I'm not sure how they are looked after in the car, but I wouldn't have thought it was difficult to get bipartisan escorts for the boxes if you are worried about that. Maybe that's where your jury duty could come in. And the numbers on the boxes are checked off as they come in.

But the point is that the count itself is a skilled and responsible job, plus you need lots of people looking on. A school gym or town hall is perfect, because you have room for the public behind a roped barrier, and plenty of space for bipartisan scrutineers to walk up and down, checking progress. And of course TV cameras at EVERY count.

Anyway, that's how we do it. I think the size of the counting unit is worth thinking about. I think precincts are a bit small, myself.

[kster]

34. Absolutely, you get a good guestamate at a glance

at the precinct and then you transport the ballots to the main counting facility.

Secrecy? You walk behind a curtain and place your ballot in the clear box.

[Febble]

39. Well, we mark our ballots in secret

then drop them in the ballot box in clear view. It stops people stuffing the ballot box.

[kster]

41. Thats the beauty part of the clear box, you can know at a glance

if they have been stuffed, when you do the guestimate at a glance, if there is 5000 ballots in the clear box and only a 1000 people showed up to vote, that particular precinct puts little red flag sticker, on that precincts ballot boxes.

[Febble]

46. Yeah, but if you do it in full view

there's no chance to stuff. What's with the curtain?

[kster]

48. Close the curtain behind you

Close the curtain behind you.

http://manzano.org/mbps/html/firstvoter.html

[Febble]

51. Yes, you need a curtain,

or something, while you fill in the ballot. But we put it in the box in full view, which means that our ballot is secret, but our casting of it isn't.

This seems to be a good thing.

We don't want acid in the ballot boxes

[kster]

56. That makes to much sense for America to do, out in the open

America doesn't count ballots out in the open, the closest we get to the actual counting is an audit , and that audit is just to be sure the machine was counting the ballots correctly. Then when you are in position to get a hand count, the SCOTUS slaps you down, and says stop that hand counting bullshit.

Please,out in the open in full view, Not in America. LOL

[salvorhardin]

52. Hmm. Maybe a slight problem.

Humans are notoriously bad at guesstimating this kind of thing. Yes, it's easy to tell if the box had been stuffed from the start, but once it has a good number of ballots in there could you reasonably expect someone to tell? It's like the bar games where you guess the number of marbles/gum balls/pickled eggs/whatever in the jar. Few people will even come close.

[kster]

57. Take a look at these babies, even I could get a guestimate on

these boxes, they are marked

[salvorhardin]

59. Would these be the boxes voters deposit their ballots into?

Or are the ballots transferred from a temporary ballot box to this one?

Also, the design of your box looks as those it encourages people to vote a straight party ticket. Would there be separate boxes for each race and proposition in an election?

Would there be different sized boxes for small, medium and large precincts? For instance, in 2004 I voted late in the day and there were only 200 odd people ahead of me on the sign-in book.

I like the idea of a clear box with graduated markings to give a visual quick-reference, but I think as proposed it overly-complicates things, and as Febble noted I'd be concerned about the secrecy of an individual's vote.

[kster]

62. My thought is, and I'm not an expert

I would put the boxes 3 feet from the wall, with the slots facing the wall, then I would have a partition covering three sides of the box tall enough, that you could see the voter, but would not be able to see what slot they put their ballot in.

2nd question, would be for the experts, Im only one man.

3rd question, absolutely there would be different size boxes.

4th question, What the hell does Feeble know (just kidding Feeble). :rofl:

[JimDandy]

69. kster, I've always liked this idea, but

most states have laws stating that ballots can't be counted until the election closes. The graduated markings on the ballot box sides seem to me to be a form of tabulating the vote before the close of the election.

(Excuse my post if this has already been stated and answered previously.)

[kster]

84. If thats the case, then we need to

cover the boxes till the end of the day, at the close of the election, we take the covers off, get a quick estimate before they leave the precinct, then ship them to the counting room.


Just a thought.

[LuckyChoice]

107. I agree with your assertion

that "the weak link is transport".

You then make the statement, "it isn't really a problem"--even though transport introduces multiple opportunities for interfering with the chain of custody. We can't rely on written procedures, because election officials and workers violate regularions all over the place and nothing is done about it. The last thing we need to do is to introduce more opportunities for tampering.

Counting at the most local level makes it harder to interfere with large numbers of votes. Once you're dealing with a central counting situation lots more can go wrong and many more votes can be affected.

You know only too well how far we are from having any kind of meaningful audit--and no, I'm not talking about 2% or 3% because as you have pointed out, general audit percentages like this are worthless for determining fraud since the statistically appropriate audit requirement to catch frauds depends on numbers, not percentages. I don't see any likelihood of our legislators hiring statisticians to help them pick the "right" percentage to audit in any given situation to be able to detect fraud.

Much better to count the votes where they are cast. Get extra poll workers like jury duty, if more people are required to do this efficiently. Design the ballots in a way that facilitates hand-counting. This is not rocket science, it just takes good management.

[heartofthesiskiyou]

49. You're looking at

1.5 M and saying OMG, doesn't get you a clear picture.

How to count them is just like it has always been done. Precinct by precinct. Each precinct has 1-2000 votes to be counted. Takes about 15-30 minutes. Each precinct is staffed by the same folks that always come out for this sort of thing. an army of 70 year old ladies and gents in sneakers with no other fish to fry then to count every vote and over see that no shenanigans occur. This is how it's been done for 225 years and has by and large worked well. Why do you think the repukes are heading for the machines, because they want fairness, or maybe they don't want to see counters have to work so hard for the thirty minutes or so?

The 'It's too hard' argument should never have been permitted. It's not too hard.

BTW any hacker will tell you, on things that really matter, LOW TECH RULES.

I trust my little old ladies in tennis shoes any day, and I like thier welcoming smiles to boot.

[OnTheOtherHand]

53. a bit of revisionism here

Hand counts simply aren't "just like it has always been done." Doug Jones reports that by the 1930s, "essentially all of the nation's larger urban centers had adopted lever voting machines." http://www.cs.uiowa.edu/~jones/voting/pictures/#lever

I don't have historical stats, but it appears that in 1980, somewhere around 10% of voters used HCPB. Now it is closer to 1%. Treating "voting machines" as a contemporary Republican conspiracy is ahistorical.

I just don't know how you came by your conviction that "Each precinct has 1-2000 votes to be counted. Takes about 15-30 minutes." Maybe that is true where you are, but hey, it's a big country out there.

Maybe it's not fair, but when I see advocates of HCPB rewriting history in support of their views, I lose interest in the whole thing pretty fast.

[mod mom]

60. Amen. nt Low tech rules in vote counting!

[Kelvin Mace]

101. Could you supply me with the source for your figures

that it takes 15-30 minutes to hand count 2,000 ballots? Again, how complex a ballot? How many races? How many counters?

This is how it's been done for 225 years and has by and large worked well.

No, it hasn't been done this way for 225 years. Starting in 1892 we started using machines. Today only 1.3% of the voting population have HCPB, and they reside in low-population counties.

[WillYourVoteBCounted]

116. this isn't correct - NC had 3 hand counting counties in 2004

NC had three tiny hand counted paper ballot counties.

Graham, Hyde and Tyrell.

FOr the 2004 General Election, they were still counting the ballots
until about noon - 1PM the next day.

15 minutes?

What are you referring to?

[nealmhughes]

3. Didn't we manage to do it in the 18th, 19th, and a great part of the 20th

Centuries? So what if it is slow. Try it, you'll like it.

[Kelvin Mace]

24. Yes, with a population a fraction of today's

and with LOTS of rotten elections.

[JimDandy]

73. The population,

as a whole, back then is not the number on which we should focus in this discussion -- the voting population per precinct is.

As to the "LOTS of rotten elections" -- yes, both then and TODAY, in the electronic age. But in today's e-voting environment, we are faced with conditions of near zero-visibility. It is an environment in which we are forced to measure the fairness of an election, and therefore the health of our democracy, by the potential for fraud, instead of actual rottenness itself.

A 'hand-counted paper ballots' voting system is the best elections system, because "We can only agree on a vote we can see"*

*This slogan, by Barry Kerr, was posted on a recent thread on BBV.org's online forum. I've never met the folks who run BBV.org, but the visitors to their forum have outstanding, thoughtful discussions on their board. This is my measured opinion and, as an outsider to that organization, was arrived at after careful study of that board.

[Kelvin Mace]

89. I am not concerned with what anyone on BBV.org

has to say.

History is full of examples of people disgareeing over voter intent with paper ballots.

I am still waiting for someone to give me an example of a country using HCPB that has ballots as complex as ours, with no standard ballot and whose results are considered accurate.

Someone please show me a county with 200,000 or greater that is using HCPB in the US.

[LuckyChoice]

108. yes, "voting population per precinct"

is what matters.

The old, historical kinds of election fraud can be prevented relatively easily with good management and proper oversight at the polling place. There are plenty of examples of good procedures for this.

Ballots can be designed to facilitate hand counting. Just because this isn't done now doesn't mean it cannot be done. Imagine how much easier it would be to redesign ballots (perforations, different colors for different races, whatever works), compared to designing, testing and overseeing voting machines!

Making the changes to have secure, accurate hand counting at the polling place would be less complex and more reliable than any option I've seen so far using machines.

In no case should we give up transparency. Elections are too important.

[Kelvin Mace]

111. Interesting

a brand new poster jumps into the fray.

[Bear down under]

4. We do it in Australia

--pencil on paper ballots -- but then we have a population a tenth the size of yours. And a system as watertight as we can make it that keeps anyone affiliated to any political party from getting within touching distance of them. The downside of course is that in a closely-called seat it can take nearly a fortnight to get the final result. There has been some talk of introducing electronic machines ("to make elections faster and cheaper") but the idea is the reverse of popular.

[tbyg52]

10. I thought you had an open-source electronic voting machine?

I take it its use is not widespread? Or have they done away with it completely?

[eridani]

72. An open source system has been developed--

--but not used in actual elections so far, AFAIK.

[Kelvin Mace]

25. How many races/issues

appear on your ballot at one time?

[LuckyChoice]

109. What about New Zealand?

Do they hand count there?

Anybody know?

[rman]

6. sure, there were no elections before e-voting

:shrug:

[mod mom]

7. It can be done. It's time to throw out the machines and stop funnelling $

to GOP Crony corporations who produce tainted results. Hand Counted Paper Ballots, COUNTED AT THE PRECINCT LEVEL WITH FULL PUBLIC WITNESS will allow the will of the voters to be served and will regain public confidence in our system.

I don't understand how anyone, knowing the incidences of insecurity and tainted results from the machines, can still advocate using the damn things. I hope these corporations will be sued for the full amount (and more) of public dollars invested in them for delivering such a faulty product with disastrous results.

[tbyg52]

11. Throw all the machines out!

I don't care how much money has been thrown at them. We've wasted money before, with *lots* less reason.

[mom cat]

8. People in the civilized world do it all the time!

[Kelvin Mace]

26. Here's what I want to see

Some one show me a local where ballots are handcounted that has:

100,000+ ballots to count
25+ races on the ballot
"Clean" results with no evidence of cheating.

If you can find one, tell me how many people were used and how it was paid for.

[WillYourVoteBCounted]

20. can you describe your experience doing the hand counts?

I have spoken with just a few people who have done the HCPB,
some in my state.

Could you give us your first hand experience on it?

How long did it take you,
were the other poll workers enthused, or cautious,
what was the process,
how many counters were there at your polling place,
maybe provide some anecdotes of your foray into HCPB

[Kelvin Mace]

21. Let's do a comparison

The video says it took three hours to count 1,100 ballots. They have teams of three, and from what I can see of the video, they have 4 teams, or twelve people. The number of ballots counted is less than a single prfecinct in my county. They claim 100 absentee ballots to our 20,000+ in 2004. To extrapolate the people requirements, I would need 1,746 counters for my county alone. I would need around 54,545 counter for my state, and just OVER ONE MILLION counters for the entire country.

Also, their ballot was about 8.5x14 with 6-8 races, versus our ballot whcih was 11x17 with 30+ races. Places like California can have multi-page ballots.

Yes, you can count small numbers of simple ballots by hand, and do so with fair accuracy. But once the numbers increase several orders of magnitude, and the complexity and physical size of the ballot increases as well, it becomes logistically impossible pull off.

Twelve people can build a car in three hours. Building 1,500 cars a day (62 an hour) requires automation.

[Kolesar]

29. It's a scalable problem. 10x the ballots, use 10x the number of counters

[Kelvin Mace]

37. No, it is NOT a scalable problem

because you reach a point of diminishing returns. You are talking about organizing over a million people in 3,500 different jurisdictions and training them in the vagaries of the various laws governing elections.

I have 100 counties in NC. NC election law is 1,400 pages. The section dealing with countinmg ballots is over 100 pages. Each county and city has it's own laws which augment state and federal law.

[Kolesar]

40. Rules are rules

there have to be rules that election supervisors know to manage an automated vote. Why could they not just dictate their instructions to the temporary counters? This is not designing moon rockets.

edit:you are essentially saying that it takes a million people to count 110 million ballots. Well, I could count 110 ballots in about five minutes, then we would be out of there!

[Kelvin Mace]

42. There is a big difference

between election officials managing machines and managing over a thousand people..

you are essentially saying that it takes a million people to count 110 million ballots. Well, I could count 110 ballots in about five minutes, then we would be out of there!

Have you ever participated in an election? Worked a precinct? Observed a count?

Again, the ballots I work with are on large sheets (11x17), and can cover 25-35 races. Each race must be counted separately. Each must be counted by three people. This means that each ballot will be handled from 75-105 times. The more ballots are handled, the greater the chance of error, damage to ballot or cheating. In some jurisdictions, ballots are MULTI-PAGE.

Theory and practice are two different things.

[Kolesar]

75. Cleveland voting activist Victoria Lovegren is an advocate of paper ballot

so I will go with the advice of someone I know who has experience in voting issues

[Kelvin Mace]

87. So, she is the only person on the planet

with this experience? No one else.

Well, be my guest. You go right down to your BoE office and tell them you want HCPB and you want it NOW.

Do get back to me with what they say.

[Kolesar]

91. You need a beer

[Kelvin Mace]

94. I may yet have to take up drinking

[kster]

38. We could look at the second video see how they do

it,I think the second one counts out 22,000 ballots.

[mod mom]

55. Funny, Europeans which if I recall correctly have large cities can handle

it just fine, but not where you live.

[Kelvin Mace]

61. I'll say it yet again

Fewer people, simpler, uniform ballots.

[Kolesar]

74. Well, make the ballots uniform, then

[Kelvin Mace]

80. That comes next

First we get HR-550 passed, then we pass a uniform ballot law. Fight one battle at a time.

[mod mom]

77. It's not uniform ballots because every area has different races. If the

uniform ballot was just the issue and standardized form could be developed for HCPB as well. I audited optiscan ballots from Lucas County OH (Diebold) and I can tell you that although the size was uniform, the races and order were determined by precinct code. All it would take is manipulation the opti scan machines w/o a decent audit and voila another stolen election. Sorry...I have direct experience working with these ballots so your weak argument isn't going to make it with me.

Fewer people also means fewer people overseeing the procedure and could in fact control the outcome of the election. I say let the public oversee the vote counting so we can be assured that the outcome reflects the will of the people. Don't you want the outcome to be overseen with full public scrutiny? Don't you want less $ funnelling back to GOP operatives? Don't you want public confidence in the elections by eliminating machines that have the ability to be manipulated. The small % of audits, especially when we have seen them to be NOT RANDOM, but carefully controlled, can hide election fraud.

[Kelvin Mace]

79. By uniform ballot

I also mean a uniform method of marking the ballot. Some places draw a line, some fill in an oval, some put an "x". We need one ballot style, marked in only ONE manner.

All it would take is manipulation the opti scan machines w/o a decent audit and voila another stolen election.

That's why we need to pass HR-550, to have strong auditing in place.

Fewer people also means fewer people overseeing the procedure and could in fact control the outcome of the election. I say let the public oversee the vote counting so we can be assured that the outcome reflects the will of the people.

The more people you have, the more chance you have for organized groups to cause disruptions in the cunting. Remeber the Florida "protest" where GOP staffers pretending to be "angry citizens" managed to halt counting of the vote? Worked pretty well, didn't it?

Don't you want public confidence in the elections by eliminating machines that have the ability to be manipulated.

No, I want confidence in elections using vetted hardware and software, bakced by rigorous auditing. That will yield a fair and accurate election. Universal HCPB:

1) Is not going to happen, PERIOD.
2) If it did happen, will result in more election error and fraud than we have seen since the days of Jim Crow in the South.

[mod mom]

90. I audited Diebold Optiscan ballots from Lucas Co OH and there were

definitely different methods of filling in (as you say "X" fill-in or line drawn thorugh).

Thanks but no thanks...I DON"T want our elections in the hands of only a few individuals like we saw in Blackwell in '04. I would rather have the public oversight anyday.

Optiscan have central tabulators that can be manipulated.

NO MACHINES IS MY MANTRA! Hand counted paper ballots, counted at the precinct level with full public witness!

[Kelvin Mace]

96. Then again, I say to you

go fot it.

Instead of helping pass a law with teeth that will greatly enhance accuracy and fairness in elections, you push for your voter utopia.

In theory, you can build a perpetual motion machine that will work, as long as you don't have to deal with any of those pesky laws of thermodynamics. These machines work beautifully on paper.

[WillYourVoteBCounted]

50. do you have any legislation introduced?

I don't think anyone doubts that people can count ballots by
hand, but making it happen is another story.

Our elections have much more complicated ballots,
we elect our officials differently.

Are you planning to do anything to get HCPB in your own
area?

Have you spoken with your state lawmakers?

Have you spoken with your county election officials?

Have you talked with your political party chairs for their support?

People can run elections all sorts of ways.

We had 3 tiny counties counting paper ballots by hand, under 7,000 voters per county,
and they got through the day after the election.

A lot has to change to get HCPB accepted or even legislated.

Right now there is still a tremendous resistance to paper even with
computerized voting, so its a gigantic leap to HCPB.

But if you are doing any work with lawmakers or even your local officials,
please tell us more.

[kster]

64. I'm not sure of the number, I think 84% of the people

want to get rid of e-voting, with them kinda of numbers, you would think that the Politicians that were "supposedly" elected, would be all over this, BUT SILENCE, being as they choose to REMAIN SILENT on this issue, I can not see how anyone could get anything done having to go "FIRST" thru this group of THUGS, that we call Politicians.

NGU!! Keep spreading the word!! WE WILL GET THEM!

[Kelvin Mace]

82. So far, 87% of the folks on this forum

want HR-550 passed.

It depends on how you define "e-voting", which I view the same way I view BBV. The main creiteria is an system that doesn't produce a paper ballot at the time the vote is cast that can be seen by the voter.

[kster]

121. Where did you come up with that figure? Tea leaves?

[Kelvin Mace]

125. It's called a poll

It has been running for several days on the EF.

Again, the insinuation is that I made up the figure. I deal in hard fact, unlike many folks such as Bev. If you are going to call me a liar, then come right out and say it, rather than play coy.

[kster]

130. LINK TO POLL?

[OnTheOtherHand]

132. this is not at all hard to find

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=444359&mesg_id=444359

It's on the board's first page.

I assume you didn't actually intend to insinuate that Kelvin was a liar; if I'm right, it would be helpful to say so. (If I'm wrong, well, get over it already.)

[WillYourVoteBCounted]

67. Thinking You Are Right is Not a Strategy

I guess I am more inspired by the Diva's thread,
where she actually has the gumption to do what it takes to
get change implemented.

She is preparing a thorough argument in favor of HCPB
and coming up with rebuttals to the arguments that
would be incurred.

That is the smart way to do it.

Plus, she will be covering all bases by working to
see if HCPB can be implemented in some small localities while
exploring the possibility of having legislation written.
This is a plan that maybe if it can be started somewhere,
it will spread.

I would think that it is simplistic to assume that the
decision makers we face are going to be persuaded with
the limited arguments of "because it's the best way".

But thoroughly researched arguments with expert information
that addresses security, accuracy, procedures, the value of -
that is going to have a better chance of being heard than
just a demand being made.

The opposition to HCPB will be with the administrators of elections,
your elected officials, political hacks, and the voting machine vendors.

If you can't make it sound reasonable, if you can't make it sound
workable, and if you can't make it sound valuable, then you won't get
anywhere.

Since half of the country's lawmakers still think paperless electronic voting
is AOK, imagine how far they have to go to even consider HCPB?

Really far.

And, even in the 27 states with VVPB legislation, HCPB is not considered a
viable choice by decision makers. Just ask them.

This isn't about being right -
Thinking you are right is NOT a Strategy.

If you want to win, you have to do a lot more work than just posting your opinions.

[kster]

102. I was also also inspired by divas thread, but this

thread is showing video that the votes can be counted by hand, because in my opinion it can be done. This video is proof of that. The video in this thread may assist Diva in his/her quest.

Are you suggesting that anyone who makes a thread should have all the answers, or shouldn't have an opinion, maybe you can get the DU to make new rules, personally I like to hear other peoples opinions,but if you don't, then maybe see if the DU can have a non-opinion forum.

Hope this helps you.

[OnTheOtherHand]

103. try reading the post again

You offered an opinion that HCPB is not impossible (complete with video of it working in two small New Hampshire towns). WYVBC offered an opinion that "If you can't make it sound reasonable, if you can't make it sound workable, and if you can't make it sound valuable, then you won't get anywhere." I happen to agree with both those opinions, FWIW.

[kster]

104. Watch the videos again, it does all those things and more

it shows that counting the ballots by hand is doable(because it shows it on the video) reasonable, workable, valuable. What more do I need to do, the video says it all.

[Kelvin Mace]

113. It shows that HCPB

is possible in small venues with simple ballots.

We knew that, and it has no application to larger counties with hundreds of thousands or millions of complex ballots.

[OnTheOtherHand]

115. what more do you need to do? it depends

I'm afraid I will sound sarcastic by saying the obvious: If you are content to state an opinion that convinces you, then you are done. If you want to convince others, then you need to pay attention to their arguments, not just your own.

Since most of the arguments being made on the other side have to do with size (and complexity) of jurisdiction, videos from small towns in NH (or single-race recounts) aren't likely to cut it. They just won't.

[eridani]

119. It says nothing relevant to this particular elections activist--

--who happens to live in the county with the third highest population of voters in the country, with 7000+ different ballot styles. That's 3 to 7 times more STYLES than those two very small towns have total voters!

[Kelvin Mace]

114. The video

proves HCPB work in small towns/counties. Not applicable to what we have said over and over again. Show me someone counting hundreds of thousands/millions of complex ballots by hand and we will talk.

[LuckyChoice]

106. yes, people need more information

if they are to make an informed decision.

So far there's a massive deficit of information. Many people change their stance once they know more.

[Kelvin Mace]

112. Exactly what infomration is missing?

[Bill Bored]

70. I don't think it's impossible

If you can audit, you can hand count, esp. if it's done by precinct.

To audit 2% of precincts, you need at least 2 auditors at each precinct. Whatever that number is, it can be multiplied by the number of precincts. It's a distributed processing sort of model. Like thousands of PCs across a state, each doing their own calculations in parallel. Except they are people and not computers.

That said, the absentees are a problem and so are the early votes and any other votes where there aren't poll workers at the precinct on election night.

So I think if we are going to hand count, we should probably consider making absentee voting a lot more difficult (like actually requiring voters to be ABSENT), and also do away with early voting, which we probably should do anyway because of the chain of custody problems, etc.

If we can do that, then hand counting could probably be done.

But I also think it's a false choice. We should only need to hand count ENOUGH ballots or precincts to be sure the machines haven't screwed up. What we need is a REAL auditing protocol, and perhaps a different protocol for the absentees if they occur in large numbers.

[autorank]

120. kster makes a positive post and gets savaged. It's the 550 crew!

I'm not talking about everybody who supports HR 550, I'm talking about people who simply savage other people over an issue that really shouldn't generate much conflict. Everybody should want hand counted paper ballots. However, there are clear differences as to the approach: a) focus efforts on that now or b) make what we have more accountable through HR 550.

I'm strongly in the former camp however, i don't question eomer or garybecks devotion to the cause of voting rights for their support of HR 550 but I do question the shrill and uncivil dialectic When all else fails there is always the invocation of the incendiary former user here, "the Seattle based activist."

It started in an OP attacking a long term user here as a 'bev bot' and it has been all down hill from there. I was getting nasty comments on the news thread based on articles I'd post even though I'd run a news thread a day or two earlier COMPLIMENTING the legislative efforts of the people who were going after me. I've never had so many people on "ignore" (and you know who you are so stop posting replies to my messages, I don't read them).

This can go on for ever but that's ridiculous. If you don't agree with hand counted paper ballots, fine. I don't agree with HR 550. It's not the end of the world. This forum is starting to look like the epic wars in the Israel/Palestine forum.

We'll see if HR 550 is past. Those of us who oppose it are NOT going to Capitol Hill and tell in anybody NOT to vote for it. If that were the case I could understand the bitterness. If the standard for support is absolute silence in the case of those who oppose it, that's simply not going to happen.

This forum is DEMOCRATIC Underground and there is a large allowance for views representing the Democratic Party here. I'm not aware of any part of current Democratic politics that says you are somehow a traitor if you fail to support H.R. 550. In fact a few days ago Sen. Kerry, our presidential standard bearer, the one who actually won, said that any new voting legislation was a bad idea between now and the 2006 election. So he actually opposes HR 550. Our position, that there is a better way, is not even to that point. It simply says, lets advocate for the real solution now rather than fixes. It's middle of the road Democrat...so ease up and maybe something productive will come of the dialogue.

[OnTheOtherHand]

122. well, that's the problem

You have so many people on ignore, you really aren't in a position to speak credibly about the tone.

How about you quote the post where you think kster gets "savaged," and then I post what you wrote about me when I disagreed with TIA, and everyone can compare?

EDIT TO ADD: Here's another thought about civility: next time someone you generally agree with writes that anyone who disagrees with him or her "would seem to me might have some financial reason to do so" (see post #45), how about you step in and say, hey, no need to go there? Or, say, suppose someone writes, "one cannot but wonder whether you are (expressing your opinions) with clean hands." Would you be inclined to step in and say, gee, that was over the top? or would you chime in with something to the effect that disagreeing with your own opinion is "the height of intellectual decadence, a casual dilettantism that justifies the perpetuation of tyrants"? Just a wild hypothetical that I'm sort of curious about.

[autorank]

124. If you're on "ignore" leave me alone. n/t

[Kelvin Mace]

126. I love it when people deal

with facts and reality by sticking their fingers in their ears and singing.

[OnTheOtherHand]

128. or by calling people out and then asking to be left alone

As I understand it, the "ignore" feature does not confer carte blanche to have the last word on every subject. autorank has it within his power to ignore everyone who notes the inconsistency in his position, but that doesn't give him the power to prevent anyone from mentioning it. AFAIK.

[Kelvin Mace]

127. Please show where he was "savaged"

[kster]

133. kick.nt