Snohomish county, Washington - evidence of systematic machine fraud

[tex-wyo-dem]

This was posted by Land Shark on http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x283060 and I thought it was important enough to make it it's own thread (with permission from Land Shark).

Land Shark wrote:

In Snohomish county, Washington I just reviewed the audit logs for all 933 Sequoia machines. The logs feature state that at least 98 different "screen calibration" routines were run on Election Day among 933 machines. Accounting for machines that had more than one calibration, 81 different machines had to be recalibrated because the touch screens were "anomalously" selecting Republicans.

That's about 9% of the machines actually getting intervention. An unknown additional number were just allowed to sit or the problem wasn't noticed on.

These audit log numbers overlap reports from election workers, so 81 machines out of 933 appears to be a HARD and MINIMUM number of "screen calibration problems".

Details on how the machines that were switched off due to malfunctions and voted heavily for republican Dino Rossi is in the detailed study at
http://www.votersunite.org/info/SnohomishElectionFraudInvestigation.pdf

Quote in report written by Land Shark:

"However, to use the misleading terminology anyway, if “the screen” was off by half an inch, it remains unexplained why the machines that were taking Democratic votes to be Republican votes (the line just below Democrats), were not also taking Republican votes and making them Libertarian votes, (the line just below Republicans) with similarly
large numbers of complaints from Republicans about “screen calibration issues”. Indeed, it might also be asked why no Republican votes were recorded as Democratic, though a true screen calibration defect would not produce errors in both directions at the same time, one would expect a calibration issue to occasionally go in the up direction and not just the down direction. Interestingly, the reports of these screen calibration problems" were virtually all on major ticket races, and not on the various issues and candidates that were lower on the ticket, even though some of the lower races are in the same position on
the screen as the earlier major ticket races, and thus should be subject to the same “miscalibration” problems, even if the miscalibration was somehow isolated to one portion of the screen.

[uppityperson]

1. Does fraud for/by repubs mean we need another election?

On the other hand, if we could include federal elections too? (Dream on) I really wish this would get more MSM coverage and changes made.

[Land Shark]

3. Ironically, WA Repugs want a re-vote

Best response to that is that seeking electoral justice is a good thing, but we'd need to fix the problems before a fair re-vote could occur.

[dmac]

6. So, lots of us

want a re-vote! Why do WA Republicans think they have grounds for one? (and do they stand a chance of getting their way?)

[Land Shark]

8. The Repugs hope to find 129 or more questionable votes

in order to revote the gubernatorial election ONLY, and in a runoff only, without the libertarian candidate or write-in.

So, they hope to find nonregistered voters voting, dead people voting, and felons voting, in amounts over 129 votes. Problem is, the only dead person who voted that was reported, voted for Rossi via her surviving husband and allegedly at her request (there are allegedly others voting while dead, but votes weren't specified).

The types of defects they are searching for tends to invalidate EVERY race, though the other races weren't as close. That's why they want to find relatively small numbers of "discrepancies" and not large numbers, so the Republicans may not join us in exposing more systemic discrepancies, irregularities and cheating.

I wouldn't assume without further research that "voting while dead" is a problem, so long as one was alive and registered at the time of the absentee vote was voted and signed.

[Amaryllis]

25. Paper ballots, hand recount. Gregoire wins by a much larger margin.

[tex-wyo-dem]

2. kick - check it out...important stuff.

[Land Shark]

4. Clarification

I stated that these machines were all "selecting Republicans". In one case i've documented, the voter attempted to vote Democratic and the checkmark dropped down TWO levels to Libertarian, rather than dropping down ONE to Republican. In a few cases, the screens froze up when people tried to vote Democratic and in a few cases there is no record of what people were trying to vote for when the "wrong candidate" was selected. But in all cases where the voter made a report on Election Day specifying their candidate choice, it is all AGAINST Democrats, though not 100% in favor of Republicans, just a high percentage in favor of Republicans.

Although my original phrase "selecting Republicans" is still largely correct, I'd like to be precise with the above information.

[d.l.Green]

22. Playing devil's advocate,

but being biased, I really would like to believe that the people reporting problems were being truthful. There are an awful lot of us, at least here, who believed some kind of tricks would be pulled and pretty much expected them. There was a link going around the web with a sample evoting screen that made it impossible to successfully enter a vote for Kerry- and even if you succeeded the screen still verified a vote for *. I'm not aware of a similar link going around in R circles with the opposite result. The power of suggestion amplified by expectations may possibly have resulted in at least some of these complaints. The trick would be to impound a few machines in the specific districts where this was recorded to have occurred and do some tests- or read the events on the log- if the actions of choosing before confirming shows up on a log. (I don't know a lot about computers but couldn't a machine be made to think it was November 2nd again, so that any patches that may be date sensitive would reactivate?):evilfrown:

[demo dutch]

5. I hope all this evidence is making it's way to Conyers as he is

continuing his investigation

[wiggs]

7. Questions

What are the audit logs you mention?

When was the "audit" conducted?

If the machines have not been subsequently tampered with, wouldn't inadvertant "switching" of votes be repeatable if, today, someone turned the machines on to conduct a vote drill?

Have the machines been secured?

Is law enforcement or Justice Department or FBI looking into this?

Thanks. Seems important, not only for the state race but in terms of national perception of voting integrity. It would be very very important to show that machines in Washington had been rigged, because switching of votes is suspected in several states.

[Land Shark]

9. Audit logs are internal computer records

Switching of votes may or may not be repeatable, depending on whether any code that may have activated the switch is set to run only in full election mode or on certain dates, etc.

Requests have been made to preserve evidence, but no special seal or items beyond ordinary "security" has been instituted, to the best of my knowledge. It's possible, but unlikely, that the county has taken some extra steps in response to my report and requests. I certainly have not been informed of any steps.

No law enforcement investigation. Press releases were sent out to the AP and everywhere else by votersunite.org but only scoop in new zealand passed the release on in any way. Local press has the report, and initially (before the report) claimed an interest, but I am still waiting for [e-mail] to follow up as he said he would.

[dbDESIGN]

10. The software could behave differently

depending on the date or which user is logged in. I write programs that customize themselves just based on who is logged in to use them.

Also if remote access is available via the Internet or a modem (and a cellular modem could be hidden inside the box without a physical and obvious land line) then the software or parts of it (.dlls for example) could be invisibly replaced and therby completely alter the behavior of the machine.

It is a big job to secure computers and software even if all involved (programmers, network techs, users, managers) have made security the top priority. We don't have that case here.

We need to see the source code. The only way that will happen is if someone on the 'inside' brings it out.

[tex-wyo-dem]

11. Agree, we need to see the source code...

But, it looks like the only way that will happen is either a leak from the companies themselves or by court order.

Question for you dbDESIGN...I've looked at pictures of touchscreen machines from the different companies (i.e. Diebold, Sequoia, etc.) and they all appear to be similar in hardware, at least externally (similar mechanics, similar screen type/size, similar form factor). Do you think it's possible that one of the companies developed the software, and possibly the hardware, for these machines and is just licencing it out to all the other companies? In other words, different companies making machines with different plastics and logos, but the exact same hardware and software?...Interesting thought.

[dbDESIGN]

12. Sorry, I don't know.

I would guess there are probably fewer hardware manufacturers than systems distributors. That's the case for desktop PCs but I really don't know. Maybe the vote-cheating is built into the BIOS or an on board chip and even the companies distributing the systems don't know. Paranoid or what!?!?

[Land Shark]

13. The main thing is that the party holding back information is the

one that needs to answer the questions. And the party being denied the information can not be held to the full standards of scientific proof. In law, it's called the "spoliatory inference". If a party has control of certain evidence and does not make it available, the jury is allowed to infer that the evidence, if produced, would favor the party from which it is being withheld.

[tex-wyo-dem]

15. Interesting...

so the withheld code could still be used in a court of law as "inferred" evidence. Does is carry the same weight as if the evidence were available?

[Land Shark]

16. Rules on spoliation vary from state to state

A case can not be based entirely on the spoliatory inference, but it's rare that there would be an absence of any other evidence. Most likely some version of the code that may or may not have all of it would be produced under court order, and the debate would be about whether something was NOT produced that the spoliatory inference can attach to. A plaintiff would then need to prove by a preponderance of the evidence that something existed that was destroyed or withheld by the other party, under circumstances when such evidence should not have been destroyed (such as rules requiring the retention of evidence or votes).

[d.l.Green]

23. But wouldn't the proprietary nature

of the voting machine programs trump discovery? Even in the best case scenario, unless there was compelling evidence- that exposing the code would be necessary to prove- wouldn't a judge just refuse such a request?

[Land Shark]

26. No, but the judge might enter a protective order for nondisclosure n/t

[Land Shark]

28. or otherwise limit what has to be produced n/t

[Carolab]

38. I don't see how the standards for "auditability" set by HAVA are met.

If one cannot look at the software source code, how can one audit anything? The paper records will simply be a record of whatever went on "inside the machines" or via modem "retallying". There is no audit capability if one cannot see what was producing the paper record in the first place or if there is no individual paper record of each vote that can be audited.

[tex-wyo-dem]

14. When it's not such a far fetched possibility...

I don't think that qualifies as paranoia :)

Imagine the BIOS being programmed by a single individual in a single company with vote-cheating built in, chips get burned and licensed/sold to other manufacturers of touchscreen voting machines. No one except the programmer and anyone else orchestrating this knows, including the companies buying these chips.

Certainly cuts down on the amount of people you would have to keep quiet.

[understandinglife]

17. Yes; this is one strategy that should be considered as probable...

...as would certain other 'firmware' and network-based strategies. You would expect that those involved would have thought carefully about minimizing the number of 'informed' participants.

In addition, given the lack of technical sophistication evidenced in comments from election board folk in places like OH, MD, NM, WA, ....., it seems that what I'd characterize as "naive complicity" is rampant and those involved in the e-vote fraud enterprise took full advantage of that fact.

Peace.

"When Did Bush Know?"

[tex-wyo-dem]

18. Yes...one problem is that most people are not technically savvy...

and just trust that the systems do what they are supposed to do. There are just way too many ways for computerized systems such as this to be compromised.

[understandinglife]

19. Like I've said before: "a child can pull the trigger on a AK-47"...

...and our elected representatives can pull the trigger on approving and installing touchscreen vapor ballot devices, optical digital scanners, central tabulating computers and do so with zero clue as to the consequences -- most notable of which is that a tiny number of individuals can then select our government and blow-away our franchise of democracy.

We're facing live-fire, everyone, and if we don't destroy the vapor-ballot, in all its manifestations, real soon now our Constitution and franchise are doa.

Peace.

"When Did Bush Know?"

[Land Shark]

20. More data on screen calibration to come n/t

[d.l.Green]

24. Only solution, back to paper ballots with receipts n/t

[berniew1]

21. Many machines in Snohomish were programmed Default to Bush

just as they were in Florida, Ohio, Texas, Calif.,Pennsylvania, New Mexico, etc.
http://www.flcv.com/snohomis.html
http://www.flcv.com/fraudpat.html

[Land Shark]

27. Thanks berniew1 n/t

[Land Shark]

29. New Data on When Screen Problems are "Fixed"

"Screen recalibration" is the routine poll workers are supposed to run if they spot a "screen" behaving strangely, though the screen recalibration is really a program that is run so that the computer can more properly interpret the data coming from the screen. In any event the following data shows what time of day poll workers ran these screen recalibrations, as evidenced by the audit logs:

Well my chart wouldn't paste, so here's the data:
6:00 4
7:00 3
8:00 10
9:00 10
10:00 14
11:00 15
12:00 8
13:00 6
14:00 2
15:00 11
16:00 0
17:00 4
18:00 1
19:00 2

There's a late morning peak and a 3 p.m. peak, with fewer recalibrations when the poll workers are busy with rushes starting at 7 a.m. (opening) and 4 p.m. and noon (dinner and lunch rushes) But basically the need for recalibrations is occurring throughout the day, so one of two things must be true. Either (1) moving the machines on Election morning is not the cause of the recalibration difficulties or (2) lots of voters are voting on machines in need of recalibration, but not getting it until late in the day.

Of course, it is also possible that apparent screen recalibration is the product of vote switching, and not bouncing in the truck on the way to the polls.

[LeeB]

30. "vote switching"

Do you mean that cute little trick where a vote for Kerry (or Gregoire?) turns into a vote for * or Rossi??? Sorry if you already explained that . . . and thank you for the work you have done. I've been nagging my email list with links to your report.

[Land Shark]

32. Thanks, yes, "vote switching" is when a diff candidate gets checked

compared to the one that is pressed by the voter

[dbDESIGN]

31. The visibility of the vote switching puzzles me

in that a well executed electronic vote fraud scheme shouldn't be waving flags and saying,"Hey I switched your vote!" The electronic switching of votes from one candidate to another would be done without a trace. The machine would merrily indicate that, indeed, you voted for who you thought you did while quietly recording the vote as it was programmed to do. Isn't that what black-box, touchscreen machines were designed for? So why the noticeable vote-switching problems with touch-screens?

Greed. The hardware/software is crap, cobbled together to exploit an outflow of money to 'modernize' voting in the absence of any decent performance criteria and certification process.

These touchscreen problems are a golden opportunity to break open the whole voting machine scam from the shoddy products to the outright election theft. Even those who will not consider fraud could understand not getting value for money.

Now if there were just someone with some clout to investigate......

Copied from my post at http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x286489

[Land Shark]

33. Visibility of vote switching: why visible if fraud the intent?

Given that it is difficult for a vote switcher to know how absentee votes are going from the same precincts (since many of them are not even voted and dropped off until election day here) there will be at the very least statistical discrepancies between the switched votes and other current or past voting behaviors. These discrepancies may be small and fully consistent with political changes in the district, or they could be large and harder to explain.

In the event that the swing is harder to explain or can't be explained with the available political explanations, then it would be useful to have a seemingly neutral voting defect explanation available. It gives the intentional fraud plausible deniability.

Or, it could just be a machine defect that strays only one way based on a structural consideration like ballot order.

Although it matters greatly for political consequences and potential criminal penalties, it matters little WHY our election results are off in one direction,simply that they are inaccurate even for reason of malfunction is a huge problem in itself because it means the election is not a valid measure.

[Wrinkle_In_Time]

35. Plausible deniability?

I was also puzzled by this... why make it visible? Then I realised that a touch-screen "calibration error" may not leave a footprint, unlike source or object code. Sure, there is a log entry for running a re-calibration routine (and I would love to see that code too) but you have no record of how far off the "calibration" was or even if it was a simple offset.

Maybe whoever chose to use this method of fraud (in addition to all the other methods used) was relying on people not noticing it all the time and/or banking on blaming poor calibration by under-trained election staff.

[dbDESIGN]

36. Sounds like a hide-in-plain-site plan

[Land Shark]

37. 'Spose it could be that too n/t

[LeeB]

34. Definitely a golden opportunity . . .

. . . and you're quite correct that "a well executed electronic vote fraud scheme shouldn't be waving flags and saying,'Hey I switched your vote!'"

Pardon my mirth, but I think the short answer to this is they had the same folks planning the rigging of our elections as did the planning for invading Iraq. :evilgrin: . . . just a thought.

[Ryder911]

39. I once owned a touch-screen kiosk company and...

although I don't know much about code, I can tell you that it would be very easy to intentionally mis-calibrate a touch-screen monitor to favor one selection over another. Since there is no way to visually see where the touch activated zones start and end, all you would have to do is put two graphic buttons very close together whereby one button's touch zone (the one you favor) crosses over (or near) the graphic button of the opposing selection.

The beauty of a such cheat is that if it is done correctly, only a very small percentage of those using the machines would have to make the error (e.g. 2%) for you to make a huge difference in the final results of the election. Keep in mind that this 2% error factor gets doubled to 4% in the results because not only are you stealing two votes for your candidate, you're also stealing two votes away from your opponent.

And, it doesn't take much for such a cheat to work. If the buttons were placed closely enough together, it might only take a millimeter or two to get the 2% error factor needed to swing an election.

Lastly, unlike writing a cheat in the source code, it's virtually impossible to prove someone has cheated by intentionally mis-calibrating a touch-screen. This is because touch-screens are notorious for developing calibration problems and (unless things have changed in the last few years) there is nothing built into these monitors to guarantee that the visual graphic buttons on the screen line up pixel to pixel with the touch activated points on the screen. So, even if such an error was detected, the cheater could reasonably blame it on an equipment malfunction or simple human error.

[Land Shark]

40. excellent post, thanks for the input Ryder n/t

[Ryder911]

41. thanks. /eom