Honest question: if the vote is on paper but tabulated by computer, are we

[helderheid]

still screwed? Wasn't it op-scan machines in Florida that had all kinds of anomalies?

[shraby]

1. Yup. It's that central tabulator that needs to go.

Wouldn't hurt for all the touch screens to go also.

[helderheid]

4. I agree

[FogerRox]

2. yeah U R right-- to a point

85% of AMericans vote in 2004 was counted by computer.

Yeah it could be were screwed-- But I aint going down without a fight--

SO you like Richs section 301 letter? He writes a good letter eh?

[Ian_rd]

3. Optical Scans are hardly an improvement ...

True, they leave a paper trail. But what good is a paper trail if we're still letting the computers count the ballots for us?

IMO, it means that we're "still screwed" as far as open and honest elections go.

[endarkenment]

13. Optical scanners are not computers.

They are fairly simple mechanical equipment. They are also pretty easy to test for accuracy and pretty difficult to tamper with. Why use them? They count ballots quickly and accurately and they are cheap enough for every precinct in the nation to use.

As I've said elsewhere, this thread seems to be confusing two parts of the election vote counting system: voter recorders and central tabulation systems.

[Stevepol]

80. Optical scanners are just easy to tamper with as touchscreens,

aren't they?

Why are they "pretty difficult to tamper with"?

They are programmed just like computers aren't they? Don't they use the same basic set up as computers? So why are they more difficult to tamper with than touchscreens?

The fact there's a paper ballot to audit and recount is an obvious plus but look at MN in 02. Mondale up by 5% in pre-election polls pretty much across the board, and yet he loses by 7%, about a 12% swing?

And MN was using largely optiscans I believe.

Of course the result could have been, and likely was, engineered by the central tabulators, but the fact remains that the optical scanners aren't any more secure as machines than any other computer-based system of vote counting.

Or am I mistaken?

[riqster]

82. They are quite easy to tamper with, but...

It is hardly worth it. Far better to hack the tabulation device.

[Stevepol]

81. Optical scanners are just easy to tamper with as touchscreens,

aren't they?

Why are they "pretty difficult to tamper with"?

They are programmed just like computers aren't they? Don't they use the same basic set up as computers? So why are they more difficult to tamper with than touchscreens?

The fact there's a paper ballot to audit and recount is an obvious plus but look at MN in 02. Mondale up by 5% in pre-election polls pretty much across the board, and yet he loses by 7%, about a 12% swing?

And MN was using largely optiscans I believe.

Of course the result could have been, and likely was, engineered by the central tabulators, but the fact remains that the optical scanners aren't any more secure as machines than any other computer-based system of vote counting.

Or am I mistaken?

[FogerRox]

15. paper ballots-- please-- a paper trial is somhing left behind--

Like kids in Grimms fairy tales, walking thru the forest

[whistle]

5. Yes, the best solution is electronic touch screen with paper trail

....or just plain paper ballets hand counted.

[nicknameless]

10. Touch screens? WHY ???!!!!!

All you would be doing in that case is have a computer print out a piece of paper. It would have NOTHING to do with how the votes would be counted. ... And Diebold has a great paper-trail design that spits out little receipts that are difficult for those without good eyesight to read. Plus they print them on THERMAL PAPER that doesn't even last the required 22 months without fading!

This would also stick us with touch machines with all of their freezes, crashes, vote-switches and paper-jams.

[endarkenment]

6. It was not optical scan equipment

in florida it was punch card readers. Not the same thing.

Optical scan equipment is pretty reliable and cheap and there is a paper trail and if there are disputes the paper ballots can be hand counted.

There is simply no reason to use touchscreen voting equipment. It is expensive, and it is difficult to audit, even with a paper output of some sort. These devices are being imposed on us because there is a lot of money to be made selling them and the potential for corruption is vast.

[helderheid]

7. I agree with you on the touch screens - I guess my question goes for

op scan machines too - are the results fed into a central tabulator which can be altered (like the other machines) ?

[endarkenment]

12. OK two different issues here

1) vote recording equipment.

2) central tabulators.

Yes the central tabulator is almost always a computer and as it is a computer it can of course be hacked and compromised. Even if humans or trained monkeys were doing the central tabulation, this is always a critical point in the voting process that can be targeted and compromised.

How the votes are tabulated is pretty much independent of how the votes are originally recorded.

The best vote recording systems have a paper trail and can be recounted repeatedly by hand if need be. The worst kind have no paper trail at all and cannot be recounted. Mechanical (lever) and touchscreen systems are the absolute worst.

If the tabulation process is suspect (as it should be) having an verifiable recording system is essential. Individal precincts can be recounted by hand and compared against tabulation records. Discrepencies should show up if they exist.

[nicknameless]

23. Yes!

See reply #16

[merh]

27. helderheid - let me tell you what I have learned from working with

my clerk on this. Our county has used opti-scans and paper ballots since 1999. There have been 4 contested elections and when the paper ballots were counted and compared to the opti-scan results there was 100% accuracy. Now that is in my county, but that is 100% accuracy too.

I assisted in counting ballots years ago when counting was done by hand. It was a time consuming, confusing circus. The costs involved in hand counting, the stress on the individuals involved, the worry about the security of the ballots, the worry over the accuracy of the hand counts all lead election officials to search for a better way to count ballots and to run elections.

If you research case law regarding election contests you will find that most precedent involves hand counts and ballot stuffing, dead men voting and stolen ballots. To understand the issues one needs to study the history of elections and the evolution of elections systems. One needs to know why we went to automated counting over hand counts.

Think of the ballots today. Elections in most states include all federal, state and county elections, including elected offices, taxing, bond and gaming initiatives, and proposed amendments to the state constitutions. Ballots are not just a page with a name. Ballots are multiple pages with multiple names and issues.

Do you remember when the Florida recounts were done by hand in 2000? Do you remember that circus. Imagine the efforts involved if all counties throughout the USofA had to do hand counts. Sure, for the counties with small or minimal populations, hand counts work, but what about the large counties like LA county?

Folks like to say, time be damned, but that isn't the case. Do you remember how antsy folks were waiting on the results of the Florida recounts?

Elections are not adequately funded. The manpower necessary to conduct hand counts only is overwhelming. And how do you guarantee accuracy when you are dealing with humans, an imperfect being that makes mistakes.

We cannot get folks to turn out to vote, how do we get them to volunteer to help in the hand counts?

In my mind, the bottom line is to ask the election officials what they consider to be the best way to conduct elections. They are the true experts as they have the experience and most have been dealing with it for years.

Touch screens and e-voting are out of the question. But opti-scans with the proper auditing and voter verified paper ballots has been proven to be the most accurate for the counties with the large populations. If you go to the clerks of the large counties and insist they switch to hand counts, they will raise holy hell. They will turn their backs on election reform advocates that promote hand counts only and we NEED the clerks on our side. We need to embrace them, tap into their resources, their lobbying strengths and their experiences to be in a position to see true election reform.

[kster]

8. Only way to make it safe

fight for this http://www.wired.com/news/ebiz/0,1272,61045,00.html and if they won't give it to us demand, paper ballots hand counted.

[merh]

21. E voting - are you saying go with E voting?

Is that what you are advocating now?

[kster]

45. Paper Ballots Hand Counted And Recorded By The People

At The Precinct Level is what I advocate, e-voting to me Reeks. But the person who made the thread asked a question.

[diva77]

85. Are there innovations in hand-counting of paper ballots?

I thought I heard something about that somewhere. Also, I saw a picture of something that looked like a mechanical card-shuffler that was used to count paper ballots long ago...

[riqster]

89. Innovations? Sure

They lock us out while they do it

[Land Shark]

9. I think of optical scans as a potentially rescuable system

if there are real time audits by hand counting groups of randomized ballots. This way the machine is checked and the human counting groups are checked as well (any one of them that differs substantially from the others would be grounds for not releasing the results until a Count is done again, at the pre-recount pre-release of results stage).

Since recounts are expensive and rare, we can't credit recountability too much in terms of the pantheon of election itegrity.

yet it's still a lot better than touch screens, which are beyond all hope.

[CrispyQ]

11. Once the votes are digitized they can be manipulated.

The scanner system gives a paper trail but there is still opportunity for fraud.

[yowzayowzayowza]

14. IMHO, ...

COUNTING DEVICES: Any VVPB system (including optiscan) can be safe with proper manual quality control auditing.

TABULATION COMPUTERS: Safe as long as the base numbers are publicly available for verification; ie, the tabulator is essentially "computer assist" only.

[FogerRox]

17. Please stop using paper trial-- Opscans scan paper ballots-

trail is not in Blacks law--- record or ballot is in Blacks LAW-

even a DRE with a VVPR printer it should be called a ballot or REcord--

[yowzayowzayowza]

18. R U talkin to me?

n/t

[FogerRox]

26. Any one in genral that use the term paper trial--

We fought hard in NJ to get the sponser of our VVPR law to cahnge it--
But she did, ANd it passed -- its a VVPR VoterVOerified paper record-- the legal implications are huge--
Look up trail in a law dictionary-- it aint there

Look up record its there --

Please-- ?

[nicknameless]

16. Optical-scan machines have screwed us over big-time. Think 18,181.

All scams below occurred on optical-scan vote tabulators.
Excerpt from:
http://www.pww.org/article/articleview/4642/1/197

A Nov. 7, 2002, Associated Press story datelined Snyder, Texas, reported, “A defective computer chip in the county’s optical scanner misread ballots Tuesday night and incorrectly tallied a landslide victory for Republicans.” Poll workers became suspicious. As a result of those workers’ inquiry, a new computer chip was flown to Snyder, Texas, from Dallas. Once the new chip was installed, the computer verified that the Democrat had won the election. The question remains: Was this an innocent computer glitch or something far more sinister, an attempt to steal that election for the Republicans?

In a July 30, 2003, article published by Alternet.org titled, “The Theft of Your Vote is Just a Chip Away,” writer Thom Hartmann reported on another “Texas anomaly.” In the November 2002 election in Comal County, Republican state Sen. Jeff Wentworth won with 18,181 votes, Republican Carter Casteel won a state House seat with 18,181 votes, and Judge Danny Scheel (a conservative) won his race with 18,181 votes. All three in the same county, same day, same year, all three with exactly the same number of votes. However, no poll workers in the county asked for a new chip.

http://www.independent-media.tv/item.cfm?fmedia_id=8705&fcategory_desc=Under%20Reported

In the Alabama 2002 general election, machines made by Election Systems and Software (ES&S) flipped the governor's race. Six thousand three hundred Baldwin County electronic votes mysteriously disappeared after the polls had closed and everyone had gone home. Democrat Don Siegelman's victory was handed to Republican Bob Riley, and the recount Siegelman requested was denied. Three months after the election, the vendor shrugged. "Something happened. I don't have enough intelligence to say exactly what," said Mark Kelley of ES&S.

<snip>

This crushing defeat never happened: Voting machines failed to tally “yes” votes on the 2002 school bond issue in Gretna, Nebraska. This error gave the false impression that the measure had failed miserably, but it actually passed by a 2-to-1 margin. Responsibility for the errors was attributed to ES&S, the Omaha company that had provided the ballots and the machines. According to the Chicago Tribune, “It was like being queen for a day -- but only for 12 hours,” said Richard Miholic, a losing Republican candidate for alderman in 2003 who was told that he had won a Lake County, Illinois, primary election. He was among 15 people in four races affected by an ES&S vote-counting foul-up.

<snip>

According to the Wall Street Journal, in the 2000 general election an optical-scan machine in Allamakee County, Iowa, was fed 300 ballots and reported 4 million votes. The county auditor tried the machine again but got the same result. Eventually, the machine’s manufacturer, ES&S, agreed to have replacement equipment sent. Republicans had hoped that the tiny but heavily Republican county would tip the scales in George W. Bush’s favor, but tipping it by almost 4 million votes attracted national attention.

<snip>

It’s great that ES&S optical scan ballots are paper, but it’s the way they’re counted that screws Dems over. That’s how Chuck Hagel got his amazing upset Senate win in 1996. (ES&S counted an estimated 80% of his votes.)

Much, much more at the link. Chilling information ... once you get past who the author is.

[yowzayowzayowza]

19. Proper QC auditing would have..

caught all those situations. The machines are not the problem. The lack of system-wide consistent quality control is the problem.

[nicknameless]

20. AGAIN ??!!!!!!

Oh right, "Quality Control" is the answer!

:rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl:

[yowzayowzayowza]

22. Yep.

Nothing but histrionics from you again.

[nicknameless]

24. "Histrionics" -- Where?

I documented optical scan FRAUD and YOU call it "HISTRIONICS?"

Why exactly are you such a defender of these election fraud machines?

More hooey from you about "quality control."
These machines are doing EXACTLY what their makers intended for them to do -- STEAL ELECTIONS!

[yowzayowzayowza]

25. Machines don't steal elections...

PEOPLE DO because the election process does not check the machines work. Not rocket science; qc processes are used in manufacturing the world over.

Looks like ridicule to me> :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl:

Main Entry: his·tri·on·ics
Pronunciation: -niks
Function: noun plural but singular or plural in construction
1 : theatrical performances
2 : deliberate display of emotion for effect

[nicknameless]

28. Those doing "Quality Control" in the arenas you cite INTEND for their

machines to function PROPERLY!!!

Those making voting machines INTEND for their machines to malfunction, vote-flip, count backwards, boot up with negative numbers for the Dems, etc.

I WAS ridiculing your post. Not for effect -- for REALISM -- which your post, as usual, was lacking.

[yowzayowzayowza]

32. QC requirements ...

...are decided upon based on many parameters including duty, risk, failure modes and even the possibility of industrial espionage. QC is designed to capture the malfunctions regardless the cause.

Please explain how a 2% manual precinct audit would not have caught the FAILURES in the situations you posted above.

Answer: YOU CAN'T.

[nicknameless]

39. Where a recount is done, errors CAN be caught.

(I would certainly recommend a higher than 2% audit.) But WHO is doing the audit?

And what do you do, when in state after state -- nearing if not already in all 50 states -- DREs and tabulators designed to defraud the Democrats are in place?

Look at the supposedly random recount in Ohio in 2004. MORE fraud was committed!

I don't think that this problem will EVER be solved until the vote-stealers are removed from the issue.
No more Diebold, ES&S, Sequoia, etc.

[yowzayowzayowza]

42. Um....

I don't think that this problem will EVER be solved until the vote-stealers are removed from the issue.

The human condition makes that a pipe dream. ...so much for you vaunted REALISM

Look at the supposedly random recount in Ohio in 2004.

We're not talking random recounts, rather inline precinct-level qc auditing.

Without VVPBs and random precinct-level audits, DREs are an abomination.

[nicknameless]

44. So ... we're stuck with the vote-stealers?

Why? Why do you want it to be that way?

Yes. I want hand counts. Take the MILLIONS spent on vote-fraud machines and spend it instead on hand counts. Just as is done in other countries. You want to go tell them that their reality is a "pipe dream?" "You're not really counting your votes by hand. You're deluded."

With numerous machines, in maybe all 50 states, you're going to audit them how? By machine?

[yowzayowzayowza]

47. Yep.

There will always be potential thieves that the system must be capable of dealing with no matter the level of automation. Hava problem with that? I suggest prayer. Good Luck. Keep me posted.

you're going to audit them how? By machine?

My quick and dirty recipe for secure elections:

I. Precinct Process:
1. EVIDENCE - collect VVPB from each voter whether hand marked or computer assisted
2. COUNT - tally the total number of voters
3. AUDIT - count the votes manually of a specified percentage of ballots
4. REPORT - post publicly the total voters and audit counts

II. Automated Counting: (whether centralized or precinct)
1. TOTALS - publicly post counts for each precinct in total and by race
2. AUDIT - publicly post counts of each precinct's audit ballots

III. Tabulating: From DREs to central vote tabulators, automated vote totaling should NOT be an integral part of the election system, rather a convenient adjunct.

IV. Solid "Chain of Custody" & "Many eyes"

V. Full Manual Recounts:
1. AUTOMATIC - close races within a specified percentage
2. AUDIT - Precincts whose manual and automatic count differ
3. PAID - fee/precinct/race

In short: VVPB + (precinct-level QC audits) + (Public Posting)

For most precincts the manual audit would total only a few dozen ballots, not a major burden on election nite.

[nicknameless]

49. It’s fascinating how you continue to defend the vote-stealers.

FYI, they're criminals. They should be serving time, not raking in profits off of defrauding the voters.

“There will always be potential thieves that the system must be capable of dealing with no matter the level of automation.”
When we’ve removed the automation. we’ve removed the majority of the fraud.

“Hava problem with that? I suggest prayer.”
No thanks, I prefer hand-counts. It works fine in other countries -- places where the counts actually DO match the exit polls.

“For most precincts the manual audit would total only a few dozen ballots, not a major burden on election nite.”
A “few dozen ballots” … on “election (night)”, no less. Democracy isn’t worth much to you, is it?
Impatience over counting votes, from people like you, has given us the Administration From Hell. Thanks.

Having ANY machines do ANY vote-counting will assure us margins that don’t require a recount.
And what about the FIRMWARE?
See: http://www.onlinejournal.com/evoting/042804Landes/042804landes.html

Two voting companies & two brothers will count 80 percent of U.S. election using both scanners & touchscreens

<snip>

Even if states or counties hire their own technicians to re-program Diebold or ES&S software (or software from other companies), experts say that permanently installed software, called firmware, still resides inside both electronic scanners and touchscreen machines and is capable of manipulating votes. For those who are unfamiliar with the term “firmware,” here's a definition by BandwidthMarket.com: "Software that is embedded in a hardware device that allows reading and executing the software, but does not allow modification, e.g., writing or deleting data by an end user."

The ability to rig an election is well within easy reach of voting machine companies. And it does not matter if the machines are scanners or touchscreens, or are networked or hooked up to modems.

So, for those states and counties who think they're dodging the bullet by not buying (or not using) the highly insecure and error-prone touchscreen voting machines (which will process 28.9 percent of all votes this year), a huge threat still remains—computerized ballot scanners. They will count 57.6 percent of all votes cast, including absentee ballots.

And don't count on recounts to save the day. In most states, recounts of paper ballots only occur if election results are close. The message to those who want to rig elections is, "rig them by a lot." In some states, like California, spot checks are conducted. But, that will not be an effective way to discover or deter vote fraud or technical failure, particularly in a national election where one vote per machine will probably be enough to swing a race.

<snip>

[kster]

50. Indeed

Snip..... All of the machines at the demo provided a VVPR and met HAVA requirements for the disabled. ((((((((((((((None of the vendors said they were willing to make their software source code public – they say it is proprietary, a trade secret.))))))))))))))
Open source code would be required by SB1683 which was introduced in the State Senate by Don Harmon on 2/24/2005.


http://chicago.indymedia.org/newswire/display/56327/index.php


Why do we let them get away with this, if they are selling the machines for PUBLIC elections, they should be open to the PUBLIC?

If I was responsible for buying the voting machines I'd be :rofl: if the salesperson told me that. And they would have definitely lost the sale.

[nicknameless]

53. Thanks and Amen!

If they're so worried about trade secrets, they need to get out of the business of public elections.

And what about the FIRMWARE as mentioned in my last post's excerpt? Beyond the software and the firmware, what else is there? A nice shiny computer case.

Bring back paper ballots and hand counts!

[yowzayowzayowza]

55. You two together at last.

A match made in ... er sumwarez. Can I get an AMEN?

[kster]

56. Hey,You stole my line

I could not even explain to you how obsessed I am with counting methods right now.

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=386947&mesg_id=386951

[yowzayowzayowza]

60. The only method I've seen you obsess over is...

"Paper Ballots Hand Counted and Recorded By the People." What other "counting methods" have you obsessed over? I mustu mist it.

[merh]

64. You get legislation passed that does not allow them to

keep their "trade secrets" - you know, you have yet to reconcile your stance with what I asked below. And yes, I intend to continue ANDY's message and not distort it.

Not all optical scanners are bad devices, not all election officials are corrupt, and not all precincts/counties can afford in time, manpower or money, to count by hand. Not all states have had corrupt elections.

Your broad brush approach and total failure to appreciate (or even try to find out) all that is involved in conducting elections clearly reflects your tunnel vision.

And before you start throwing out your usual insults, let me say that:
(1) no, I am not for the evil machines that have stolen elections
(2) no, I am not for the repukes and their theiving ways; and,
(3) yes, I believe that misguided and/or ignorant "election reform" activitists actually do the effort more harm than good.

I am for fair and open elections, I believe what Andy stressed about opti-scans is accurate and is what our focus should be (so does John Dean, the leader of our party, but what the heck, Andy said it first).

I believe in reaching out to the clerks/election officials and in trying to work with them and to try to help them, in an effort to have them (and their lobbying base) on our side, in our corner.

I believe in trying to write and then fighting to pass election legislation that will provide for open and honest elections, everything transparent, no trade secrets, no partisian ownership of businesses that provide the equipment, no campaigning for candidates or endorsement of candidates. Election day should be a legal holiday or held on a Saturday so that folks can actually take the time to go vote.

I believe in adequately funding elections.

And I believe in trying to have some understanding of the process that I am trying to reform or correct.

:hi:

[kster]

71. You got that right

them e-voting machines are like buying that fancy red sports car, you put special locks on it, you buy the best alarm system, you push for tougher auto theft laws, and as sure as I am sitting here the damn car still gets stolen, there is just no real way to secure it from the crooks, thats the same situation we are in with them damn e-voting machines, there is no real way to secure them from the crooks.

PBHC in public view, only way to go!

[GentryLange]

101. Firmware...

Couldn't we just reflash the firmware with publicly owned open sourced code?

As someone who worked in embedded systems, this seems easy. Then the Optical Scan machines have a VVPB, if used. So the hand audit is essential too.

But I don't think machines are incapable of being put to good use. Just not with the corporate control and disengenious safety and accuracy procedures.

[yowzayowzayowza]

54. WOW.

FYI, they're criminals. They should be serving time, ....

Thieves are criminals? NO WAY!!! Thank you soooo much. I never would have figgered that out on my own.

Democracy isn’t worth much to you, is it? Impatience over counting votes, from people like you, ....

As I have too much respect fer this place to say what deserves to be said...

Discussion Forum Rules
...
2. Civility: Treat other members with respect. Do not post personal attacks against other members of this discussion forum.
3. Content: Do not post messages that are inflammatory, extreme, divisive, incoherent, or otherwise inappropriate. Do not engage in anti-social, disruptive, or trolling behavior. Do not post broad-brush, bigoted statements.

[Bill Bored]

65. You have to audit complete precincts or at least complete machines

or scanners.

The only way to tell that the count matches is to count all the ballots cast on a randomly selected machine or scanner and compare the totals. You can't count half a machine because you won't know:

a) that the other half of the ballots were counted correctly

b) that the half you count is comprised of the same ballots that were counted on the machine (unless you give each one a serial number).

As far as a 2% audit catching all problems in a jurisdiction, see this paper by Kathy Dopp to view the odds:

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

I agree you don't have to hand count EVERY ballot, but you do need to be able to count enough to verify that the election outcome was correct. And this depends on:

- the number of precincts
- the number audited
- the number of problem precincts found
- the impact of the errors found
- the margin of victory overall, in the audited precincts, in the unaudited precincts
- the number of votes in the non-audited precincts

(Thanks to Ellen Theisen of Voters Unite for this list!)

So what you need, short of some formula that could take into account every possible scenario, is a LAW that forces a BoE or a Court to order enough of a hand recount to mathematically rule out the possibility of an incorrect outcome.

Make sense?

PS - Oh, and open source code won't do anything to prevent accidental or deliberate mis-configuration! Bi-partisan auditing of that might though.

[yowzayowzayowza]

67. Yes and No.

Industrial QC has been VERY successful the world over by monitoring portions of EVERY production lot. You don't allow a wholly independent process to complete without verification. The potential cost of a mass failure is too high.

The problem I have with the Kathy Dopp analysis is that it assumes that the audit is performed after the election after the damage is done. Effective QC must be an integral part of the process, not simply an annoying adjunct. Every single count MUST be audited as a routine part of the election counting process. The QC audit should be performed as part of the routine election closing procedures of EACH precinct, not some selection of precincts after the election is already reported.

If the average precinct has 1000 ballots, a 2-4% precinct-level audit would be every 50th-25th ballot or 20-40 ballots. To insure a homogeneous test the batch of audited ballots should be counted in the same series of batches as the balance of the population. Hand counting that quantity of ballots by the already trained and in place precinct personnel is a VERY small price to pay for the knowledge that the remainder of the ballots for each precinct are accurately counted ON ELECTION NITE.

[Bill Bored]

68. You still need serial numbers on every ballot to compare the VVPAT

to the DRE vote database. In fact, the database has little to do the the ballots. With Diebold's junk, the candidates' names in the database don't even have to match those on the ballot!

You are suggesting that we need to be able to track individual ballots straight through to the database where the votes are counted? I don't know if this can even be done on the junk that's out there now.

There may also be ballot secrecy problems with making this possible. You can already track individual ballots by the order in which voters cast their votes at the precinct, esp. with VVPAT paper rolls. There are already ballot secrecy issues with that, but extending it to the counting software is what you're talking about in order to be able to do a partial audit of a machine.

I'd rather see them pull random machines out of service during the election and test them with test votes all day long, using the actual ballots and database. You'd have to be careful not to count them though!

And none of this addresses the upline tabulator issue whereas random VVPAT audits and manual recounts would.

[yowzayowzayowza]

69. VVPAT is garbage.

To me, an election is about the collection of evidence of the will of the people, not building an unverifiable database; that means VVPBs. Afaik, currently deployed DRE-style databases would not be allowed as evidence in a criminal or civil case in ANY court. The only reliable method do validate DREs would be with BOTH a full automated VVPB count and an audit batch. IMHO, automated tabulation should not be integral to the election process, rather a convenient adjunct.

[Bill Bored]

72. DREs do not have VVPBs. They have VVPATs.

Do you understand the difference?

You can hand count either one, but the voter fills out the VVPB while the DRE prints the VVPAT. The law determines when either is actually counted or is used for the official count.

[yowzayowzayowza]

73. Yea, I've no interest in a trail.

The VVPAT is intended to be deposited into a ballot box by voters prior to leaving the polling area and thus creates a paper trail that can be used for audit in the event of a recount or machine malfunction.

As a voter I deserve the EVIDENCE of my will, a ballot, to be paramount; not merely a trail subject to some nebulous database application. The DRE must print a completed BALLOT for each voter(VVPB), which are in turn counted to decide the election and verify (audit) the counting process. The DRE's database and resultant "upline tabulation" should have NO STANDING in the election what-so-ever, rather merely an adjunct reporting convenience (similarly GEMS tabulation of opti-scanz). As your previous post aptly describes, anything short of a complete manual recount of a VVPAT requires a ridiculous amount of juggling. I've no reason to trust ANY unverified counting solution. NEVER. EVER. No mandatory system-wide QC = NO TRUST. PERIOD.

On edit: To put it another way, the only way I could stomach a DRE VVPAT:

1. The VVPAT must print in a format recognizable by man and separate counting machine.
2. *ALL* VVPATs must be counted by a separate counting machine to verify the DREs numbers.
3. A quantity of each precincts VVPATs must be hand and machine counted to verify the counting machine.

Otherwise as I stated above "VVPAT is garbage"

...and Good Evening fer now.

[demodonkey]

75. NO! And do not call it a "trail"... remember ANDY'S WARNING!

DRE certainly CAN print a VVPB.

What comes out of the DRE can be a BALLOT, or a "record". Good legislation will at LEAST require the VVPR (and if not a ballot at least call it Voter-Verified Paper RECORD, as used in Rush Holt's bill HR 550) to take precedence over the machine count and be used as the official record (sometimes this is worded "become the ballot of record".)

So whether or not it is a VVPB or a VVPR is a matter of the law.

NOW -- as to the "trail" word. DO NOT USE THIS. "Ballot" or "Record" are a better choice. If I learned ANYthing from Andy it was NEVER to say paper "trail" for the reasons in my SIG, which was really his.

Andy always warned that "trail" is not a legal term and allows the voting machine companies too much wiggle room (such when they claim that the worthless printout tapes inside the machine, which the voter never gets to see, or after-the-fact printouts of electronic "ballot images", are an official "paper trail".)

HERE -- AGAIN in case the sigs are not always visible -- is my (Andy's) SIG:
When no statutory definition for a term exists, the first fallback in all courts is to consult the definition in Blacks Law Dictionary. a compelling case can be made for the fact that ballot is defined in Blacks Law Dictionary...Paper is defined in Blacks Law Dictionary but "Trail" is completely absent. Collectively the definition for ballot and the definition for paper generate a clear and easily understood legal definition of the paper ballot needed to ensure the integrity of our vote. Paper trail produces no such record. (Andy Stephenson)

[merh]

29. I don't understand..

Why do you keep Andy's image up yet you continue to counter, hell laugh at, everything he said about proper audits, opti-scans and voter verified paper ballots. :shrug:

Nothing you profess has answered any of the concerns regarding the logistics and the costs involved in "hand counts only". Can you answer those concerns? Have you taken the time to ask your clerk what all is involved in conducting an election and how the clerk feels about your stance "hand counts only"?

With proper auditing and truly random recounts, optical scans are the safest and most accurate way to count an election. The trouble starts when the votes are sent via electronic means to a central tabulator. I would also add that the tabulator needs to be as secure as Ft. Knox because after all, our votes are more precious than gold." ~Andy Stephenson, 4/29/2005

Edited to add the link to Andy's course notes about auditing elections.
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=384082

[FogerRox]

31. Diebold ATM use open sorce software

Diebold DREs use propiretary source code

An Accountant that has 2 sets of books goes to jail-
DREs have 2 sets of books too-

Why would a DRE need a front end and a back end? unless like an accountant with 2 sets of books who was skimming off the top--

You all get the idea--

DREs are built to cheat---- ANd the ballot images are stored in corporate owned software so NO ONE can EVER EXAMINE the vote counting mechanism.

But the Banking industry wouldnt stand for that soooooo they got good equipment.

[merh]

34. I'm not advocating for the Diebold DREs

On the contrary, I am trying to keep them out of my county.
Don't confuse the DREs for the properly audited and controlled opti-scan systems that are in use and that have been accurate.

With proper auditing and truly random recounts, optical scans are the safest and most accurate way to count an election. The trouble starts when the votes are sent via electronic means to a central tabulator. I would also add that the tabulator needs to be as secure as Ft. Knox because after all, our votes are more precious than gold." ~Andy Stephenson, 4/29/2005

Explain to me how you would conduct elections in the counties with large populations. How do you suggest that those counties conduct the counting by hand?

[nicknameless]

33. I HAVE DONE NONE OF THE THINGS YOU'RE SAYING!

I agree that optical scans, because they have paper ballots, are the best that the voting machine companies offer us.

How DARE you say that I'm laughing at the things that Andy promoted!
You obviously have appointed yourself as the crusader for his cause, but WITHOUT his capacity for reasoning.

You disingenuously pretend that hand counts can pull off the same massive fraud that machines do. Andy wouldn't do that.
Why are you so completely dishonest about that?

I would be in favor of audits, if the Repugs would allow them. Look at what went on in Ohio in 2004 when they were supposed to do a random recount. MORE FRAUD!

I have answered NUMEROUS TIMES your BS about the cost of hand counts. Look at the MILLIONS spent on vote-stealing machines. What if that money were instead spent on honest counting? You obviously prefer the Repug landslide tabulators.

[merh]

36. The opti-scans with paper ballots is what the other poster is

suggesting. He said that quality control is the key and you scoffed at that. Proper auditing is quality control, just as keeping the "central tabulator ... as secure as Ft. Knox".

Hand counts are not practical in the counties with the large populations and to revert to hand counts is reverting to operations of the past that did not work. They didn't work and that is why automated systems have been developed.

They conduct the auditing in the counties in my state that use the opti-scans. It is not an either or proposition as you would like to believe. Hand counts only will not endear you to the elected officials that conduct elections.

If legislation is needed to provide the proper audits and quality control, then let's write it and push for its passage. Let's get the clerks on our side and work with them. Let's be practical about it and work to improve elections, not work to return to the problems of the past.

I see the issue from a practical standpoint, you see it from pie in the sky, this is what I want. Money be damned, well gee, that will endear the clerks who have to budget their funds to conduct the elections.

Do you know what all is involved in conducting an election? :shrug:

[merh]

37. Funding has been a number one problem for election officials

for years.

LEAGUE OF WOMEN VOTERS OF OHIO FILES HISTORIC LAWSUIT

In a complaint being filed today in federal court in Toledo, the League of Women Voters of Ohio joined the League of Women Voters of Toledo-Lucas County and a dozen individual plaintiffs in suing the governor and the secretary of state for 30 years of dysfunctional election administration. Represented by a team of attorneys led by the Lawyers’ Committee for Civil Rights Under Law, the League is claiming a pattern of failed election management, chronic under-funding of county boards of elections and inadequate training for poll workers.

http://www.lwvohio.org/press/suit.htm

[helderheid]

30. OK, so we should add the tabulation machines to those we want with open

source code and mandatory audits, from what I'm understanding here.

[yowzayowzayowza]

35. IMHO, with proper auditing...

escrowed source would be satisfactory. Open source is no panacea.

[FogerRox]

38. Source code to be held in escrow if irregularities a rise

[merh]

40. Yes.

We need legislation passed in every state that provides for the open source codes, the mandatory audits and the quality control necessary to keep the elections honest.

To get the legislation written and passed, we need to work with the election officials in our states. Not the SOS, the county folks that have to conduct the elections, that have conducted the elections and know what all is entailed in conducting an election.

We need to be their advocates, not their opponents. We need to be working with them and for them. We need to help them get the proper funding so that they can provide the quality control procedures necessary, so that they can have the manpower necessary to conduct elections.

If we don't work with them, our efforts are doomed. There are those that profess to be election reform activists that have frightened, bullied and doubted otherwise honest election officials. Their actions have officials throughout the country skeptical of election reform activists. We need to try to overcome that by going to the officials in our states and working with them. First we have to ask them to tell us what they need, what would be the best way to conduct an election, all that is involved in conducting elections.

We need them in our corner if we are ever going to be successful, imho.

[adolfo]

83. Agreed!

Merh, I've always enjoyed your posts and appreciate your insight.


:toast:

[GentryLange]

102. IMHO that depends on the election officials you encounter..

Some elected officials are your friends, some like the free Wine, the nice conferences, and the cruises bought and paid for by the Corporations, generally these second types aren't too interested in election reform activists.

I think there's two clear types of elected officials, those that fight against corporate power, and those that fight for more corporate power. Working with the first type is a great goal, but that second group needs removed from power.

[nicknameless]

57. But what do you do when the firmware is rigged?

See link & excerpt in post #49

[yowzayowzayowza]

58. Whether the software is written...

to a hard disk or a chip, you do the same thing: MAKE THE MACHINE PROVE ITZ VIABILITY AT EACH AND EVERY USE. I know itz a difficult concept that makes you yearn for the simplicity of the dark ages, but I'm sure eventually you'll catch on. Keep up the good werk.

[nicknameless]

59. I certainly wasn't asking YOU.

[helderheid]

63. Anyone who knows me on this forum knows how anti diebold, etc., I am

I am just trying to make sure that if the firmware is gotten rid of, that our votes don't continue to get manipulated at the tabulation level. Hope this is clear now.

[nicknameless]

66. I think most DUers are anti-diebold etc. vote-stealing machines.

I would never suggest that you were otherwise. Your thread seemed to be an honest quest for answers. I'm also looking for answers. (And not looking to be involved in this Punch-and-Judy devolution that's occurring so many places on DU.)

My question is a serious one that I've not heard addressed. If you also don't know the answer, that's fine. Just asking. I probably need to track down someone who is actively working on this issue on the technical side, as I'm desperate to make changes in my state.

[timewellspent]

41. Clermont was not the only county to use Opti - Brown did also

I have been researching this subject and it is amazing to me that everyone thinks it is a scam. It is beautiful. Hackett did what he should of done and it is a tribute to the hard working Democrats in the second district of Ohio. They worked hard and almost made it happen.

I was helping in Brown County and it is not much mentioned in here that they also used a precinct count optical scan system that worked all day long in the heat and humidity.

Brown County had a good showing of Democrats defending the district. It is time to do the work before the election instead of complaining about it afterward.

That being said, Clermont had a central count opti system and brown had a precinct count system. Humidity could of been a problem in Clermont, but they ended up getting them all thru the reader.

I have to commend the Dem's in this one. They made it happen. Dem's snuck up on the Rep's and made it happen. We were going to the polls and seeing who hasn't voted yet and getting those people out to vote. It happened in Brown County.

Let's not get down about this election, but let's learn from it. Go to your Election's office, BEFORE the election and check it out, get involved. Make sure you ALL do that before the election. That is the time to make noice. Once it's over, it's over.

[merh]

43. Welcome to DU, timewellspent!

:hi:

I am so glad that you are advocating folks get involved and learn about the process now, before it is too late. We have to work with the folks that conduct elections. We have to know what it all entails so that we can be on the lookout for irregularities or attempts to corrupt the process.

If we want election reform, we have to work with the folks that conduct elections, not work against them.

Thanks for your post! :patriot:

[Bill Bored]

46. Precinct totals should be compared to central totals on the Internet.

See Section VIII about this in the DNC Ohio report.
http://www.democrats.org/a/2005/06/democracy_at_ri.php

We also need random auditing of Op Scan ballots and DRE VVPATs alike. This will happen in NY according to the new law which makes no distinction between voting machines (DREs) and voting systems (Op Scanners). A good thing, although it would have been better to simply ban DREs.

Also, we need recount laws which allow hand counts of the paper ballots or VVPATs under conditions where the possibility of an incorrect outcome of the election is unlikely to be detected by random auditing or any other means. NY has a law about this too, but it needs some tweaking. See this thread:
<http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=386269&mesg_id=386269>

Op Scans are still better than DREs because the ballots are better. Voter error rate is lower. Machine error rate may not be any better than DREs because they are both programmed the same way through GEMS or similar Election Management Systems. But at least with Op Scans, the ballot design itself is frozen when it goes to the printer. You can't change the ballot on the screen at the last minute to confuse the voters and you can test the scanners with large numbers of real ballots fairly easily.

Op Scan ballots are easier to hand count too. If the paper is of a certain weight, you can even count the ballots by sorting and weighing them! As long as no one puts their big fat Republican thumb on the scale! (Sorry, I couldn't resist.) :)

On edit: having read the above debate about hand counting, I'd say Andy was right, because of the number of races on some of these ballots. But when it comes to a recount of a single race, Op Scan ballots are very easy to count by hand and could actually be weighed!

[merh]

48. Thank you for posting a reasonable post.

I do appreciate it!

Howard Dean said opti-scans are the way to go! :patriot:

[garybeck]

51. opscans with random audits

I think if there are random audits we are not screwed.

but i agree, without audits we are screwed.

------------------------------------
the solar bus
ELECTION JUSTICE CENTER
your home for updated information on the fight for democracy in America
http://election.solarbus.org
------------------------------------

[Wow]

52. If we want a printout

Can't they just print out as many copies of the vote the way they want it as they please. Then what we are recounting is what they had their printer spew out.

[eridani]

61. Two advantages of opscan

1. Voting is separate from tabulation. Voting is slow, and tabulation fast, so that people with opscan ballots who find the privacy booths full can just go somewhere else in the school library to fill the ballots out. One optical scanner took care of 10 precincts at my polling place. DREs, in addition to all their other faults, clog up the system because only one person at a time can use them. Therefore, they are essentially one more form of vote suppression for busy people.

2. With precinct auditing, you are counting two different ways, which serve as crosschecks. If the results come out the same, that increases credibility, just like when paleontologists find that their tree-ring data and their 14C data agree with each other.

Consider that Canadians count rapidly because they never ever have complicated splits, like our legislative district electing state representatives and also voting in three separate Congressional districts. Provincial elections are separate from national ones, and they never try to do more than two things per election.

[riqster]

62. The only way to make any voting system work is transparency

If you have hand-counted paper ballots, you have to have the tabulation process viewable by the public, and audited real-time by impartial observers-or, failing that, representatives of each party involved.

If you use any other method, the black box tabulating model WILL NOT WORK. It has to be open-source and every step of the process has to be quantified, verified, and monitored real-time and after the fact. Any technology that does not allow for these common-sense security measures should be tossed out, and urinated upon into the bargain.

DREs and lever machines are inherently flawed because there is no auditable record of voter input, only the machine output. Punch card and optical scan devices at least offer both datasets, IF controls are in place. These controls must have the force of law and be unambiguous, or else they are worse than useless.

[kster]

70. KICK.NT

[kster]

74. Helderheid, great question

this thread got us in to the "nuts and bolts" of e-voting, whether people are for it, against it,or just plain not sure. There was a lot to be learned from this thread. Good Job.

[helderheid]

76. Thanks :)

I just wanted to make sure nothing "slipped by" if we got "almost" all we needed for fair (fairest possible) elections. :)

[kster]

77. kick.nt

[Time for change]

78. Central tabulators can be screwed with just like regular voting machines

In fact, it appears that in Ohio, that was the major mechanism by which the election was stolen.

The DNC report concluded that "there was no widespread fraud" solely on the basis of high correlations between the Kerry vote and the Fingerhut vote and other variables, such as Issue 1, percent African Americans, and the Hagan vote. But they didn't EVEN CONSIDER the possibility that the central tabulators were used to decrease the total votes in Democratic precincts. That could have been done without affecting the percentages within precincts, and without affecting the correlations that they used to claim "no widespread fraud". That appears to be the main mechanism that was used in Ohio.

It seems to me that one way to guard against this would be to have one person in every precinct document the final count as the polls close. This could later be compared to the counts calculated by the central tabulators. If they widely differed, that would tell you where the problem was.

[riqster]

79. Actually, that is only one method that was used.

One of the biggest scams was voter registration fraud. If your records were messed with, you were either not registered, or had to 'prove eligibility', both of which required a provisional ballot.

Then, they didn't count the provisionals....

[Time for change]

95. Absolutely -- thank you for stressing that

Registration fraud was almost certainly an issue in this election.

I believe that it could have involved one or a combination of two different mechanisms: Massive purges of registration in Dem. areas, or fixing of the registration numbers after the election to hide the fact that tens of thousands of votes were deleted electronically from Democratic precincts.

[kiwi_expat]

84. Calling liam_laddie

Liam_laddie should be able to tell us if this is possible.

Liam, based on your understanding of how Hamilton County checks their totals, do you think it is possible that the central tabulators were used to decrease the total votes in Democratic precincts?

If the SoS' total vote for a precinct was lower than the number of votes cast for that precinct (i.e., the number of cards in the precinct deck), wouldn't the county staff notice that? The entire county vote would be affected.

[Time for change]

86. I have discussed this issue with Voters Unite!, among others

I have asked them if we can compare the numbers noted at the time of poll closing with the final official numbers, as reported by the central tabulators. I have been told that this is generally not possible because those numbers were not obtained, either because nobody thought of it, or because it was tried, and the numbers were not available.

[helderheid]

87. Just seems to me this would be the easiest way to swing an election as

demonstrated by our own DNC Chair (regardless of what you think of the woman showing him how):

http://www.udpc.org/evote-lowband.htm

[Time for change]

96. I think so too

By having a central tabulator in Democratic counties delete votes or having one in Republican counties add votes in the same proportion as the actual count, you can cause a net change of tens of thousands of votes (if a large county such as Cuyahoga) without having to initiate fraud in the individual precincts.

As I explain in this thread (http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=132&topic_id=1974460&mesg_id=1974460), the registration data suggests that either this is what was done, or it was done through massive purges of voter registration in Democratic precincts -- as riqster points out.

The added advantage of doing it this way is that it won't change any of the correlations between the Presidential vote and other variables when analyzed by precinct. Then, if an analysis is done such as that used for the DNC report, it won't find any evidence of fraud.

[kster]

97. This is why

I question Deans role in all of this, he witnessed it all being done FIRST HAND.

But yet silence from Dean, and the DNC report about the machines.

What gives?

[liam_laddie]

88. answering the call, sort of...

This may be old news to most posters on these ER&D threads,
but anyone who hasn't, should log on to Prof. Douglas Jones'
website <http://www.cs.uiowa.edu/~jones/voting> and read his
work, especially "A Brief Illustrated History of Voting," "Counting Ballots," "Counting Mark-Sense Ballots" (about opscan), "Keeping Electronic Voting Honest," and "Evaluating Voting Technology." Incredibly useful to me when I began dipping my
political toe in the muddy waters of current election controversies. Pedantic mode off...

First, an election result is not considered complete (and not yet certified) until ten days past the election day, which allows time for absentees to arrive and be processed. Overseas and military ballots fall into this category; if they're postmarked by the election date (Aug 2 in the recent CD-2 race) - they're counted. I do not know if provisionals were used anywhere in CD-2. Will check.

Comparing totals between county and SOS office. I don't know
whether the SOS receives the result from a county file and adds it to other counties' totals, or if the SOS is tied directly into each county's compiler and thus is in real-time contact as the votes are adding up. Will check.

There is no way of knowing whether the result reported out of the
county's central compiler (the CPU) matches the input
from the tabulators. By that I mean, the digital data can be revised via software and short of a sizable audit/recount (is there a legal distinction?) I can imagine no way to have 100% confidence that the reported result is an accurate count of each and every voter's intent.

I wonder if the recounts done in Dec. '04 were merely repeating
the pre-programmed revisions possibly used in Nov. I know
Hamilton County seems to do a thorough pre-test and handles cards as accurately as possible, but once those votes are converted into electronic data, how can we be certain, without
hand audits? With Clermont County, the opscan sheets would seem to make an audit somewhat easier than with Ham's punch-cards.

I believe it is possible, perhaps probable, that county BoE
Board members, Directors and employees do NOT know
if their software is rigged. It would seem that the fewer who
know, the less chance for leaks. Thus it is likely that only the
vendors - ES&S, Diebold, etc. really know all the facts.

Decreasing Dem votes? Done by many means: registration files, provisionals, maybe electronic subtraction somewhere along the wires or web-packets, can't be sure,
I will ask the BoE about reporting procedures.

Sorry that I can't be more informative at this point, but will follow up. Three cheers for Paul Hackett and to all patriots! ...liam

PS - I think the MSM should start using the term "the Presidency"
to separate the office from the effing idjit sitting there.

[kiwi_expat]

90. Thanks liam. We just need the pre-tabulator TOTAL votes for each precinct

I had the impression that the following pre-tabulator totals are recorded for each precinct: (R) total regular ballots used; (S) total "spoiled" (damaged) ballots; (A) total legitimate absentee ballots; (P) total legitimate provisional ballots. If that is true, then the total pre-tabulator votes can be easily calculated: V = (R - S) + A + P.

That total (V) can then be compared to the SoS' tabulated results for the precinct (the sum of the SoS' totals of votes for each candidate and under/overvotes) to see if any votes have been deleted by the tabulator. Whether the counties actually compare those totals is another matter, of course.

Cheers.

[Time for change]

94. That's what I thought too

In fact, a conference I went to recently sponsored by Vote Trust USA handed out some material on that.

Whether that kind of accounting was actually done is another matter.

In any event, I think that this is a terribly important issue. If it wasn't done, then we should ask, "why not". If it couldn't be done, then I think that we need to make sure that we have that capability in future elections, and that we use it.

I believe that this could be a very important mechanism for decreasing the likelihood of fraud in future elections, if we make sure that the mechanism exists and that we use it wherever we can.

[kiwi_expat]

99. Ohio statute re. recording number of regular ballots used

I finally tracked down the Ohio statute involved:
http://onlinedocs.andersonpublishing.com/oh/lpExt.dll?f=templates&fn=main-h.htm&cp=PORC

"3501.26. Procedure when polls are closed.

"When the polls are closed after a primary, general, or special election, the receiving officials shall, in the presence of the counting officials and attending witnesses, proceed as follows:

"(A) Count the number of electors who voted, as shown on the poll books.

"(B) Count the unused ballots without removing stubs.

"(C) Count the soiled and defaced ballots.

"(D) Insert the totals of divisions (A), (B), and (C) of this section on the report forms provided therefor in the poll books.


"(E) Count the voted ballots. If the number of voted ballots exceeds the number of voters whose names appear upon the poll books, the presiding judge shall enter on the poll books an explanation of such discrepancy, and such explanation, if agreed to, shall be subscribed to by all of the judges. Any judge having a different explanation shall enter it in the poll books and subscribe to it.

"(F) Put the unused ballots with stubs attached, and soiled and defaced ballots with stubs attached, in the envelopes or containers provided therefor, and certify the number."


* * *

This does not cover absentee or provisional ballots, of course.

I do know that the total number of absentee ballots is available (somewhere) by precinct. That would have to be a pre-tabulator total. Tabulators can not distinguish between regular and absentee ballots.

Provisional totals by precinct might be harder - or impossible - to find. Richard Hayes Phillips indicates that the number of uncounted provisionals might not be available by precinct
http://web.northnet.org/minstrel/provisional.htm

[kiwi_expat]

100. "Valid"-prov totals ARE available by precinct. That's all we need!

Richard Hayes Phillips does indicate that the total of "valid" provisionals is available by precinct. He uses the term "valid" to refer to the provisionals that were not rejected for being cast in the wrong precinct etc.

It is just the total of "uncounted" provisionals that is not available by precinct. We don't need that total. (The "uncounted" provisionals are included in the SoS' under/overvotes for that precinct.)

* * *

So, by George, we do have enough information to verify the SoS' tabulator total-votes for each precinct:

Pre-tabulator total-votes = Regular ballots used (minus "soiled and defaced" ballots) plus "valid" absentee ballots plus "valid" provisional ballots.

Compare with tabulator's candidate totals' sum plus under/overvotes.

[Time for change]

103. That sounds great!

Are you going to pursue this?

Is there anything I can do to help?

[kiwi_expat]

105. Liam-laddie indicated he might try again to check the Ham County books

Hamilton County appears to run a clean operation, but they can't prevent the tabulator from being hacked.

Liam has been discouraged from recounting ballots by the Hamilton County staff, who are always claiming they are under pressure preparing for the next local election - the last one being CD-02.

Cincinnati's inner-city precincts' poll books would be a good place for Liam to start, when he gets the opportunity. Fortunately, this investigation does not involve handling the actual ballots.

[Time for change]

107. Let me see if I understand this correctly

If Ohio law was followed, there should be one or more witnesses at each precinct in Ohio who was given enough information that we should be able to use it to check whether or not the central tabulator in each county changed the final vote count for that county after it was reported from the precincts. Is that right?

And that's what Liam-Laddie will be checking on with regard to Hamilton County -- to see if the votes match up?

[kiwi_expat]

108. We are just talking about the TOTAL votes for the precinct, of course.

Yes, by inspecting the books, Liam should be able to get enough information to determine if the total number of precinct votes from the county tabulator is the same number as the pre-tabulator vote total. However, he could not verify the precinct's candidate totals without an actual manual recount of the ballots.

* * *

I only know the Ohio law with respect to totals of regular ballots. The precinct totals of regular ballots should definitely be recorded in the poll books.

I'm not sure where the totals of "valid" absentee ballots and "valid" provisional ballots are located. I only know that Liam was able to find the total numbers of "valid" absentee ballots for certain precincts and that Richard Hayes Phillips was able to find the total numbers of "valid" provisional ballots for certain precincts. I assume that the recording of both totals is covered by Ohio law, but I haven't been able to find the actual statute section(s) yet.

[Time for change]

109. I am very excited about this. I think this could be an important key to

figuring out what happened.

I'll bet that the totals don't match -- in fact I'll bet that they're way off, with the central tabulator results greatly in favor of Bush -- unless of course the data is obtained from people whom we can't trust.

[kiwi_expat]

110. I sure hope you're right.

"I'll bet that the totals don't match -- in fact I'll bet that they're way off, with the central tabulator results greatly in favor of Bush..." -TfC

I'm feeling more pessimistic. Hamilton County appears to have run a pretty clean and transparent 3% manual recount. There were few problems (other than the over/undervotes not being inspected for voter intent - which all counties failed to do). If the tabulator had deleted votes for any of those randomly selected precincts, it would have been discovered in the manual recount. Of course it is possible that the tabulator deleted some of the 97% votes that were not manually recounted.

One of the great things about your idea of just verifying the TOTAL votes for a precinct, is that a significant number of precincts can be checked with relative ease.


"...unless of course the data is obtained from people whom we can't trust." -TfC

The regular-ballot totals are signed off by 4 election workers. Let's hope they are honest.

[Time for change]

111. I don't know much about Hamilton County

The counties I'm most suspicious of are Cuyahoga and Warren Counties. My guess is that there was fraud involving the central tabulator in several counties, but probably not every one.

You've probably seen this, but see what I have to say about Cuyahoga County in these two posts. Even the DNC report recognized that there were some very strange things about Cuyahoga County:

Registration anomalies, especially in Cuyahoga County:http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=1974460

See item # 2 here, regarding DNC report:
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=380878

[kiwi_expat]

112. I agree with you about Cuyahoga and Warren Counties - and Clermont...

etc.

The problem is that we do not have any volunteers to check those counties. Liam has been amazingly persistent with the Hamilton County staff, and has had some limited success as a result. Anyone with a halfhearted approach to auditing a county's books is not likely to get anywhere.

Maybe someone (you?) could initiate an exchange with a county's staff by e-mail. Liam has a good contact in the Cuyahoga County staff who might be able to assist setting up an appointment. Perhaps someone from Kucinich's staff could then be persuaded to actually go to the BoE office and check the poll books.

[Time for change]

113. I will bring this up with Ray Beckerman's 'Ohio project'

I do volunteer work for them. Do you know them? Hopefully we can work out a way to do this.

[kiwi_expat]

114. Good idea.

I don't know anything about the "Ohio project". Is there a website?

[Time for change]

115. Here ya go:

http://ohioelection2004.com/

And here's a copy of the e-mail I just sent them:

I believe that there is a good deal of evidence that fraud in the 2004 Ohio Presidential election was perpetrated largely by having central tabulators electronically change the vote counts that were reported to them.

In this thread, see item # 2 for evidence that this happened in Cuyahoga County and item # 5 for evidence that it happened in Miami County (both using evidence reported by Richard Hayes Phillips): http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=380878

And see this thread, especially the section titled "what is the significance of this massive new increase in Democratic registration", for a discussion of how this mechanism for fraud could make the DNC report claim of "no evidence for widespread fraud" useless: http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x1974460

The main reason I am bringing this to your attention now is that a discussion I have been having on the DU suggests that it may be possible, by comparing pre-tabulator vote totals to post-tabulator results, to show whether or not the totals were fraudulently changed. I am not very good at describing the central issues here, but a fellow DUer, kiwi_expat, seems to know a lot about this. Please see the discussion we have been having on this thread: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=387455&mesg_id=387455, starting with my post # 78, and going down to #114.

Would any of you know how to pursue the information necessary to obtain the data we need to show whether or not there was fraud involving central tabulators? I believe that Cuyahoga, Warren, Miami, and Clermont would be especially important.

[kiwi_expat]

116. Liam_laddie knows the most about the pre-tab totals.

Liam is the one who first told me about the "ballot accounting charts". He sent me a PM saying "Ham Cty has a page at beginning of each poll log book which has a chart to fill in - starting and ending total of punch-cards, ballots cast, spoiled, unused. The ranking member of the poll worker team records these, and at end of day, all four workers sign the page with name & address. One hopes they're honest". (I then found the appropriate reference in the Ohio Statute.)

Liam also knows how to get the valid-absentees total for a precinct. He got it for Cincy4M. And I'm sure he can find out how to get the valid-provisionals total for a precinct. The Hamilton County BoE staff have been very helpful to him.

I will be off-line for a week. Please PM Liam if you need more detail.

Cheers.
kiwi

[Time for change]

118. I wasn't able to get the needed info from the Ohio Project

So I'll pm Liam_Laddie about this.

[kiwi_expat]

119. How about contacting Kucinich's office for help with Cuyahoga?

Does anyone you know have any "in" with Kucinich? (Mgr, perhaps?) Perhaps someone on Kucinich's staff - or one of his keen volunteers - could be persuaded to check the Cuyahoga poll books. That would be ideal.

[Time for change]

120. I'll look into it, thanks n/t

[marions ghost]

91. helderheid

ABOUT FLORIDA--if you look at the screw-ups in the Florida counties that had newly acquired opti-scan you will find that a lot of the problems that were reported at the precinct level had to do with the fact that the operators were not trained correctly. The machines were pushed through very quickly, and they even had problems with the ballots specs being incorrect for the machines and mundane issues like that. In states where the scanners have been used for years, the error rate is extremely low.

OK that point about general incompetence having been made--it DOES ALSO appear that the most recent Diebold opti-scan machines sold to Florida had new "memory cards" that can be altered. The problem thus is in this new design, not in the scanning technology. (See link below)--an important distinction.

THEREFORE IT IS NOT CORRECT to say that opti-scan machines per se were the problem in the case of Florida in recent elections. It was a combination of new Diebold scanner that was purposely made very vulnerable to hacking, the poor training of the users primarily due to lack of time, and of course the unknowns of the central tabulators.

THE ANOMALIES reported (in comparing opti-scan counties with DRE counties in FL) DO make opti-scan look bad in comparison. But without having any access to the central tabulators, it's impossible to draw any real conclusion, except those based on statistical anomalies. I think they were trying to get their hands on the memory cards from the precincts in question in FL to look at that variable independently. Not sure where that went.

The two points of vulnerablity are:

1. The NEW memory cards in precinct-based opti-scan machines
2. The Central Tabulators

This article has more:

"Computer expert Harri Hursti gained control over Leon County memory cards, which handle the vote-reporting from the precincts. Dr. Herbert Thompson, a security expert, took control of the Leon County central tabulator by implanting a trojan horse-like script."

http://www.onlinejournal.com/evoting/060305BBV/060305bbv.html

----------------------------------

My problem is with using Florida as an example of where opti-scan failed, and then saying that means all opti-scan is no good. There are many states where it has worked very well for years (using less souped up older systems). It's a relatively simple technology, but Diebold and the election "helpers" in Florida have done a lot to discredit a system that, with the right procedures and checks generally works very well. If you research it, I think you'll find what I'm saying is true.

Of course at the same time, I feel we need random hand counts during the voting process and much more transparency in general. I'm just defending the basic opti-scan technology against blanket disparagement. It is MUCH better than DREs with paper receipts. It gives us real paper ballots, and used correctly provides a machine count that can be used in conjunction with hand counts. Obviously we must look at the NEW opti-scans very carefully--for one thing that's probably what Ohio is now getting--I wonder who the vendor might be for that contract(!?!) :(

[riqster]

92. Blackwell would like it to be Diebold

But ES&S won the right to bid for county business.

Don;t get too comfy, though-Diebold and ES&S are run by two brothers named Urosevich. How competetive are these two companies, really?

[marions ghost]

93. thanks for that info

I didn't know it was ES&S for sure. They have the right to bid--so who else is in the bidding?

I'm hoping that those on the ground in Ohio will be inspecting the new opti-scan systems very carefully, based on this experience from Florida about the "memory cards..." I'm sure we can trust ES&S about as far as we can trust Diebold.

[helderheid]

98. thanks for this - a very important clarification and much appreciated!

[mod mom]

122. MG. do we have a date when these new memory cards have become suspect?

(ie Optiscan macines manufactured or purchased after X date?)

[Cocoa]

104. punch cards were by FAR the biggest problem in FL

opti-scans had problems, most famously in Volusia county, but they have paper ballots which can be (and were in Volusia county) recounted by hand.

Anyway, punch cards gave Bush the win over Gore.

[PATRICK]

106. They overreach because they have to

so they don't like paper evidence even if they are protected from an automatic recount or able to mangle the paper eventually with some degree of impunity. Cyberspace invisibility is their Unholy Grail, The vestiges of traditional tampering sully their whitened sepulcher voting booths. Yes they could defraud many many ways, but eliminating ALL evidence and the need to soil their hands, even to having dumbed down organizations who "don't need to know" how they win.

Eventually this would create a class of vote controllers, the Bush Election Cartel and a lazy ignorant bunch of electees who lose real skill at cheating and might even believe they are totally above board.

In this case the Bush dynasty OWNS the GOP and the fight over the system controls can narrow. I think Gingrich was very much part of it and is increasingly frozen out, as Hagel will be at some point to the dismay of his ill sold soul. Such is the price for turning away your eye, even if you think you are on the same side, even if you are part of the current crime apparatus.

In the end their will be only one imperial center. The GOP wannabes are dumber than than DLC collaborators. Corrupt blinded, enslaved as ineffectual tools, the mass of vile corrupt pols probably is not even up imagining their own peril. Hard when you are well off admired, powerful, in a nice suit on TV.

[hunter]

117. Hi, Patrick, I haven't seen you around.

But I haven't been hanging out much in the DU election forums lately. Even for purpose of arguing about election technology, it has become difficult for me to believe we live in any sort of Democracy or Republic. (I remember when it was still okay to laugh at elections in the Soviet Union... now I have very grave concerns about our own.)

It's a sad thing to live in a nation where people believe COUNTING MARKS ON A PIECE OF PAPER (1,2,3,4,5,6,7....) requires high technology.

If I went back to the nineteenth century and handed Herman Hollerith a bag of photodiodes, he could probably come up with an optical scanner that wwould be more trustworthy than some of the crap high-tech optical vote counters we have today.

As a practical matter paper ballots read by optical scanners are the only economical solution to our vote counting problems. But we must be very careful that these optical scanners don't have any more processing power than that required to simply count the votes. An optical scanner with mechanical counters might be preferable to something requiring an "operating system" or any proprietary software. (The suggestion that we might put such software in "escrow" is utterly appalling to me.)

The function of software is to manipulate data. There is no good reason to manipulate votes. The electronic path between your vote on your ballot and the count reported to the registrar needs to be entirely transparent. The optical scanners that read our votes have one job, and that is to simply count votes. These vote counting machines need to be as stupid and as low-tech as we can possibly make them or they will not be trustworthy.

[kster]

121. Kick,for old times sake..nt

[mod mom]

123. kick for transparent elections.