The Trouble With Tabulators

[ChrisPhx]

(The same message I have sent numerous times to numerous people and places since before the election)

I have completed 95% of my work towards a Masters of Computer Science and Engineering. My familiarity with computers is what led me to learn about and post in various arenas about the potential problems with the Central Tabulator, among other things.

While Bev Harris of blackboxvoting.org has been addressing this, she has also been in contact with experts such as Dr. Avi Rubin of Johns Hopkins, Dr. David Dill of Stanford, and Dr. Rebecca Mercuri of Harvard. They are all professors of computer science. And experts on voting systems.

My biggest frustration has been in attempting to convey the potential problem with the Diebold GEMS Central Tabulator software. It isn't a terribly difficult concept to grasp, I don't believe. But for some reason many people have been and remain resistant to it or incapable of grasping it.

The tabulator may be used whether touch screen voting or optical scan voting is used at the voting machine. It doesn't matter.

Basically the data (votes tabulated, totals) is stored in a Microsoft Access Database, a primitive Database as compared to many that are available. Because it is stored in Access, the computer language "Basic" can be used to manipulate the data. As you might guess from the name, Basic is a very simple programming language.

One person reportedly wrote a 6 line program or script to alter the votes.

There are other ways into the data, I believe, that are not legitimate.

Wouldn't you expect that our voting software might be developed and tested as rigorously as say the software that supports the space program at NASA? That mission critical software is subject to the most intensive standards and testing.

Should it be any less so for the electronic systems that not only are involved with individual voting machines, and even more so for the Central Tabulator that deals in thousands or millions of votes at a time? The software that guarantees our most basic of rights in our democracy?

Dr. Rubin is an expert in Computer Network Security, and nobody like him was involved in the development of the software.

Because the software (Diebold GEMS Central Tabulator) runs on computers with modems, they can be hacked into remotely if the modem is available on election day. But even if the modem is not available, as it shouldn't have been, anyone with malicious intent who obtained physical access to the computer could have changed the vote totals AFTER tabulation.

Bev Harris taught a chimp in ten minutes to delete votes and to not leave behind any trace that the change had been done.

You know, my theory is that the term "conspiracy theorist" is a tool used by the Political right to halt discussion of topics they don't want to have examined too closely.

I therefore propose that we ourselves cease the use of these words and the concept. Let them call it whatever they wish to, but don't let them stop us from examining scenarios which may be beyond our comprehension in terms of the lack of scruples the perpetrators would have to have.

I expect more and more information to emerge from Bev Harris as she continues to analyze the information received from her massive request for information under the FOIA (freedom of information act).

I believe the election was stolen, which is more depressing than simply losing. But for that very reason we have to find out if it is true.

[Wilms]

1. Tabulation sure does seem to be overlooked around here

I think it is not well understood, which may be both a cause and a result of this circumstance.

Holt's bill contained seemingly strong language with regard to safeguarding hardware and software in "Voting Systems".

Not one to assume that it really covered the entire "Election System", I wrote to Holt's office and asked about that. I hope they specify tabulation.

ChrisPhx, can you explain, or can you link us to sites dealing with, the step by step procedures that ballots take toward the final tally?

[ChrisPhx]

2. Here is a link to one state's manual for using the tabulator

This isn't quite what you asked for, although it does show if examined in detail how the votes are gathered in precincts and sent on to the next level.

http://www.azsos.gov/votesend/electronic%20election%20reporting.doc

None of it matters too much if someone can hack in at any point to change the total votes, which they can.

[rigel99]

18. You have no idea how helpful this is..

I'm working on Diebold problems in GA, started doing a deep dive Open Records (FOIA) request and have lots of Diebold data..
check out countpaperballots.com if you wanna see the latest letter exchanges with Diebold (not) and the KSU and Cox office.... very intersting stuff...

but was having an citizen audit party today with my most awesome countpaperballot friends, and we kept having these questions like what does system counter and public counter really mean in a Diebold machine, what do they track..

and then I zoomed in on your post... VOILA the answer in black and white because of you and your wonderful post!! brava brava..

you see we are preparing for legal action.. we've found lots and lots and lots and I mean lots of fraud... the hard part is getting access to the CD to prove the really intentional stuff...

so we are waiting for the right timing.....

this document is very helpful from Arizona.. is this an official, publicly released document? it looks like they have it on their website free for the public to consume... i cannot tell you how much this really really makes my day.. really... you have no idea....

bellisimo...

[ChrisPhx]

26. It is straight from the SOS web

site for Arizona. It was there prior to the election too.

A Republican who could not, would not answer questions before the election as to how security was to be maintained for the tabulator.

I'm so glad it was useful to you!

[Amaryllis]

5. I called and talked to them about this. They told me that "voting

systems" according to HAVA's definition also includes tabulators. His bill prohibits wireless connections to voting systems:


`(9) PROHIBITION OF USE OF WIRELESS COMMUNICATIONS DEVICES IN VOTING SYSTEMS- No voting system shall contain, use, or be accessible by any
wireless, power-line, or concealed communication device at all.

However, this part ONLY applies to the actual voting machine, not to the tabulator:

`(12) PROHIBITING CONNECTION OF SYSTEM OR TRANSMISSION OF SYSTEM INFORMATION OVER THE INTERNET- No component of any voting device upon which votes are cast shall be connected to the Internet.'.

[ChrisPhx]

6. Physical access is also an easy way in

Bev did a demo and taught a chimp to alter the totals, leaving no trace, in ten minutes.

Howard Dean was present for the demo.

[Wilms]

7. Thanks. I'll check HAVA.

And that helps explain why 9 and 12 are separate.

I also asked about improving the language of those two sections so that it would read more like...


`(9) PROHIBITION OF USE OF ALL COMMUNICATIONS DEVICES IN VOTING SYSTEMS- No voting system shall contain, use, or be accessible by any wired, wireless, power-line, concealed, or other communication device at all.

Sec 2 (c) (12) could be strengthened to read:

`(12) PROHIBITING CONNECTION OF SYSTEM OR TRANSMISSION OF SYSTEM INFORMATION OVER THE INTERNET OR OTHER NETWORK- No component of any voting device upon which votes are cast shall be connected to the Internet, or other network.'.

If networks are allowed/used in elections administration they could be exempted, but should also come under regulation and certification requirements.

[Amaryllis]

8. You can call and talk to MIchelle.She is very knowledgable about the bill

and why they did it the way they did. If you find out, let us know.

[Wilms]

10. HAVA 's definition of "Voting Sysytem" seems to cover the entire system

That's good news, though HAVA really doesn't speak much about counting or tabulation.

But this means Holt's bill's requirements for security would cover tabulations

From HAVA:

(b) Voting System Defined.--In this section, the term ``voting system'' means—

(1) the total combination of mechanical, electromechanical, or electronic equipment (including the software, firmware, and documentation required to program, control, and support the equipment) that is used--

(A) to define ballots;
(B) to cast and count votes;
(C) to report or display election results; and
(D) to maintain and produce any audit trail information; and

(2) the practices and associated documentation used—

(A) to identify system components and versions of such components;
(B) to test the system during its development and maintenance;
(C) to maintain records of system errors and defects;
(D) to determine specific system changes to be made to a system after the initial qualification of the system; and
(E) to make available any materials to the voter (such as notices, instructions, forms, or paper ballots).

Thanks for helping me through that, Amaryllis. :)

[Amaryllis]

11. But look at the part of the bill that deals with no internet connection.

It only applies to voting machines, not tabulators and scanners. That makes me nervous. All they need is a modem phone number and password to hack it.

[Wilms]

16. Makes me nervous, too.

That's why I brought up the subject in my email to Holt saying:

"If networks are allowed/used in elections administration they could be exempted, but should also come under regulation and certification requirements."

I really wouldn't care if they used all the electronic junk to provide "Preliminary Results", and wait a day for a hand-based count AND tabulation.

[rigel99]

19. Wow! Thanks Amaryllis...

I'm catching all kinds of goodies on this thread..

your (E) to make available any materials to the voter (such as notices, instructions, forms or paper ballots), this is excellent... very helpful. What do you think they mean by instructions?

also under (D) maintain and produce any audit trail information, if they don't accumulate the DRE tape printouts at the precinct, would that be a violation? I found multiple counties that don't accumulate at the precinct.... it's technically required for a full audit... but this is just a technicality.....

[Amaryllis]

23. Good questions; don't know, but the more I look at this the more I don't

want any e-voting systems at all.They are just too damn hard to secure and we shouldn't have to consult IT experts constantly to know if we have secure elections, dammit!

[ChrisPhx]

27. Not when manufacturers are big partisan

supporters of W.

"want any e-voting systems at all.They are just too damn hard to secure and we shouldn't have to consult IT experts constantly to know if we have secure elections, dammit!"

[Wilms]

25. (e) May refer to "Instructions" printed to train voters & poll workers.

With (d), note the word "any". Sort of like "if any", I think.

[ChrisPhx]

3. If you're concerned about tabulation check this out...

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x330246

[Philosophy]

4. They may have poor security but I don't think that was the problem

if the problem were hackers I would think we would have heard something about it - not all hackers are experts and some would have left tracks, some would have on purpose just for the notoriety. I think its far more likely that backdoors were built in to the software to enable Repub political operatives to tweak the results. No amount of security on the machines will fix that.

[Amaryllis]

12. That's what really freaks me out about all the legislation. I keep thinkin

we are just putting bandaids on a festering wound, and the voting systems companies and repubs in the know are just laughing at us. Did you see this thread?
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=330275&mesg_id=330275

[Amaryllis]

13. When I say hackers, I mean anyone who goes in and tweaks the

results. Maybe I am not using the term properly. At any rate, it's too damn easy to get in and tweak results, whoever is doing it. And it takes so few of them to change enough records to impact the election.

[Wilms]

17. How well is the problem addressed with these Holtian provisions?

From HR 550:

`SEC. 247. (c) '(8) PROHIBITION OF USE OF UNDISCLOSED SOFTWARE IN VOTING SYSTEMS- No voting system shall at any time contain or use any undisclosed software. Any voting system containing or using software shall disclose the source code, object code, and executable representation of that software to the Commission, and the Commission shall make that source code, object code, and executable representation available for inspection upon request to any person.


`(10) CERTIFICATION OF SOFTWARE AND HARDWARE- All software and hardware used in any electronic voting system shall be certified by laboratories accredited by the Commission as meeting the requirements of paragraphs (8) and (9).

[sacxtra]

9. IT'S THE ELECTRONICS, DIGITIZED DATA, AND INSECURE NETWORKS

All of which have now denied you to have your vote count, or be audited.

This happened at the EXACT moment they were converted to digitized data your ballot literally disappeared at light speed.

Now you need to fill out an election/fraud complaint form.

Your vote didn't count.
Nobody can tell you your vote counted, because nobody can see it.
Therefore your right to vote was violated.


Also, You don't need a masters degree to manipulate digitized data.

Bring me ANY PART of the electronic, digitized data, or insecure networks voting system. Give me permission. I will break it. And I only have a GED.

[berniew1]

14. Main factor in stolen election was computer tabulator default/glitches/swi

switching.

There was widespread touchscreen machine fraud/switching in about 15 states documented.
http://www.flcv.com/fraudpat.html (Florida, Ohio, N. Mexico)
http://www.flcv.com/ussumall.html (other states)

and default programming of tabulators to count straingt Dem ticket votes for other than Kerry or no one in 10 states (above)

and documented that some of the manufacturer reps had remote patches to interact with the election result compilers on election day
(affidavits in Ohio)

and widepread rigging of ballot/tabulators/compilers for optical scan and punch cards(documented by Greens in Ohio, and etc.)
http://www.flcv.com/greenrc.html


There is not enough understanding of the magnitude of the widespread fraud/vote switching that occurred in the 2004 election due to the programming and proceedures used to count the votes.

I'm not sure its possible to have reliable/secure results using touchscreens no matter how much effort is made, and all systems can be easily manipulated to affect the vote count if there isn't an adequate system of checks/balance.

[ChrisPhx]

15. Lack of widespread knowledge is the problem

"There is not enough understanding of the magnitude of the widespread fraud/vote switching that occurred in the 2004 election due to the programming and proceedures used to count the votes."

Very true, probably the biggest problem, challenge we face.

[farmbo]

20. Here's an oldie but a goodie from the DU archives about Triad tabulators:

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=104&topic_id=2768979


Triad tabulated 42 Ohio punchcard counties. They alone awarded Bush the presidency.

[passy]

21. WAN GSM

"Because the software (Diebold GEMS Central Tabulator) runs on computers with modems, they can be hacked into remotely if the modem is available on election day. But even if the modem is not available, as it shouldn't have been, anyone with malicious intent who obtained physical access to the computer could have changed the vote totals AFTER tabulation."
What do you think about using something like this for remote access : http://www.digi.com/products/wireless/digiconnectwangsm.jsp
Could the same technology be hardwired?

[ChrisPhx]

28. This would be a good question for someone

like Avi Rubin, an expert at Johns Hopkins in Computer Networks security.

[ChrisPhx]

32. Avi Rubin's web page url:

http://avirubin.com/

[bailey77]

22. The Diebold tabulator has now been hacked in a live system

The hack was done in conjunction with research by Black Box Voting, and took place in a live election setup using a real county system. This is a recent development, and it was filmed by two different entities.

Black Box Voting is working on replicating the hack with other versions, and extending the hack into various remote access configurations with other GEMS versions.

They will be in contact with the GAO and John Conyers this week about this new penetration of a live county election system.

Erin

[BushSpeak]

24. The chief programmer of Diebold GEMS software is a convicted felon

In the early 90's, Jeff King was convicted of embezzlement in the Seattle law firm he worked for. He did this by installing a back door into the company's accounting software.

He later ran a ballot printing company that worked for Diebold. He was then hired by Diebold, given a seat on the board of directors and named senior programmer of their Diebold GEMS software that is used to operate their central tabulators.

I got this off off an Andy Stephenson radio interview.

According to Bev Harris's research, their are three copies of the database. The election results are written to the first one, then copied to the second with whatever modifications deemed "necessary". The election officials only can see the second database and these are the results that are printed out.

Apparently, Bev also found that they are equipped with remote control emitter/receivers. During one California primary, the results were shown in real time at a Diebold site. The only way this could have been done would have been through direct wireless access to the machines.

A lot of this info disappeared, when Bev's site was hacked just after the election.

Looks like we should also insist that the votes be counted at the precinct level before transmission.

[bailey77]

29. A few minor corrections

Jeffrey Dean, not Jeff King. He was based in King County, Washington.

He later ran a ballot printing company that worked for Diebold. He was then hired by Diebold, given a seat on the board of directors and named senior programmer of their Diebold GEMS software that is used to operate their central tabulators.

I got this off off an Andy Stephenson radio interview.

Jeffrey Dean was not hired by Diebold. He did become the owner of Spectrum Print & Mail, a printing company shortly after he got out of prison. This company got the contract to print King County Washington's ballots.

Global Election Systems (a Canadian company founded by stock market manipulators and felons) got the contract to supply voting machines for King County in 1998.

Global Election Systems acquired Jeff Dean's company, Spectrum Print & Mail, in 2000 shortly before the presidential election. Jeff Dean was given the position of chief programmer and was also given a seat on the board of directors of Global Election Systems.

In January 2002, Diebold acquired Global Election Systems. The sale was consummated on Jan. 23, and Jeff Dean was no longer a director and was taken off as an employee and put on as a consultant.

He continued until April 2002, and rumor has it he still is on the payroll somehow, but that has never been confirmed.

According to Bev Harris's research, their are three copies of the database. The election results are written to the first one, then copied to the second with whatever modifications deemed "necessary". The election officials only can see the second database and these are the results that are printed out.

Not exactly. The election results are written to the first one and also to the second one. The election results with precinct-by-precinct results come from the first one, and the summary results come from the second one. By doing some very simple sleight of hand, it is possible to manipulate the second one and leave the first one intact. This allows the overall results to be changed, yet even a hand count of randomly selected precincts will always match the precinct-by-precinct report.

Apparently, Bev also found that they are equipped with remote control emitter/receivers. During one California primary, the results were shown in real time at a Diebold site. The only way this could have been done would have been through direct wireless access to the machines.

No. Bev Harris and Kathleen Wynne interviewed some of the people who order parts and assemble the Diebold machines, last July, in McKinney Texas and elsewhere. All of them said that wireless was ordered for the machines but they differed on whether it was to be installed in the past or the future.

You are probably thinking of the San Luis Obispo problem, which you can find documented in Chapter 13 of the Black Box Voting book, online for free at http://www.blackboxvoting.org.

That problem was initially thought to be a case of voting machines in the field calling home via telephone modems (not wireless). Then it was determined that the votes were absentee and early votes, though the accounting and reporting had some anomalies.

The vote file did show up on a Diebold FTP site, and it appears to have been saved at 3:30 p.m. on election day, a violation of California law. No one at the elections office or Diebold will admit to copying the vote file out of the San Luis Obispo County GEMS computer, nor to placing it on the Internet.

The file is tagged with a password ("sophia") and a Diebold tech named Sophia Lee was at San Luis Obispo that day, but she denies taking the file.

A Diebold memo also indicates that less than 24 hours before the election, the San Luis Obispo GEMS system was tabulating votes incorrectly.

A lot of this info disappeared, when Bev's site was hacked just after the election.

This is correct, but the hacking took place shortly before and just after the election, and the database containing the story files on this was corrupted and is now unusable.

However, you can still find the details on this, thoroughly documented, in the Black Box Voting book. Primary researchers on the San Luis Obispo story were Harris and one of the directors of Black Box Voting, Jim March.

Looks like we should also insist that the votes be counted at the precinct level before transmission.

Yes. But most places are moving away from this. When you count at the precinct, the vulnerability is the attack point during transfer of the data, or ballots, to a central point. However, if you post the precinct count publicly, you have at least some checks and balances.

When you count at a central point, as is done with all mail-in systems, punch card systems, and absentee votes, you get into the dog and pony show where it is relatively easy to do a snow job on observers.

The poll tapes, which should be signed by poll workers at the polling place and should contain both the zero report (beginning of election day) and the results (end of election day) are a good check for this, but if no one signs them, or they go missing, or chain of custody is weak, it is not difficult to game the system.

[rigel99]

30. WHAT DOES STOLEN DIEBOLD EQUIP WITHOUT SERIAL NUMBERS

Mean?

What would that mean in terms of potentially invalidating an election. For every state that uses Diebold, they need to reference this police report and start a lawsuit to begin invalidating the erroneous Diebold data... there can be NO WAY to trust the data when there is equipment (including full network technology and software) stolen and at large, with no serial numbers and therefore no way for Diebold to secure this technology from being able to plug into any state's Diebold network and upload fake data.

Please get this information into every state that is investigating Diebold, and encourage aggressive litigation... Here is the link to the Stolen Diebold Police report.. PM me for a full copy of the report..
http://www.countpaperballots.com/whatstolenequipment.htm

[ChrisPhx]

31. Uhhh...better hurry

They're prolly about to outlaw any litigation that threatens national security or IOW that they don't like.

:toast: