Forum Navigation
Topics
Log In
Log Out
:
Forum Search
New Today
New This Week
Advanced Search
Tree View
Forum Account
Edit Profile
Register
Forgot Password
Forum Tools
Help/Instructions
Policies
CLICK STATE TO SEE:
"WATCH LIST"
Marked with: 
"OPEN & HONEST"
Marked with: 
...
|
| (US) 4/06 - New Voting Machine - Both... |
|
| Author |
Message |
    
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 1
Registered: 03-2006
Best of Black Box? 
Votes: 7 (A keeper?) | | Posted on Thursday, March 30, 2006 - 8:45 pm: | 
|
I am in the process of programming a simple and secure election system that can replace the currently used electronic voting machines. This system is extremely affordable as it will use commonly available PC equipment.
During the voting session, the voters' ballots are randomly placed and printed on one continuous paper form. Each voter is able to verify his or her selections on the print-out, and gets to vote again if selections were incorrect. With all ballots on one continuous sheet of paper, there can be no stuffing or skimming
of ballots.
If you would like to find out more, please visit my blog: http://newvotingmachine.blogspot.com/
(Message edited by bobn on March 30, 2006) |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 106
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Saturday, April 1, 2006 - 7:38 pm: |
|
What happens if an "alternate" roll is printed up and swapped for the legitimate roll when no one is looking? In that case we'd be right back to where we are now.
Now, if you were to also anonymously "publish" all votes on the Internet, so that all voters could "verify" that the vote they cast was the vote that was counted, then I'd say your system would be pretty much goof proof/mac |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 3
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 4, 2006 - 4:12 pm: |
|
If an alternate counterfeit roll were printed and succesfully inserted, it could be recognized immediately if an additional on site computer received all the subtotals from each booth and also printed them out. Any difference in subtotals would be immediately noticed. Poll workers would sign in on each continuous form ballot both at the beginning of the day and after verifying the numbers at the end of the election day.
As to your question on publishing the results on the internet, thus giving eachvoter a post-election accounting capability. This could be done by placing a unique number on each ballot for the voter to record if desired. Here a double randomization would have to occur so the anonmity is maintained. The position of the ballot on the continuous form and also the unique number would have to be issued randomly.
I believe that this would satisfy the conditions for anonymity, However it would be up to the election commission to make that decision. Some people would argue that a controlling person would than be able to force another to vote the way they specified. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 107
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 6, 2006 - 12:36 pm: |
|
>>>If an alternate counterfeit roll were printed and succesfully inserted, it could be recognized immediately if an additional on site computer received all the subtotals from each booth and also printed them out. Any difference in subtotals would be immediately noticed.<<<
What if all of the poll workers were in cahoots such that they modified the orignal roll and all "subtotals"
>>>Poll workers would sign in on each continuous form ballot both at the beginning of the day and after verifying the numbers at the end of the election day.<<<
See above; with heavy emphasis on "if all of the poll workers were in cahoots"
>>>As to your question on publishing the results on the internet....This could be done by placing a unique number on each ballot for the voter to record if desired.<<<
I know. You see this is something I proposed years ago.
>>>However it would be up to the election commission to make that decision<<<
Yes, of course, and it is somewhat analogous to asking the fox "Mr. Fox, would you please help us install systems and measures that would prevent you from stealing all our chickens, Mr. Fox?".
>>>Some people would argue that a controlling person would than be able to force another to vote the way they specified.<<<
"forcing another" is a crime best dealt with by the police and courageous citizens who come forth to report that crime.
May I suggest you read some of the material avalable on this site concerning alternative vote systems. I guess in the last year there must have been at least a dozen different "systems" proposed similar to yours, all with one inherent flaw. That flaw is that those "systems" ultimately rely on humans of potentially questionable intent, and having access to critcal though subtle parts of their "system", "not tampering" with that system. Indeed, any system would work, including the infamous Diebold monster, if humans who had access to it would simply "not tamper" with it.
Now, at this point you may feel a sense of running up against a wall of the intellectual box we all live in, but there is hint in that previous paragraph of a way to break thru that wall. It is the part about "humans of questionable intent". If a human could be found whose intent was beyond question, then that human could be enlisted to "verify" critcal aspects of any "new" system you might devise, even the system you presently propose.
But, what human would possess unquestionable intent? A Mother Theresa? The Pope? Pat Robinson? A child? I think there would be someone at some point who would question the integrity of even these people.
There is however one person whose intent is beyond question, at least so far as the vote is concerned. That person is, the voter. If the voters desire is to skew the result of their own vote, they would simply have to vote that way initially. Thus the voter has no motivation to "tamper with" their vote later on.
This and other fundamental principles/ideas have been proposed by myself and may be found at my website www.howtorescueamerica.com . The ideas detailed there must have potential, as there are people now running around the internet proposing ideas remarkably similar to mine, all without crediting me for being even the "inspiration" for their "brilliance". It seems reasonable to question the intent, if not the integrity of such people/mac
(Message edited by littlemac on April 06, 2006)
(Message edited by littlemac on April 06, 2006) |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 4
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Saturday, April 8, 2006 - 7:09 pm: |
|
>>>What if all of the poll workers were in cahoots such that they modified the original roll and all "subtotals"<<<
The argument you present on your website www.howtorescueamerica com/htra.html in favor of printing the ballots and the unique ballot numbers on the internet is soliid. That is, the government has already implemented the absentee ballot which practically invites vote selling . So now let's continue to generate a workable system using the internet.
>>>at least a dozen different "systems" proposed similar to yours, all with one inherent flaw. That flaw is that those "system" ultimately rely on humans of potentially questionable intent, and having access to critical though subtile parts of their "system", not tampering with that system.<<<
By taking your suggestion of incorporting the internet, let us see if we cannot eleminate these potentially questionable humans. The tightest loop (i.e. with least human discretion) , to enable validating the continuous-form ballot, would be for an onsite summarizing computer to receive and immediately retransmit to the internet the subtotals together with ballots and unique identifying numbers from all polling booths. You refer to a group of people known for their good character at your website who will recognize the value of being honest especially where the law is being strictly enforced with appropriate penalties, and I would add representatives of each of the main political parties. By having such a group at the each voting location, they could validate the continuous-form printouts with the internet totals. This loop could be fed back within and hour at voting places which maintain a full queue of voters. This timing would allow for sufficient randomization of ballot position on the continuous-form ballot.. The resulting procedure should not only give the voter ample reason to trust the system, but greatly improve voter turn-out as the voter is fully participating in the process. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 108
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, April 9, 2006 - 10:04 pm: |
|
This was a miss post - I was unable to delete it, so I simply erased it. OOps!
(Message edited by littlemac on April 09, 2006) |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 109
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, April 9, 2006 - 10:17 pm: |
|
>>>The resulting procedure should not only give the voter ample reason to trust the system, but greatly improve voter turn-out as the voter is fully participating in the process.<<<
There are priorities; things that come before other things. Most voters presently believe they have "ample reason" to beleive in the present vote system when in fact they do not. Our task is to create an infallible system first and foremost and worry about voter acceptance last.
>>>The tightest loop (i.e. with least human discretion) , to enable validating the continuous-form ballot, would be for an onsite summarizing computer to receive and immediately retransmit to the internet the subtotals together with ballots and unique identifying numbers from all polling booths.<<<
Once "transmitted" it is "received" at a point removed from the original act of voting in time and space (distance) thus giving anyone intent on altering the vote the "time" as well as the "distance" from watchful eyes, to "make the switch".
>>>By having such a group at the each voting location, they could validate the continuous-form printouts with the internet totals.<<<
They would only be validating a "facsimile" removed in time and distance from the actual vote. Further, the group of "known good character" would only operate in a supporting role, and is relatively insignificant compared to the one most important verifier of all, the voter. The voters integrity, at least so far as their own vote is concerned, is beyond question.
The problem here may be that you and others do not understand the subtleties of all this. May I suggest you take a blank piece of paper. Delineate the top in increments of time from left to right. Left is the zero point in time, right is say 12 hours later.
Now, on the extreme left of the line make a dot that will represent the actual vote. Now draw a line from that point downward to the bottom of the page and delineate that line in terms of "distance" from zero at the point of the actual vote to say 5 miles at the bottom of the line.
If you use graph paper you can actually x-y the "potential opportunity" for the alteration of the vote. Any "sample" that on the graph falls away from the actual point in time and distance where the vote was made, must be open to question. The greather the distance the more opportunity. In other words, for purposes of verification, you can not sample the vote "after" it has been made and expect absolute certainty. It must be sampled at the exact point in time and space (distance) where the vote is made, in order to deny the vote thief the time he will need to make the switch, and the distance he will need to hide his actions.
>>>So now let's continue to generate a workable system using the internet. <<<
It exists. See my website. If you can better it I am always open to improvement, but so far no one has been able to "break" my system/mac |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 6
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, April 10, 2006 - 1:08 pm: |
|
>>>If you use graph paper you can actually x-y the "potential oppoutunity" for the alteration of the vote. Any sample that on the graph falls away from the actual point in time and distnce where the vote was made, must be open to question<<<
So lets describe the steps of the process and plot the time and distance graph:
Time= 0:
The inbooth computer is used to allow voter to make a hard copy, "the official record", that conforms to voters intent on the continuous-form printout. The computer incorporates open source programs, uses no input data lines, and being in public view is protected from humans of questionable intent.
Time=0 to 1 hour:
The plural cast ballot is accumulating audited votes from many voters and creating the official record. The summarizing computer is recording the actual ballots. The inbooth computer is randomizing ballot locations on the continuous-form and compiling vote subtotals. Similar to the inbooth computer, the summarizing computer incorporates open source programs, uses no input data lines, and being in public view is protected from humans of questionable intent.
Note: Up to now no humans except the voters know the content of the official record, and the inhouse computers have not yet released the content of the ballots. (Time = 1 hour and distance =zero)
Time=1 hour + a few seconds
Inbooth computer sends a group of unique ballot numbers together with subtotal to summarizing computer which in turn adds the ballots and broadcasts info to internet. Trusted group at voting site starts to compare internet data with official record at specific voting booth.
Note: Time for humans to intervene and change data = a few seconds, distance depends on how fast the vote thief is traveling
Time=1 hour+ a few minutes.
Trusted group either confirms internet data with official record, or observing corrupted internet data, initiates action to correct hacked transmission.
Result: Any human or computerized corruption is detected within minutes, and in any case the official record is still intact and safe. |
Rick Simon
Voting Rights Forum Participant
Username: Rsimon
Post Number: 2
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, April 10, 2006 - 5:39 pm: |
|
What if the pollworker destroys all of the printed ballots at the end of the night, then puts their spare roll back into the machine, then votes for all the people they want? They they remove the memory card and destroy it so the only record is the VVPAT?
I am terrified VVPATs can be manipulted and revited any way the pollworker, or even election officials want. What can I do! |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 110
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 11, 2006 - 8:29 am: |
|
>>>Note: Time for humans to intervene and change data = a few seconds, distance depends on how fast the vote thief is traveling <<<
Yes, and that's a few seconds too many. Remember Harry Hursti altered the vote count in the Forida test in "a few seconds".
>>>and in any case the official record is still intact and safe.<<<
Unless it has been altered by a homogenous group of vote theives pretending to be poll workers.
>>>What can I do!<<<
You can use a parallel, external vote check system that samples the cast vote at the precise moment and place it is cast. The only system I know of that will do exactly that is the P.o.O.P., or Point of Origin Poll. It will irrefutably check the veracity of any system (paper, computer or anything else) using a relatively small sample, with as few as four people (including the voter) and does not require permission or license of government. This system is here, now.
It not the system you use to vote, it's the certain knowledge that whatever that system is, it has not been compromised/mac
(Message edited by littlemac on April 11, 2006) |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 7
Registered: 03-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, April 11, 2006 - 11:05 am: |
|
>>>I am terrifiied VVPATs can be manipulated and revited any way the pollworker, or even election officials want. What can I do?<<<
During the day of voting the "Trusted Group" will be comparing and validating every subtotal that occurs on the website for that particular polling location. All interested voters will be confirming their vote using their unique number, and many will be downloading an exact copy of the continuous-form ballot.
Now the pollworker providing the counterfeit continuous-form ballot on the next day would have many people to convience that the counterfeit copy was indeed the "Official Record". If only one vote were changed, all subsequent accumulated totals from the previous day would be off by one, causing immediate concern by many. Of course only the person whose vote was changed would know it. Even with a receipt printer, someone else could come up with a fake receit.
However, any group of politically like minded voters can, on the day of voting, get together and share there unique numbers, tabulate swear to and record the results, and be forarmed for such a malicious event
>>>and in any case the official record is still intact and safe.<<<
>>>Unless it has been altered by a homogenous group of vote theives pretending to be poll workers. <<<
At the polling place we have computers running with validated open sourced programs and without any input data lines. They have been recording the voters choices on the continuous-form ballot for only the voters to see. Seconds after the first transmission to the internet, the trusted group of observers is verifying that the subtotal on the continuous-form is identical to the internet data. Now your're implying that in those few seconds the vote theives pretending to be poll workers will be able to beat the trusted group to the continuous-form printer, swap out the official record and replace with a counterfeit one, and not be noticed by the trusted group or voters standing in line. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 111
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 12, 2006 - 8:27 am: |
|
>>>Now your're implying that in those few seconds the vote theives pretending to be poll workers will be able to beat the trusted group to the continuous-form printer, swap out the official record and replace with a counterfeit one, and not be noticed by the trusted group or voters standing in line.<<<
If there is one thing free enterpise capitalism has taught us it is that; if there is a need (legal or otherwise) that exists in the marketplace, entrepreneurs (legal or otherwise) will find a way to satisfy that need. So far your system seems overly complex and underlying that I'm sure is that the basic premise of the system is less than adequate. You therefore keep adding "patches" to compensate for its weaknesses. Each patch brings with it an entry point by whcih a clever hacker might enter.
All attempts to build the perfect mousetrap here have focused on the mousetrap itself (the vote system). Any such system will work, if you have in place an irrefutable alternate vote check system that "verifies" the vote count (similar to a banks recept/statement). That is the direction we should be taking. If such a system shows a discrepancy, the election would have to be rerun and in doing so the electorate becomes fully recognizant of vote theft and associated problems/mac |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 8
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 12, 2006 - 10:39 pm: |
|
{ If there is one thing free enterpise capitalism has taught us it is that; if there is a need (legal or otherwise) that exists in the marketplace, entrepreneurs (legal or otherwise) will find a way to satisfy that need. So far your system seems overly complex and underlying that I'm sure is that the basic premise of the system is less than adequate. You therefore keep adding "patches" to compensate for its weaknesses. Each patch brings with it an entry point by whcih a clever hacker might enter. }
I dislike patches too, so let's return again to the original proposal: An open sourced electronic voting machine with output lines only, one that can relay subtotals and random images of the ballots to a phone line somewhere, and can also print a plural cast ballot on one sheet preventing ballot box stuffing and skimming. This results in a much less expensive and more secure system than the commercial systems presently being purchased. I can't dispute the possibility that theives acting as pollworkers could insert a counterfeit continuous-form printout, even if stiff legal penalties were provided.
Enter then the one and only patch: As you suggest, incorporate the internet to create a valid copy of the accumulating subtotals and ballot immages. This allows the voters to verify that their ballots have not been altered, allows the world to know the latest subtotals during the election day, and provides the ability for anyone to create a duplicate of the plural cast ballot.
I am not specifying how the information is being transmitted from the voting computer in each booth to the internet, whether directly to the internet, or via some interveining voting commission office. It is only after the "trusted group" consisting of members of all major political parties have verified the equivalence of the internet posting with the official plural cast ballot, and the internet posting changed from "not yet verified" to "verified" that the internet ballots will be considered valid. Any subsequent changes to the verified internet posting, observable by the world, will immeditely signal an attempted thieft and would be investigated and prosecuted by the proper authorities. In this case the persons holding a changed counterfeit ballot are suspect.
This results in a system more secure than hand written ballots placed in a ballot box as ballots are constantly being audited by the world during the election. Further, recounts are not necessary as every ballot has been accounted for and has been counted.
{the group of "known good character" would only operate in a supporting role, and is relatively insignificant compared to the one most important verifier of all, the voter. The voters integrity, at least so far as their own vote is concerned, is beyond question.}
With the internet "patch", the group of "known good character" consisting of representatives from all major political parties would be operating in a decisive roll as they would be judging the validity of the internet information throughout the voting day. By their actions the internet has now become a validated copy of the continuous-form printout. |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 502
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 13, 2006 - 4:58 am: |
|
A phone line is not output-only. No ports that can implement error checking or retransmission are output only. You need some form of signal going to the sender to indicate what he's sent doesn't check out and must be retransmitted.
If you want a system that can be safe against insider cheating the voter must have a receipt printed at the time of voting that is legally binding and he must have the right to correct his vote using his receipt. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 112
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, April 17, 2006 - 4:08 pm: |
|
>>>This results in a system more secure than hand written ballots placed in a ballot box as ballots are constantly being audited by the world during the election. <<<
You don't know that. That's an assumption stated as a fact. Without outside, parallel verification, security will never be verifiable, and will always be only a guess.
Elsewhere on this website I wrote about an original idea I called a P.O.P. , or Point of Origin Poll. I claim the results of a POP would be irrefutable and more robust than even the registrars count. A discrepancy between a POP and the registrars count would result in the registrars count being thrown out by default. This is the kind of "outside, parallel verification" I was peaking of above. It is also an area that most people don't seem to want to explore for some reason, except for Mr. Lamb and a fellow that used to post here but hasn't been around in a while, Mr. Trent Wittenbach.
If you really want genuine election reform, may I suggest that it is as close to you now as the archives of this website.
>>>With the internet "patch", the group of "known good character" consisting of representatives from all major political parties would be operating in a decisive roll as they would be judging the validity of the internet information throughout the voting day<<<
All of those "representatives" could be coerced. Take the democrat/republican candidates for President. Both were former "bonesmen", the democrat still had $64 million he hadn't spent as of the night of the election (why?) and that same democrat sure threw in the towel mighty fast, especially considering the fore knowledge he had of major discrepancies between the count and the exit polls. No, the only sure way is to verify whichever system is used, using an external and independent "reference" such as the voters themselves/mac |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 113
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, April 17, 2006 - 4:11 pm: |
|
>>>This results in a system more secure than hand written ballots placed in a ballot box as ballots are constantly being audited by the world during the election. <<<
You don't know that. That's an assumption stated as a fact. Without outside, parallel verification, security will never be verifiable, and will always be only a guess.
Elsewhere on this website I wrote about an original idea I called a P.O.P. , or Point of Origin Poll. I claim the results of a POP would be irrefutable and more robust than even the registrars count. A discrepancy between a POP and the registrars count would result in the registrars count being thrown out by default. This is the kind of "outside, parallel verification" I was peaking of above. It is also an area that most people don't seem to want to explore for some reason, except for Mr. Lamb and a fellow that used to post here but hasn't been around in a while, Mr. Trent Wittenbach.
If you really want genuine election reform, may I suggest that it is as close to you now as the archives of this website.
>>>With the internet "patch", the group of "known good character" consisting of representatives from all major political parties would be operating in a decisive roll as they would be judging the validity of the internet information throughout the voting day<<<
All of those "representatives" could be coerced. Take the democrat/republican candidates for President. Both were former "bonesmen", the democrat still had $64 million he hadn't spent as of the night of the election (why?) and that same democrat sure threw in the towel mighty fast, especially considering the fore knowledge he had of major discrepancies between the count and the exit polls. No, the only sure way to verify, no matter which system is used, is by using an external and independent "reference", especially if based on the voters themselves/mac |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 506
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 5:03 am: |
|
As you have printed the ballots 'randomly' on the roll, there isn't any immediacy to seeing an error, you have to go back and count every vote on the roll to find the problem, and the evidence is in the hands of the elections workers and subsequently the Board of Elections, not the common voting citizens.
Now, the elections workers and the BOE may have the same interest in counting votes fairly and honestly or not, you can't ever be sure. And all of them must have that motivation. If they don't, then you may never get to see that printed strip as it originally existed. Unless there is a complete code 'image' review just before voting, you can't be sure that they didn't adjust the votes in the machine and 'journal' the changes so as to be able to produce a tallied change on the fly and have the journalled information to produce the supporting paper roll at a later time.
If you're going to tighten up elections, the only way I can see to do it is if people can verify their vote. Only their intent can be trusted regarding their specific vote. You also need a correction mechanism that allows the voter to correct their vote. Everyone is trying to protect the ballots that are untraceable and unknown. This makes them staked-out goats. Much of what is done thereafter is to attempt to punish the tiger for killing or eating the goats, when what we really wanted at the end of things is live, healthy goats. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 3
Registered: 04-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 9:17 am: |
|
Brant,
I'm sorry to get into this with you. I know your passion on this. I GET IT! But you will never ever get anything through any legislature in this country that lets any voting system happen that allows a physical piece of proof of how someone voted leave the polling place with the voter. It just politically is not ever going to happen.
You may get paper trails, even paper ballots in the traditional sense, but you're just never going to get a receipt that can be carried away. Just get used to that fact.
Besh wishes otherwise.
Kurt |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 4
Registered: 04-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 9:23 am: |
|
I just read the whole thread above.
You're never going to see running subtotals during the day either. The future trend is to delay results until later. There is no apetite for partial counts or early counts.
Where do these ideas come from? I know you guys are "outside the box" thinkers, but things like partial counts during election day and carry away receipts are pure anathema to most who know elections and their history.
Kurt |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2126
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 10:41 am: |
|
Hi Kurt,
I agree with you 100%.
Some of these proposals or similar ones have been discussed on previous threads, so that may explain why you don't see more dissenting opinions on this thread.
Since you have so much professional experience as an elections director your viewpoint will be especially valuable.
(Message edited by catherine_a on April 18, 2006) |
mike sinder
Voting Rights Forum Participant
Username: Sinder
Post Number: 2
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 12:19 pm: |
|
Voting:
Step one: Voter casts preliminary ballot on electronic machine;
Step two: Voter confirms ballot on electronic machine;
Step three: Voter receives printed copy of confirmed ballot
Step four: Voter instructs machine to accept electronic ballot and places printed copy in secure ballot box
Tabulating:
Step one: Electronic totals are unofficial and reported as available after polls close
Step two: Random precincts/races selected for hand-recount to verify electronic results
Step three: Electronic results become official if no discrepencies are found or hand counts of printed paper ballots are ordered to get official results if discrepencies occurred.
No hanging chads, no question of voter intent, no reliance on chips or machines, accurate counting of EVERY vote. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 6
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 12:31 pm: |
|
Mike:
I like it, except for one tiny problem and/or question. Do you re-hand count if there is a one or two vote discrepancy in a random precinct?
I ask this because PA election case law is loaded with recounts done of optical scan or punchcard ballots after a recount was ordered by a court. There are FREQUENTLY small discrepancies (I mean less than enough to affect the outcome and in the single digits). Do you really trust the hand count more? I wouldn't, and neither have our courts in PA. In those cases, they nearly always have accepted the machine counts. This does vary from county to county however. The inconvenient truth is that even carefully selected teams of hand counters given pre-verified stacks of ballots get the count wrong an astonishing proportion of the time.
Kurt |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2128
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 1:37 pm: |
|
In Ireland all counting is done by hand by teams of 2 people who must each independently get the identical number. There are lots of observers to this process, including members of the general public as well as representatives of all the candidates. Counts are more complex because they involve ranked voting of many candidates (e.g. up to 20 for some elections), and there are many stages of vote transfers and recounts. So it would compare fairly well with the US ballots where you have lots of races/issues but far fewer candidates. The counting process is accurate, fair and transparent--and quite exciting.
In Italian hand-counting there is a similar process of checking and cross-checking by several individuals who must independently get the identical result. When done like this hand counting can be extremely accurate.
This is separate from ballots that are disputed for one reason or another. This is where most of the "differences" occur in Ireland, and all candidates are allowed to have their representatives participate in the process of evaluating all questionable or disputed ballots.
Re: 1 or 2 vote discrepancies--that's probably a more important issue than is often recognized, where there are electronic vote counting machines in place. I read a paper that said that in the Ohio 2004 Presidential election, changing only 6 votes at every precinct would have been enough to change the overall results. The difficulty then arises that a small difference like this at each precinct would not be enough to raise concerns of election officials, candidates or voters, since 6 votes on their own would usually not be enough to change an election result. Yet a small error, occurring over many precincts, could indeed change the results. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 114
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 1:52 pm: |
|
>>>Since you have so much professional experience as an elections director<<<
>>>I just read the whole thread<<<
So, if you are an "elections director", then that implies you are part of the system that is stealing the vote.
It's clear that the only way to make meaningful forward progess is to use alternative, parallel and independent methods to validate the vote. What most of those who, aren't "out of the box" thinkers, are trying to do, is analogous to buiding the worlds most inpenetrable vault, then not equiping it with any kind of parallel alarm system.
The Point of Origin Poll POP is just such a parallel system. A system that operates independent of government, is so robust as to be "irrefutable", and probably scares the hell out of those who are presently stealing our vote.
I challenge you to dialogue with us on alternative, parallel systems instead of attacking our work, because your comments seem disingenuous and only serve to dissuade us from getting closer to the final solution/mac |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 507
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 1:56 pm: |
|
Cheat, scenario one:
Voting:
Step one: Voter casts preliminary ballot on electronic machine;
Step two: Voter confirms ballot on electronic machine;
Step three: Voter receives printed copy of confirmed ballot
Step four: Voter instructs machine to accept electronic ballot and places printed copy in secure ballot box
Step five: poll worker replaces ballots in 'secured ballot box', reports which 'random' precinct to accessories as to which precinct to count.
Tabulating:
Step one: Electronic totals are unofficial and reported as available after polls close
Step two: Random precincts/races selected for hand-recount to verify electronic results
Step three: Electronic results become official if no discrepencies are found or hand counts of printed paper ballots are ordered to get official results if discrepencies occurred.
No hanging chads, no questions raised about voter intent, no reliance on chips or machines, inaccurate counting of what the original votes actually were. The voters have no evidence their ballots were subverted.
Cheat, scenario two:
Voting:
Step one: Voter casts preliminary ballot on electronic machine;
Step two: Voter confirms ballot on electronic machine;
Step three: Voter receives printed copy of what he thinks is his confirmed ballot, however it has not been recorded as it was displayed on the machine, and the printed ballot matches what the machine recorded, not what he voted and he doesn't check it
Step four: Voter instructs machine to accept electronic ballot and places printed copy in secure ballot box
Tabulating:
Step one: Electronic totals are unofficial and reported as available after polls close
Step two: Random precincts/races selected for hand-recount to verify electronic results
Step three: Electronic results become official if no discrepencies are found or hand counts of printed paper ballots are ordered to get official results if discrepencies occurred.
No question of voter intent, no evidence for proof of election fraud and few the wiser.
You're not going to block taking someone's ballot choices away until the voter can check on anc correct his vote.
You won't stop fictitious voters until you have a method of (at least) a voter voluntarily proving a vote is his.
And Kurt, I do know elections history. What I believe that you are failing to account for is how much the history as it was, in total, prejudices people against different choices that were made to fit previous historical context.
It is possible to use then non-existant technology, like internet access to allow people to report coercive behavior and avoid having a ballot copy become an item used as leverage against the original voter. Being able to check your vote and correct it is the only way a voter is ever going to have any certainty that his ballot was counted as he cast it.
This then leaves the worry of whether fake people have voted and that can only be beaten by strong, free, checked voter ID. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2129
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 2:07 pm: |
|
Mac:
"So, if you are an "elections director", then that implies you are part of the system that is stealing the vote."
Really? Is that what you assume? Perhaps in some cases that could be true.
Assuming someone is an ethical elections director, they will also have practical experience and yes, perhaps also some biases in terms of what is easiest to manage. But don't underestimate the value of experience including knowing one's staff, knowing one's voters, and understanding the implications of the budget one is likely to be allocated. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 7
Registered: 04-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 2:55 pm: |
|
Thank you Catherine, but mac, I WAS an Election Director (past tense). I no longer am one. I know personally about 63 of PA's 67 county election directors or their equivalent position, and ALL are the most honest hard-working folks you'll ever encounter ANYWHERE. This need of some to impune election officials is just simply insane. Now if you want to go after the vendor community for treating HAVA as a license to steal with shoddy products, I'm with you! And incredibly enough, so are some of yur most frequent targets - even people like Mike Shamos and Doug Lewis of the Election Center. I was at a conference in Greensburg, PA in 2002 where he spoke right in front of the vendors and told them they were too cheap to hire enough people to avoid a disaster when HAVA took effect. He told all counties to hire their own experts because the vendors weren't going to be there for them. Most counties haven't done it. They're cheap too. They have an excuse. They're spending taxpayer dollars.
But E.D's are underpaid overworked public servants doing their best to do what's right with inadequate budgets passed by politicians who don't usually give a rip about elections other than their own.
The Congress demands more of them and the local politicos try to make them do that more with less every year.
No one stole anyone's vote, mac. Screwed it up? Maybe. Did dunderheaded things to screw it up? Maybe. In over their heads on the law or technology? Probably. Stole? No way in hell!
E.D.'s aren't wired that way, not even the worst ones are wired that way. Their integrity is frequently all they have to be at peace with. They certainly aren't getting rich, outside of a few megalopolis locales. And most don't give a damn who wins elections, because as E.D.'s we see politicians all the time and we're not impressed with them. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2130
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 3:39 pm: |
|
Kurt,
"He told all counties to hire their own experts because the vendors weren't going to be there for them."
Was that Mike Shamos? Or was it Doug Lewis?
What kind of expert did they think the counties should hire? Was he talking about computer experts? (or legal consultants?)
I doubt the vendors ever informed the E.D.s that part of the expense of the systems would be independent IT consultants to rescue them on election day.
E.D.s must be caught between a rock and a hard place--saddled with lots of regulations, but not given the resources to do what is required. Combine that with ITAs who aren't doing their job (and are in bed with the vendors it appears from the ITA testimony to Debra Bowen's hearing), and you have a guaranteed mess.
Whatever hankey pankey might have gone on with some elections, it would be absurd to jumpt to the assumption that election officers were part of any problem.
From everything we know from the Hursti reports, and from hints about what was found in Utah, there are back doors in the hardware/firmware/software that could allow votes to be affected without any E.D. ever knowing, and that even the best internal procedures would neither prevent nor reveal.
I will be glad when the Utah report comes out so that we know more about what we are up against. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 8
Registered: 04-2006
Best of Black Box? 
Votes: 2 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 4:34 pm: |
|
Catherine,
That was none other than Doug Lewis. I have learned about Doug's role in insulating the ITA's from questions, and that stinks, but Doug Lewis is otherwise a bright man in elections, and before you get too warm and fuzzy with me, you should know that I hold Mike Shamos in VERY high regard. I know lots of folks don't like what he did re: paper trails in PA certifications, but under PA law as interpreted by the courts and the Department of State's professional staff, he had no choice.
But when Shamos criticizes the ITA's, that holds great weight with me.
If PA ever gets paper trails it will have to come from the legislature. There is no interpretation of current PA law that would allow them.
The type of expert Doug Lewis suggested each county have was to program the DRE databases themselves, without need for vendor personnel. That means programming ALL the data tables (Election, Offices, Parties, Candidates, Terms of Office, Districts, Precincts, Machine Assignment, Referenda, etc.) as well as all links between them, and all absentee ballot design and master ballot sheet design. I did ALL THAT myself without a vendor's help.
But Cath, there IS one thing we can blame E.D.'s for - lack of diligence. There is adequate reason to understand that these systems pretty much S*CK. E.D.'s can be faulted for ignoring the problem if they do. They should WELCOME, even INVITE "Clean Up Crew" scrutiny. You are right to be suspicious of any locality who fears explaining what is happening on election night when problems start to happen. In 4 years (8 elections) I never had a perfect problem free election, but I always knew how to not let problems screw around with an accurate count. I'm not sure all E.D.'s can make that statement any more, not with these new DRE's. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 9
Registered: 04-2006
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 4:59 pm: |
|
Brant,
The point is - you are throwing way too much technology at the problem. I understand that given the parameters you chose, you may not have an option. But please Brant, believe me, we already have more technology in elections than people can handle. We're pooping all over the KISS principle - Keep It Simple, Sir.
There are not enough tech savvy people in local governments to carry off DRE's as they are, much less more layers of tech complexity.
"internet access to allow people to report coercive behavior" What? Do you get that the people most likely to be coerced are the poor, ignorant, and elderly? They're less likely to have internet access than I am to sprout wings and fly.
I'm not saying your system doesn't meet its stated goals. I'm saying it's politically impossible to enact, and unduly complex. The answers simply must be less techy and more simple. True after-the-fact verification may well be a pipe dream you may find yourself having to let go of. The best you may get is "before you leave the booth" verification. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2131
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 5:13 pm: |
|
Why do you assume I have anything against Shamos? I don't think much of his $10,000 "challenge" but I don't doubt his intelligence and I'm sure he has a lot of good ideas.
You were really wise to do all your DRE programming yourself. What made you realize early on that you should do this yourself?
What would be a more effective way of communicating with E.D.s so that those that don't perceive the risks of these systems can better understand them, without them feeling under personal attack? Does it help if they are educated by one of their peers? Is it a question of supplying the right documentation in a way that non-IT-savvy E.D.s can digest? What would you recommend? Have you seen any particularly ineffective communication strategies that you'd suggest concerned citizens avoid?
Perhaps in some states E.D.s might come under pressure from other quarters that might limit their range of response. (E.g., if a SoS decides that a particular kind of equipment must be used, or will not be certified, that would limit an E.D.'s choices. Or in subtle ways an SoS or someone with budget-making clout drop a hint about what machines they'd prefer to be used.)
I imagine that old-fashioned budget or time pressure could make it hard to take in new information, no matter how well intended, if one is just clinging on trying to get from one day to the next with limited staff, limited funding and limited expertise in some cases.
It's probably different in every location. It would be interesting if you have any observations from your personal experience and that of peers. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 11
Registered: 04-2006
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Tuesday, April 18, 2006 - 7:00 pm: |
|
No Catherine,
I did not mean to impute any personal dislike of Shamos by you. I was just commenting on the general feeling I get taking the boards as a whole. "Shamos likes paperless DRE's so he must be evil" kinds of vibes.
Shamos is bleeping brilliant. BTW, do not EVER play pool for money with him. I saw him play at a conference before I knew he writes books on billiards. He is brilliant on the green felt, too.
Last year, at that conference at a hotel near Dulles airport, Shamos mentioned that he thought the only DRE he really trusted fully was the Danaher, and even then only by a hair. "Best system on the market currently." The reason was that the programming is ALL in firmware, no disks, no cards (other than the database tables - and then only in proprietary hardware cartridges) (No off-the-shelf memory cards), no way for anyone to rig the software.
I say he's right. No Danaher employee ever even SAW my databases. The very same button that was set for a candidate in one precinct could be his opponent's in the next, so how could anyone at the vendor do anything? All names, parties, etc., are string variables only, enetered by me. The only thing a Danaher machine knew was how many times, for example, button D30 (fourth column, 30th row) was pushed. The only programmer that gave a name or party to D30 was me, and it could be and was different for different precincts. In fact, I demonstated how the very same button number could be Bush's in one precinct and Kerry's in another. Now many counties using Danahers don't do that, but they can.
BTW, Danahers ARE federally certified. I don't know what Connecticut's problem is. Maybe their paper printer roll was not federally certified. I know CT wanted in on their system. PA wanted it off.
I programmed my own databases because it was expected of me. It was why I got the job. I told the Commissioners I could do it, and I could. It nearly killed me. The Danaher software stinks. It is REALLY buggy and user unfriendly, especially in how it defines offices in primaries and how its default ballot layout ALWAYS is wrong. |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 508
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 5:18 am: |
|
I have never suggested DRE's, Mr. Bellman; you either haven't read my posts or weren't paying sufficient attention.
The idea that all of the voting machine glitches favored Bush demands a willingness to believe in a tremendous long shot of coincidence that I find ridiculous. I'm not saying that elections directors cheated, but somebody did.
I find the characterization of 'too much tech complexity' to be insubstantial as the technology that I suggest is easily checked for it's correctness and verifiability and easily electrically isolable to prevent voter-undesirable reporting.
I appreciate your opinion that it would be difficult to get backing, I disagree strongly with your estimation of its merits. As to it being unduly complex to meet its goals, I challenge you to render a simpler system that actually meets its goals, which is to render a method whereby a voter can trace his ballot, verify it, and correct it without the retainer of the ballot box copy being able to trace it to the voter. I suspect that what you meant is that it's
unduly complex to meet your goals.
And forgive me for doubting this, but did it occur to you that when you enter a candidate's name in your database table, don't you have to enter the party affiliation as well, since it's going to be displayed to the voter? If so, what makes you think that the machine couldn't cheat based simply on party affiliation?
Has someone verified their source code and run the identically programmed and configured machine on a disassembling logic analyzer to see that the code that they've reported as the source code is actually the code it's running? |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 13
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 6:04 am: |
|
Brant,
Yes, because the code is unchangeable. It's like an old Commodore 64. The whole of the operating system is on firmware chips. All the software is. The machine doesn't know anything about parties. It is designed to be used worldwide, even in places that don't know what a Republican is. If I had accidentally entered "Rebublican" as the party, then the whole system would report "Rebublicans" everywhere. The machine simply does not know about, test for, or even acknowledge anything about any parties. The party label is merely a text string that the machine slavishly reports back on the print-outs. It is a variable that is never tested for or used in the logic at all, anywhere.
The conspiracy types in this movement who believe that somehow Republicans are out to steal elections are just really getting too wacky for words.
Now is it POSSIBLE, on SOME SYSTEMS? I suppose it might be, but not on a Danaher 1242. They are designed to be used worldwide, including in 3rd world countries that build their ballots based on PICTURES of candidates and party LOGOS. There simply is nowhere in the system's logic where any U.S. party labels are ever considered.
And while in PA elections, I had to enter a party label for each candidate, that was merely because the law requires it to be in the candidate's box along with his name in a November election, and so the name operates in the correct party in a primary, wherein the office of "Commissioner-D" and "Commissioner-R" are treated as two completely different offices.
You really do need to get over this "Republicans stole the election" rant. It simply did not happen, sir.
I oppose these new DRE's precisely because I'm concerned that it COULD and MAY SOMEDAY happen, especially when the whole nation uses 5 or 6 systems, as opposed to the myriad systems we had up to now.
The best protection we had was our electoral system diversity, and we've now blown that up.
Sad.
By the way, the Danaher uses a Motorola 68000 processor to run that chip embedded firmware code, if that means anything to you. And no, I don't mean a 68000-series chip. I mean the old, same as an old Macintosh, 68000. It is O-L-D technology, and it works like a charm. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 14
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 6:42 am: |
|
Brant,
One more thing for you to consider. The Danaher 1242 is used by Pennsylvania's most overwhelmingly Democratic county, Philadelphia. In many precincts in Philly, the Green Party candidate for Governor outpolled the Republican candidate in 2002.
The entire Philadelphia election office is crawling with Democratic patroage employees. In 2004, Kerry carried Philly by 542,205 to 130,099. Do you really think that Philly would be using Danahers if there was ANY CHANCE IN THE WORLD that they could be hacked to favor Republicans? C'mon. Your paranoia is showing, Brant. By the way, the other two PA counties that use Danahers are one that is slightly Democrat, and one that is slightly Republican. So when you consider that Kerry won PA by about 120,000 and he had a 400,000 vote margin on Danahers, gee, the Danahers ought to be your heroes. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 15
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 6:52 am: |
|
Brant,
To put an even finer point on it - if the 2004 PA election were limited to only the places with voter verified ballots, Bush would have won in a walkover. Only the paperless DRE's created Kerry's PA win. The lever machine counties were close to a wash.
The point is this - electronic machines and Republican success in elections has nothing to do with each other. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 16
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 7:06 am: |
|
Another case in point: Montgomery County, PA
(3rd largest in PA after Philly and Allegheny)
Overwhelmingly Republican by registration. Election office run by a self-described Republican patronage employee. Quote: "If they want to get me out, get more Democratic votes."
System used: Sequoia Advantage full-face DRE (paperless)
2004 Presidential Result in county:
Kerry 222,048
Bush 175,741
Any questions? Funny how when you look around enough, many facts contradict the "prevailing wisdom". |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2139
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 7:24 am: |
|
Kurt,
There is evidence of both major parties manipulating elections. This probably goes back as long as elections have existed; it is not news. I see no reason to expect that this would change when electronic machines are used.
Bev's book, which you can download free on this website, documents some of the many examples from recent decades, many of them including results from electronic machines that are impossible or unbelievable, or were known to be faulty (but with no way of correcting them after the fact, or with people who refused to allow inspection to occur).
This has happened with both main parties. Some groups in some locations may have more expertise or opportunity, and levels of integrity vary. A partisan person could throw an election without the knowledge of the candidate or the local election official. Most of the modern electronics make this scenario easier to do and with less risk of detection.
One should assume that attempts to control elections will occur regardless of the voting technology in use or the party in power. The job of citizens, E.D.s and legislators should be to put all possible safeguards in place to ensure accuracy and transparency, such as well-documented chain of custody of ballots. Thsee must be backed up with effective and even-handed scrutiny, accountability and enforcement.
There are questions raised about elections that favored Democrats as well as elections that favored Republicans. It is likely that elections have been stolen by both sides. Pointing to any party as winner is not a reliable indicator of whether fraud did or did not occur.
Anyone who tries to frame this as a partisan issue is missing the point. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 17
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 8:20 am: |
|
Cath,
Once again, dead on!
"A partisan person could throw an election without the knowledge of the candidate or the local election official. Most of the modern electronics make this scenario easier to do and with less risk of detection."
This one is great (and troubling). In the aforementioned Montgomery County, PA, the previous version (pre-HAVA) of their tabulation software (Sequoia) passed the security tests. When the new (post-HAVA) version was submitted for Dr. Shamos to review, he brazenly tried to demonstrate how the results file couldn't be hacked (as in the past and as in the Danaher) and found, SHAZAAM, it could be hacked in the new version! He just opened the mdb file in Access, changed the results and whammo, new results! No lock-outs, and no mention in the audit logs of the intrusion into the file. That's bloody scary, and Sequoia left that puppy in there! Why is that?
Try to imagine being a programmer for one of these firms. EAC and NASED standards were horribly late. ITA certs horribly late. Rush to get the things out. Mistakes happen. Shoddy, poorly documented code gets put out. Security holes left open for easy development work get left open. All because no one in Congress had the b*lls (use an 'i', if you like) to delay HAVA's deadlines so that this could be done right. They preferred to keep the devil they knew. Shame on them all. They were so afraid of reopening Pandora's box that they couldn't fix an obvious mistake.
HAVA is, as the Brits say, a dog's breakfast. It will result in the ruination of public's confidence in elections for a generation. All because some Congressmen couldn't admit they went off half-cocked in 2002 when they cobbled together this piece of trash called HAVA.
But what troubles me here is that not all of your members have the judgement of you and Bev and a few others in leadership. For many, this is just another "damned Republicans stole the election" rant site. That's just crazy talk, in my opinion. The problem is that these machines MOSTLY ARE hackable. I don't think it has happened yet, I really don't. Cheaters usually aren't this far ahead of the curve. But these machines make an overwhelmingly tempting and easy target, and if they are not done away with, the cheating WILL happen and WILL be darned difficult to detect. (I won't go to impossible.)
The thing that annoys me is all the charges flying around about Ohio (phantom votes, ridiculously long lines, registration fraud) are decried in the blogosphere because they all seemed to favor Bush. Well Cath, all those same things happened here in PA, too, and they all benefitted Kerry. I can document that with proof. Remember, I was there, in the trenches. No one talks about PA, because Kerry won it. But he won it with almost the exact same margin by which Bush won OH. Had Kerry won Ohio, I suspect we'd have the same ranting from Republicans about PA, and not without reason.
In my county, lines in heavily Republican precincts were up to 4 hours long, while in Democratic precincts, lines were maxed out at 30 minutes. Bush carried my county by about 10,800. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 18
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 8:38 am: |
|
Cath,
Responding to this:
"What would be a more effective way of communicating with E.D.s so that those that don't perceive the risks of these systems can better understand them, without them feeling under personal attack? Does it help if they are educated by one of their peers? Is it a question of supplying the right documentation in a way that non-IT-savvy E.D.s can digest? What would you recommend? Have you seen any particularly ineffective communication strategies that you'd suggest concerned citizens avoid? "
Wow, Cath, you ask really good tough questions. I'm really afraid that most have been brainwashed by the vendors and frightened by the state and federal authorities. The immediate past threat was the loss of the HAVA dollars. No E.D. could risk losing millions for a county that the USDOJ would then require them to spend anyway but without the federal dollars. Not many E.D.'s could keep a job under that scenario. They had to get the federal bucks or be gone.
Unfortunately, it may require a Cook County-esque meltdown more places before people "get it". You read the stories. I read the stories. I'm not sure most E.D.'s read electionline.org. They should. So should BBVers. Some of that stuff will scare you! And it should! These new systems are B-I-G problems. They are difficult to use and poorly conceived. They intimidate poll workers, and those poor folks are just getting over provisional balloting. And now this??
I read in my local paper that my successor is having an awful time getting poll workers. I'm not surprised. All this is getting to be too much for these 2-times-a-year folks to deal with. What's worse is under the PA Constitution, almost no public sector employees are allowed to work in a polling place. That leaves private sector or retirees, or the unemployed. Guess what we get. They're usually not top notch, sorry to say. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 115
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 8:44 am: |
|
>>>You really do need to get over this "Republicans stole the election" rant. It simply did not happen, sir<<<
See what I mean. This is typical neo-conservative double speak. This will do nothing to convince anyone actively engaged here that the republicans didn't steal the election.
What this kind of rhetoric will however do is influence those "lurkers" who haven't delved into this subject deeply. Lurkers take there "clues" of whose telling the truth; not by evaluating the truth, but by noticing who is behaving in the most aggressivly "hurt" manner. The ones speaking of their tireless diligence and hard work. All of it seemingly designed to create an image of being put-upon. Reading the above will certainly persuade many lurkers of the validity of "You really do need to get over this "Republicans stole the election" rant. It simply did not happen, sir."
Further, notice there is no mention of the simple, no cost method of verifing elections; i.e. Anonymously Publishing The Vote, even after an open challenge to engage in dialogue regarding them.
These forums aren't just areas for people to express their opinions, share information and generally nudge forward the human condition. No, it is a battlefield. A battle fought for the minds of those who view it all, not unlike that other battlefield called, "television".
Sir, if you are sincere than accept my challenge. You're an "expert", tell us where's the flaw in our thinking regarding the POP and Anonymously Publishing the Vote. Two simple, neglible cost alternative methods of verfying the vote. If you're not up on them, then view the original article at www.howtorescueamerica.com
...
By the way, Ms. Ansbro.
It's so lovely to have you in on this conversation. Tell me, how is your native Ireland in Springtime? Is it just as beautiful as the pictures we've seen? It must be wonderful to be there this time of year/mac
(Message edited by littlemac on April 19, 2006)
(Message edited by littlemac on April 19, 2006) |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 116
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 8:51 am: |
|
>>>You really do need to get over this "Republicans stole the election" rant. It simply did not happen, sir<<<
See what I mean. This is typical neo-conservative double speak. This will do nothing to convince anyone actively engaged here that the republicans didn't steal the election.
What this kind of rhetoric will however do is influence those "lurkers" who haven't delved into this subject deeply. Lurkers take there "clues" of whose telling the truth; not by evaluating the truth, but by noticing who is behaving in the most aggressivly "hurt" manner. The ones speaking of their tireless diligence and hard work. All of it seemingly designed to create an image of being put-upon. Reading the above will certainly persuade many lurkers of the validity of "You really do need to get over this "Republicans stole the election" rant. It simply did not happen, sir."
Further, notice there is no mention of the simple, no cost method of verifing elections; i.e. Anonymously Publishing The Vote, even after an open challenge to engage in dialogue regarding them.
These forums aren't just areas for people to express their opinions, share information and generally nudge forward the human condition. No, it is a battlefield. A battle fought for the minds of those who view it all, not unlike that other battlefield called, "television".
Sir, if you are sincere than accept my challenge. You're an "expert", tell us where's the flaw in our thinking regarding the POP and Anonymously Publishing the Vote. Two simple, neglible cost alternative methods of verfying the vote. If you're not up on them, then view the original article at www.howtorescueamerica.com
...
By the way, Ms. Ansbro.
It's so lovely to have you in on this conversation. Tell me, how is your native Ireland in Springtime? Is it just as beautiful as the pictures we've seen? It must be wonderful to be there this time of year/mac
(Message edited by littlemac on April 19, 2006)
(Message edited by littlemac on April 19, 2006) |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 19
Registered: 04-2006
Best of Black Box? 
Votes: 3 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 9:58 am: |
|
mac,
The essential problem is that your system is illegal, and actually blatantly unconstitutional in my state. No system that compromises secrecy of the ballot is allowed in PA. Secrecy of the ballot is the legal "prime directive" in Pennsylvania. It trumps all. It trumps verification, it trumps correction, it even trumps accuracy. Until the feds forced it here, it even trumped assistance for the blind or handicapped. No one was allowed to observe your vote, even if that meant you couldn't vote.
Secrecy of the ballot has been held by our courts to have two components.
1) No one may learn how any person has voted, without the voter's consent.
2) No one may be allowed to prove to another how he has voted, other than through actual presence in the booth during assistance pursuant to Section 208 of the Voting Rights Act. His only option is to state how he voted. Possession of a ballot, or its facsimile, outide of a polling place is a crime in PA, except for absentee ballots, which must be submitted by three days BEFORE the election.
You see, there is a near-fetishistic fear of vote buying and selling and coercion in PA. Why? Because its history in PA is legendary. It used to be impossible to get a job in anthracite coal operations unless you proved you voted Republican, for example.
As far as any method to be able to discern ANY results before all votes are cast, that, too, is a crime. No partial returns are EVER allowed until the polls close. We have no "early voting" in PA, and one reason is because people are afraid that partial results will somehow get out.
The problems are primarily legal mac, not primarily technical. But as a general rule, technology in voting needs to be minimized, not overdone.
So, as tempting as your solution may be, and meets the needs you set for it, you don't get to spec the system, and neither do I. Voting systems are ultimately speced by politicians and judges, through statute and case law. It is an interesting mental exercise (one you've done well) to spec a system with goals and needs outside of that legal framework, but ultimately not very useful in a practical sense.
Now if you are working with a really long-term time horizon, and are aware that the journey of a thousand miles begins with a single step, then your efforts may not be in vain. You may begin soemthing that you are not around to see the end of, and if that's okay with you, good for you.
But you really do delude yourself if you think (and I am not presuming here that you do) that anything like this will see fruition anytime soon. The politicians, and the rank and file voters, simply won't be able to deal with it. Now if you have an anarchist bend, and "being right" is all that this about, then have fun.
As for believing the 2004 election was stolen, if believing that helps get you through the night, it's all right. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2141
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 10:03 am: |
|
Mac,
You're wrong to characterize Kurt Bellman's remarks as neo-con type double-speak. This is the kind of name-calling rant that drives many folks around the bend, and rightly so. It's rude and inappropriate to this website.
Keep your passion but marry it with manners.
As Kurt points out, in PA he saw signs of impropriety that benefitted Kerry. And PA's not the only place, I'm sure. Just ask Washington State voters. I'd love to know more about what Kurt observed or heard--it would be of great interest to me.
This is a non-partisan issue. Elections should be fair, regardless of who wins or loses.
Might the combination of election problems have affected one, several, or many elections? Very possibly. We don't disagree about that, mac. At BBV we're all trying to get as much hard data as possible so we can move more data from the "could have happened" or "might have happened" into the "did happen" category. Then one has to get even more data so that one can determine which issues were related to sloppiness and accidents (and how might this be prevented in the future), and which were deliberate (and how might this also be prevented).
Doing this is a careful, slow, step-by-step process. There are lots of bad smells emanating from these systems but we've got to attempt to discover and analyse the sources of the smells with a great deal of care because a lot is at stake: the reputation of candidates, election staff, businesses, local/state/national/global policy, and democracy itself.
One of the reasons I've got such respect for BBV is that they have taken such care to
* always adopt a non-partisan stance on the election/voting machine issue, being willing to follow evidence wherever it leads
* always stress the need for obtaining hard evidence rather than conjecture (though acknowledging "smoke" where it is seen)
* demonstrate and share effective techniques for getting hard evidence so that any citizen can get involved in a way that fits their interest and abilities
* share the fruits of its investigations as widely as possible so that democracy benefits
Unlike other parts of the blogosphere BBV is not a place one comes to for wild accusations. The fruits of its investigations speak far more eloquently and effectively than the rants and mud-slinging on other websites.
(Message edited by Catherine_a on April 19, 2006) |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 21
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 11:09 am: |
|
Cath,
In the spirit of your post, I want to be clear about the pro-Kerry stuff in my county. The long lines in Republican precincts were unavoidable. In 2004, the ballot was short in PA (no local offices) so time in the booth was never a problem, even at the busiest precincts. The problem was voter check-in that took too long. It just tok too long to find voters in the pollbooks. Efforts to divide the lines by letter nearly ended up in fistfights because it always seems the "other line" moves faster.
We had had tremendous population growth in Republican precincts (suburban sprawl with McMansions) over the previous years. Because the PA Congressional redistricting went to the Supreme Court due to a CRAZY Republican gerrymander (Vieth v. Jubelierer), we were frozen in our existing precincts by state law. We had some precincts with over 4000 voters. The lines were horrendous. All the huge growth areas were heavily Republican.
The other "shady stuff" was not done by Kerry's folks but by 527's (IRS code) acting in his interests. About two dozen left-leaning 527's decended on my county because it was a swing county in a swing state. They illegally paid incentives to people to round up registrations. We had a duplicate rate of over 80%. (80% of registation forms were from people already registered.) Many of those forms had wrong DOB's or adresses or ZIP codes by a digit and resulted in duplicated registrations. The 527 people had refused to let voters fill out the form themselves. They insisted on filling it out themselves and just let the voter sign it. We had to hire dozens of temps to get all the forms processed. Quality of the processing suffered.
The worst part for Kerry was that the "Election Protection" squad people swarmed into heavily Democratic areas accosting all who approached a polling place asking them for their name and address, and asking to see their voter ID card. They were wearing outfits that looked intimidating. Many Hispanic voters contacted me later and said they were frightened by the "guys in the black shirts" at the polls and just went home.
What the experience showed me was that too many well-meaning advocacy groups swarming in a single area can overwhelm already concerned voters, to the ultimate detriment of their own goals. The result was that while turnout was up everywhere, it was up more in Republican areas. Democrat activists need to understand that their voter base includes some people easily frightened, easily dissauded from voting. They need to avoid being too large a crowd. All the various 527 gorups and "Election Protection" groups could have benefitted by coordinating their efforts, or at least realizing that if one Democratic advocacy group is good at a poll, two may be better, but 20 isn't necessarily.
What all those unnecessarily duplicated registrations did was make it impossible to get ID cards to all REAL new registrants. We were still processing new registrations on the Sunday before the election. At one point in October, our unprocessed registration stack was 22 feet thick. And the SURE system was a dog - very slow.
As it happened, Kerry won PA, but people working "for him" didn't help him much in my county. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2143
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 11:16 am: |
|
Thanks very much for that. Very instructive. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 22
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 11:21 am: |
|
Cath,
One more thing - provisional balloting was new in 2004 in PA. The 527's rounded up people off the street who said they wanted Kerry to win and just took them to any polling place. We had dozens of provisional ballots from people who admitted on the envelope that they lived IN OTHER STATES! The 527's didn't care. They just were rounding up votes, in their minds. "Go to any polling place. Ask for a provisional ballot!" was the cry. That's just pure irresponsibility. They had tons of people believing that anyone could vote anywhere. It was nuts! |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 509
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 11:46 am: |
|
Are you trying to say it's impossible to vote a straight party ticket by a single choice in PA?
It's done almost everywhere else in the country, in fact listing party affiliation is a requirement in most if not all of the states, isn't it? If it is in the machine and is associated with the person's name in some table it is possible to use it to affect recording of the vote. "The party label is merely a text string that the machine slavishly reports back on the print-outs. It is a variable that is never tested for or used in the logic at all, anywhere.
Yet it shows up in the printouts, associated with, I assume, the correct candidates. Interesting, if that happens coincidentally with any reliability. This answer doesn't bear up to logic.
That the programming is in firmware doesn't prevent all hackability, it just means that it has to be a hardware hack where you replace the chips or reprogram them. Execept for flash PROMS, these are nearly always socketed, otherwise you would have to de-solder a 24 or higher pin chip to be able to upgrade firmware on a machine, and this would be tremendously expensive.
While I know how to use prom readers/programmers, procedures are not in place to make this a safety feature unless access to these machines is controlled by some large non-partisan or thoroughly mixed-partisan and large committee.
This really isn't about partisan politics for me, it's about how exit polling is used even by our national government to grade elections in other countries but it just had to be wrong here. That is doubtful; Occam's razor leaves nothing bigger than a sliver of that argument. You don't address that where exit polling was supposedly 'wrong' all the deviations went to Bush. The odds of this are astronomical.
Your argument that cheating didn't seem to have happened in particular places isn't any form of rational argument that it didn't happen somewhere else, this argument holds no water. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 23
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 12:20 pm: |
|
Brant,
I do not wish to belittle your concerns, and believe me, I share MOST of your concerns about the Diebolds and most of the new generation of machines, but please, keep the "broad brush" in the tool belt, will ya'?
I know which precincts VNS chose in my county. They oversampled Democratic areas by a great deal. Can I generalize about that nationally? Nope. Just reporting on what I know.
Now to assume that my Danahers, in constant service since 1989 (then known as Shouptronics), were part of a plan to rig the 2000 or 2004 election, one would have to believe that a prescient programmer in 1988, when the ROMS were made, put malware in there to sit idly by for 12 or 16 years, electing Democrats and Republicans in roughly equal numbers, sitting idly by for the Clinton elections, just waiting for the Bush-43 elections of 2000 and 2004 to suddenly snap into action and change a predefined number of votes to other candidates so Bush could get elected. All this without having any idea what columns would be assigned to which party, whether there would be any ballot questions at the top of the machine.
Remember these machines use hard electrical switches, not touch screens. There is a 12x42 matrix of hard switches - 504 positions. The programmer decides which buttons mean what.
And yes, the very same hard wired ROMS were there in 1989, and in 2004. No one ever changed so much as one of them. Batteries, yes. Software, no.
As for stright party voting, yup we have it. But all it does is match whatever string is put into the straight party button to every candidate in that precinct linked to that party string in the database. "Does A equal B", not "does A equal 'Republican'". In fact, you don't even enter the party for a candidate as much as you choose it from the pull down menu of all parties you previously entered for that election. It just gives you back what you previously gave to it in the parties data table.
As for your exit poll problem, those who have rigorously examined it have debunked Steve Freeman. Notably I point you to www.mysterypollster.com, done by a Democratic pollster of high repute. He has lavishly explained the proven pro-Democratic bias in the exit polls. He didn't come to the realization lightly, but through scholarly examination. The biggest reason was a higher than anticipated Republican voter refusal to answer the exit polls. Why, you ask? Probably Limbaugh's influence. That last one is my speculation. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 117
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 12:50 pm: |
|
>>>You're wrong to characterize Kurt Bellman's remarks as neo-con type double-speak.<<<
Really? He wrote the following:>>>You really do need to get over this "Republicans stole the election" rant. It simply did not happen, sir<<<
He accuses me of ranting about something that virtually all of us on BBV not only believe, but collectively "proved" after the 2004 election. In Ohio's Cuyahoga County, some 27 precincts had over 93,000 more votes than registered voters. When you consider that Mr. Bush only won by 118,750 votes in that state, the most pivotal of all, any "reasonable" person would have to conceed there was massive vote fraud and that the election must have been stolen.
Further; the 3 states that used (difficult to rig) paper ballots all were within the +/-3% margin of error. Indeed, all were within 1.2% and one within .1% of the early exit poll results. However, some 27 other states (all with electronic voting or counting) were outside of the exit poll tallies by 4-17%, and all anomalies favored the Republicans.
So, do you still think I'm "wrong" Catherine? Of course you do.
Oh, and by the way, I noticed you didn't answer my query as to the lovely springs there in the Emerald Isle, Ireland from which you post. Are they as beautiful as the pictures I've seen, or have you been out lately to enjoy the countryside. I mean you've been so busy helping us across the pond here to sort out our problems. Let me hereby convey the undying gratitude of the American people for your selfless efforts/mac |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 24
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 1:02 pm: |
|
Catherine,
I'd like to provide more on the Danaher system's shortcomings but I'm hesitant to for one reason: the new HAVA-related update has finally led to a wholesale change in the software both in the machines, and the tabulation software, I'm led to believe. The machines needed to have support for "Ballot Image Retention" (the 1242-5T had it, the 1242-4T didn't) and support for the audio module for the vision impaired.
The Guardian Election Management system I used was version 2.something. I'm led to believe it's now 3.something. I hope they took the opportunity to fix the numerous bugs in the darned thing, among which:
1) Ballot formatting copied from one precinct to the next lost formatting, i.e. thickness of lines and boxes. Sometimes entire races would disappear that should have copied over from previous precinct. Had to manually re-do them.
2) Election night tabulations notoriously buggy and error-prone. That why we call them "Unofficial." I always advised everyone to take anything that came out of the election night reports with a truckload of salt.
3) Erroneous columning of candidates under correct party. Logic was right, alignment was wrong.
4) The logic of the office table was awful. On ES&S's Election Data Manager, an example of an "Office" is "Township Supervisor V:1, T:4" (Vote for One, 4 year Term). One then selects which parties were having a primary for that in which townships and VOILA! the links are automatically created for all those combinations.
Not in Danaher's GEMS. Each party and each township had to be individually created manually as a separate office, without Windows based copy and paste. ARRGH! That leads to an exponentially increased opportunity for errors. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 25
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 1:08 pm: |
|
mac,
"In Ohio's Cuyahoga County, some 27 precincts had over 93,000 more votes than registered voters."
That I'd love to see. I dispute the truth of that. Is that part of the final returns? And what does Wu say about it?
In my county if anything doesn't pass the possible test, it doesn't get certified.
Kurt |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2144
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 2:06 pm: |
|
Kurt,
I am aware of the extensive exit poll analysis done by Mystery Pollster and many others. These were later shown to have many flaws, and the hypothetical low rate of Republic responders was not supported by the data. Kathy Dopp and other statisticians have carried out further work (including integrating Elizabeth "Febble"'s work--sorry I can't remember her last name at the moment). The exit polls do support the theory that there were significant problems.
As Mac points out, there were many kinds of gross irregularities in Ohio and in many other states. Extraordinary barriers were sometimes put in place to prevent recounts, prohibit examination of the ballots, etc. States would just ignore their own statutes. (E.g., in Florida, key counties didn't respond to FOIA requests until after the deadline for challenging election results had passed, and even then only after litigation. When the documents were eventually provided there were hundreds of serious irregularities or malfunctions that came to light, enough that would have been sufficient to allow a recount had the information been provided in a timely manner.)
This is all well documented and easily confirmed by a number of non-partisan sources (VotersUnite.org, BBV, GAO Report, etc.). I am avoiding anything hypothetical, and I'm not speculating as to causes though fraud is certainly a possibility in at least some cases. I don't know if we'll ever know the full extent of the problems or the reasons behind them.
The big picture of the extent of the unexplained irregularities was persistently ignored by major media. This meant that people were likely to be aware of only the so-called "glitches" in their own state or county, without ever getting an overview of what was happening elsewhere. Indeed, some of the problems are only now coming to light.
Kurt, your voters were fortunate that they could count on you to do a smell test before certifying the results. Elections are the foundation of our democracy and they deserve to be a higher priority than they are in most counties' planning and budgeting process. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 26
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 2:07 pm: |
|
Okay mac,
I've got both the Cuyahoga County, OH registered voters report and the ballots cast report up on my PC as I write this. Which precincts are they? I'm looking, and I haven't found one yet.
Did you think I wouldn't check?
You may be able to convince other people with a claim. I need more evidence.
Kurt |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 27
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 2:20 pm: |
|
mac and Catherine,
Nope. I finished checking them all. No precincts in Cuyahoga, OH had more votes than registered voters. Not even one. None even stood out as unusually high or low turnout. They're all in the 65-70ish percent range, without actually calculating them.
How does this misinformation get spread? Catherine, mac, please don't damage the worthwhile work people like Bev are doing by continuing to repeat that which has been disproven by those without an axe to grind.
I'm not saying that there weren't problems, but I will not accept that the exit polls proved anything other than a sloppy sampling by VNS.
The specific claim made by mac about Cuyahoga (In Ohio's Cuyahoga County, some 27 precincts had over 93,000 more votes than registered voters.) is a bald-faced lie. I've proven it by going to the source documents available on the web.
Do the search again, Cath. It's all there to see. The lie is exposed.
Kurt |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2146
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 2:35 pm: |
|
I didn't saying anything in relation to the specific incident Mac mentioned. You're absolutely right, Kurt, that every specific incident must be backed up by solid facts. Mac's making a claim w/o a reference is not typical for this website.
That said, there were loads of problems in Cuyahoga County. Kathleen Wynne of BBV staff can address this most authoritatively, and I won't steal her thunder. She was there at the recount and has written extensively about the irregularities. You can also do a search of this website if you're good with search facilities. (I'll get some links if Kathleen doesn't get to it first--but I have an important project deadline so it might not be tonight.)
It is great that you are checking up on the facts using the actual records. That is absolutely critical for the credibility of anyone who is interested in election reform, and it is one of the strength of the research done by BBV.
Keep in mind that there's a difference between the formal research that BBV carries out, and an opinion expressed by a non-BBV-staff-member such as Mac or me or anyone else on a public forum. With time you figure out which posters do their homework. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2147
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 2:52 pm: |
|
http://www.bbvforums.org/cgi-bin/forums/show.cgi?tpc=72&post=20130#POST20130
"By the way, Ms. Ansbro.
It's so lovely to have you in on this conversation. Tell me, how is your native Ireland in Springtime? Is it just as beautiful as the pictures we've seen? It must be wonderful to be there this time of year/mac "
[Notice the deprecating tone of Mac's comment, and the disrespectfully snide use of "Ms. Ansbro". Notice also that he is interjecting something that is irrelevant, off-topic and inaccurate into this discussion. The inaccurate information was ignored by me.]
http://www.bbvforums.org/cgi-bin/forums/board-profile.cgi?action=rate&topic=72&p age=19804&post=20152
"So, do you still think I'm "wrong" Catherine? Of course you do."
[Notice that Mac is putting words in my mouth as to whether or not I agree with him about some of the exit poll results.]
"Oh, and by the way, I noticed you didn't answer my query as to the lovely springs there in the Emerald Isle, Ireland from which you post. Are they as beautiful as the pictures I've seen, or have you been out lately to enjoy the countryside. I mean you've been so busy helping us across the pond here to sort out our problems. Let me hereby convey the undying gratitude of the American people for your selfless efforts/mac"
[Again notice the snide comments, and referring to inaccurate information posted earlier. Another attempt is made to hijack the thread through something off-topic, and an attempt is made to detract from the significance of anything I post.]
Note also the disrespectful tone of some of his comments to Kurt Bellman.
Apologies Kurt. This kind of behavior is fortunately very rare on the BBV website. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 28
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 5:11 pm: |
|
Catherine,
I am more accepting of mac's tone than you might realize. I chalk it up to passion and anger about a perceived injustice. That fact that I don't share his perception is immaterial. His "pain" is legitimately felt, and needs to have a voice. I'm a big boy, with pretty broad shoulders.
I've been the victim of injustice, too. I don't handle it particularly gracefully always, either.
mac's okay with me, as is Brant. I actually like well thought out disagreement. And I'm a pretty forgiving sort. I can think these guys are wrong, without finding them evil. I do appreciate the apology in the spirit its offered, though.
One of the reasons I joined rather than lurking as I did for better than a year (yes, I lurked even while I was still an E.D.), was that I saw a general improvement in tone and intellectual rigor among the site's leadership. The ranters were being reeled in a little by peer pressure.
Many BBVers and I will disagree about details and specifics. It doesn't change my belief that on the "big picture", BBV is mostly on to something big here - that purely electronic voting is a VERY dangerous thing, as it is presently being done in most places. If everyone keeps their eye on that fact, we can get through the small stuff. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 29
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, April 19, 2006 - 5:22 pm: |
|
Oh, by the way..
You'll never see me defend stonewalling a recount, as long as the recount was legally called for.
In PA, that's a REALLY high hurdle. It's nearly impossible for a non-candidate to get a recount in PA, and it's fabulously expensive for anyone to get one. The fees and affidavit requirements are un-stinkin-believable. PA just instituted a 1/2% auto recount for statewide offices. But in local races, even a flat out tie vote doesn't get you a free recount. Incredible, isn't it?
When BBV came to me in October 2004, I initially blew you guys off. After I got over the panic mode in that election, I INVITED your Clean Up Crew guy that contacted me to be glued to me Election Night. He never showed. It was a shame. Except for the precincts that took til 1AM to finish, it was a textbook Election Night - the fewest problems I had in any of the 4 years I was there. I think it is because Presidential years are pretty easy in PA, because all the local offices are contested in odd-numbered years. That's where the problems explode. |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 510
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 5:02 am: |
|
So you do have a party affiliation tied to the names (through your table) so it would be possible to cheat along party lines. You've avoided saying this, but it in some way has to be true that you relate party affilation to the individuals running that are in those parties, since you can't know the party affiliation just from the name. You're avoiding this issue in a number of ways, but you're only avoiding the issue.
And you talk about election night tabulations being buggy and error prone on a DRE, yet you say that you're sure the machines don't cheat? If it doesn't produce reliable, consistent results, all the time, how could you possibly tell? On a DRE? Without a paper trail?
Your arguments to what I have stated aren't on point.
Specifically:
Whether firmware had bad functions that were called, or not called into play on a DRE would be very hard to prove or be sure of (there's a big pad of registered voters that don't vote, and if your swapping registered voters' votes you don't extras ). The arguments that Republicans weren't aswering the pollsters was disproven long enough ago that I've forgotten who disproved it, but I do remember that their evidence and arguments were credible. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 30
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 8:28 am: |
|
Brant,
There is a gaping security hole in all voting systems, if I understand your point. And that is ghost voting for voters who don't actually show up. I know that has happened in the distant past MANY MANY places in PA, and I SUSPECT it continues to this day. At least I know the charge is constantly being repeated that it's happening.
There are quite a few places where all the precinct election officials are of the same party. For my county that typically happens in the poorest sections of our county seat where the entire precinct board is all Democrats. There simply are no Republicans who will do it. PA law prohibits moving people into a precinct from another unless there is NO ONE to serve. So in a few of our most rural precincts there are solid Republican boards, and in many of our city precincts there are solid Democratic boards. So of course the "other party" always expects that there are shenanigans going on where they have no folks. But they don't have the resident folks to put in there.
You are missing the point on parties though. The parties are what I put in. If I put in "Republican" or "Rebublican" or "REP." or "GOP" or (Minnesota) "Independent Republican", those are the names the machine uses. I ran a bilingual operation. The system still works if I used the Spanish translations for the party names. How's the software going to look for that? It always kicks back what the E.D. puts in. There simply is no way for the code to examine for a particular hard coded string. It is a matching of a link, not a test for a particular label.
This is what infuriates me, if I may be frank, with the conspiracy theorists. They (note I didn't say 'you') assume that someone put code in there to look for "Republican" and jack up those numbers and tick down the other parties. I'm merely telling you that those people are off their rockers. It simply cannot be done on the system I knew and used.
Now could it be done on a machine that is an overdressed PC with an OS and disk drives and communication ports, etc.? Hell yes it can! That's the distinction I'm making. With a proper system it can't. Even Dr. David Dill realizes that. He doesn't like the Danaher for a variety of reasons, but even he admits they are orders of magnitude more secure than the Diebolds, ES&Ss, WinVotes, etc., because they aren't really even computers in the normal sense of the word. They are essentially electromechanical bucket counters - a very close analogue to the lever machines. There is no way for a lever machine to know what a "Republican" is, and there is no way for a Danaher to know it either. Its internal software merely activates the buttons that the database programmer has asked to be activated, and counts how many times each button is voted. That's all it knows. At the end of the day, it queries the cartridge that is installed in it to associate a race and name with those button numbers, and prints the results. That's all it does. They are far more analogous to an oversized calculator (all software on chips) than a computer in the normal sense of the word.
Once you convinced yourself that your calculator worked correctly, did you assume that someone snuck code in there to alter its calculations later? It just makes no sense, Brant. You don't understand the Danaher machine. That much is clear.
Now the tabulation software - that's another matter. That could be screwed with, admittedly. But there is a failsafe audit done every election. Each machine prints a paper tape at closing. You've seen them. The tabulation software had better match the totals of the 400+ individual machines, and during the official canvas that doublecheck is made. And what happens if the tabulation software says one thing and a manual tabulation from each machine says something else? The individual machine tally prevails.
My advice to the whole country is to NEVER take as final anything a DRE-based system says on Election Night. Just don't do it. Wait for the official canvas. If you live in a state where "concessions" mean something, don't concede, EVER! In PA, concessions are meaningless. A candidate who concedes believing he lost, and is later found to have won, still has won. |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 511
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 10:22 am: |
|
Mr. Bellman, follow these points closely:
1.My calculator is not a fitting analogy, my calculator isn't kept in public storage and used for something that someone else can make great gain from, nor is is working in an absence of significant testing, either. The analogy only fits to the extent that while my calculator has little intrinsic value, I've had 2 stolen from me.
2. Somewhere or other in your coding of the 'database tables' that you use on the Danaher, you have to have entered what is going to be posted to the machines' displays.
3. Data entered into any form of digital memory takes on the semantic meaning that it will ultimately have based on the context in which it is used, nothing more, nothing less. Matching could be done on the first character of the string, for instance. Your understanding of what use data may be put to appears limited by a dogmatic, perhaps paradigmatic understanding of the machine.
4. Unless you have taken this item and run it (this necessitates your knowing what a disassembling logic analyzer is) while monitoring it with a dissassembling logic analyzer, also having to hand the vendor's source code and having recompiled and assembled it yourself (to see that what you came up with as the ultimate result of this matches the binary image on the PROMs that they produced) you wouldn't know that the machines are running what you've been told they're running or not.
This prompts the following questions:
1. Have you ever seen the source code for these machines?
2. Has anybody that you have any personal experience with recompiled and reassembled that source code and compared its resultant binary image with what's on the chips from one of the machines?
3. Why do you believe that this machine is so reliable, if in fact as you said: "Election night tabulations notoriously buggy and error-prone.", if this is a hard-coded machine, how does the tabulation from a DRE get any better?
4. Any answer to 3 strikes me as impossible. Their code for tabulation is either good or it isn't, the accumulated data is either good or it isn't, where's the middle ground?
4. If this is a DRE and doesn't have a paper ballot produced, what does the canvas consist of? |
Mike Myhre
Voting Rights Forum Participant
Username: Mike_myhre
Post Number: 16
Registered: 02-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 10:40 am: |
|
Kurt,
Check out the GAO report...
Tech gives 'cheat sheet' for 'random' recount and takes hard disk from machine.!?!
Ohio doesn't match exit polls.
I don't believe it is a partisan issue, any party will have liars and cheats, but the evidence is so great that both the 2000 and 2004 elections were stolen by the Bush team that you can't ignore it or assume that it won't happen in the future. |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 512
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 11:07 am: |
|
Mr. Bellman, please follow these points closely:
1.My calculator is not a fitting analogy, my calculator isn't kept in public storage and used for something that someone else can make great gain from, nor is is working in an absence of significant testing, either. The analogy only fits to the extent that while my calculator has little intrinsic value, I've had 2 stolen from me.
2. Somewhere or other in your coding of the 'database tables' that you use on the Danaher, you have to have entered what is going to be posted to the machines' displays.
3. Data entered into any form of digital memory takes on the semantic meaning that it will ultimately have based on the context in which it is used, nothing more, nothing less. Matching could be done on the first character of the string, for instance. Your understanding of what use data may be put to appears limited by a dogmatic, perhaps paradigmatic understanding of the machine.
4. Unless you have taken this item and run it (this necessitates your knowing what a disassembling logic analyzer is) while monitoring it with a dissassembling logic analyzer, also having to hand the vendor's source code and having recompiled and assembled it yourself (to see that what you came up with as the ultimate result of this matches the binary image on the PROMs that they produced) you wouldn't know that the machines are running what you've been told they're running or not.
This prompts the following questions:
1. Have you ever seen the source code for these machines?
2. Has anybody that you have any personal experience with recompiled and reassembled that source code and compared its resultant binary image with what's on the chips from one of the machines?
3. Why do you believe that this machine is so reliable, if in fact as you said: "Election night tabulations notoriously buggy and error-prone.", if this is a hard-coded machine, how does the tabulation from a DRE get any better?
4. Any answer to 3 strikes me as impossible. Their code for tabulation is either good or it isn't, the accumulated data is either good or it isn't, where's the middle ground?
4. If this is a DRE and doesn't have a paper ballot produced, what does the canvas consist of? |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 513
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 11:21 am: |
|
Mr. Bellman, please follow these points closely:
1.My calculator is not a fitting analogy, my calculator isn't kept in public storage and used for something that someone else can make great gain from, nor is is working in an absence of significant testing, either. The analogy only fits to the extent that while my calculator has little intrinsic value, I've had 2 stolen from me.
2. Somewhere or other in your coding of the 'database tables' that you use on the Danaher, you have to have entered what is going to be posted to the machines' displays.
3. Data entered into any form of digital memory takes on the semantic meaning that it will ultimately have based on the context in which it is used, nothing more, nothing less. Matching could be done on the first character of the string, for instance. Your understanding of what use data may be put to appears limited by a dogmatic, perhaps paradigmatic understanding of the machine.
4. Unless you have taken this item and run it (this necessitates your knowing what a disassembling logic analyzer is) while monitoring it with a dissassembling logic analyzer, also having to hand the vendor's source code and having recompiled and assembled it yourself (to see that what you came up with as the ultimate result of this matches the binary image on the PROMs that they produced) you wouldn't know that the machines are running what you've been told they're running or not.
This prompts the following questions:
1. Have you ever seen the source code for these machines?
2. Has anybody that you have any personal experience with recompiled and reassembled that source code and compared its resultant binary image with what's on the chips from one of the machines?
3. Why do you believe that this machine is so reliable, if in fact as you said: "Election night tabulations notoriously buggy and error-prone.", if this is a hard-coded machine, how does the tabulation from a DRE get any better?
4. Any answer to 3 strikes me as impossible. Their code for tabulation is either good or it isn't, the accumulated data is either good or it isn't, where's the middle ground?
4. If this is a DRE and doesn't have a paper ballot produced, what does the canvas consist of? |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 118
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 12:12 pm: |
|
>>>This is what infuriates me, if I may be frank, with the conspiracy theorists. They (note I didn't say 'you') assume that someone put code in there to look for "Republican" and jack up those numbers and tick down the other parties. I'm merely telling you that those people are off their rockers. It simply cannot be done on the system I knew and used.<<<
There is no way you could "absolutly" know what could or could not be done with the "system you used", anymore than you could "absolutely" know whether money from your bank account is being stolen if your bank suddenly decided to stop issuing receipts and statements.
All knowledge is derived from a "reference", in this case "data" derived from the same "source", then run thru an alternate, independent system that elimnates the possiblity of human bias. Again, in this case there is only one such system presently in existance. It is the Point of Origin Poll, or POP.
While the POP cannot presently be utilized in your state due to statutorial constraints, it can in other states including California. If only one state allowed the POP and it uncovered vote fraud, the consequence would be the nullification of that election and subsequently the rewriting of all law that prevents alternate vote checks thru POP/mac |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 33
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 12:20 pm: |
|
Brant,
Re: your questions
1)No, the source code would mean nothing to me. I am concerned only that it do what it was intended to do. Every machine is tested prior to each election, in public (theoretically - no one has ever shown up to observe the testing in recent years although the notice is publicized), and each machine WITH ITS CURRENT ELECTION CARTRIDGE is tested with a test election script of predetermined voting sequences (not the same one each time, but a different one each election) and the results paper tape is examined for correctness. The results of that test are saved for EACH MACHINE in EACH ELECTION until all opportuinities for a challenge have past. That test includes a test to see that straight party votes are properly allocated to the respective candidates. These tests take a team of 6 or so testers several days to conduct, under the supervision of the county itself.
2) No, other than the proper testing authorities. Recall, these machines are NOT new. They have been in service almost 20 years, and have never given any cause for suspicion. In addition, when they were put into service, no change in voting patterns was observed from the predecessor system - levers.
3) The system has two major components. A) the individual voting machines, and B) the central tabulating/database creation software on a PC. I trust part A) greatly. I'm not so sold on B). I've seen enough weird stuff come out of B) to be leery of it, so I didn't ever trust it without personal parallel creation manually (actually Excel speadsheets made from the paper tapes) of the tabulations. Interestingly enough, while the election night unofficial module had errors a few times, the final official tabulation module never did. I don't really know what to make of that. Remember, it was that same election night module that burped 3000+ phantom votes in Franklin County, OH. I never saw anything that bad, but I saw enough to not trust it fully.
4) This isn't really that tough. You know that the parts are good, but the assembly of the whole is problematic, so you check every bolt and screw manually. Sorry for the mechanical analogy. Is the point clear? Their tabulation module (not code) does suspicious stuff. The data from each of the machines is good, but sometimes the election night module reports the wrong totals. Happened 5 times out of 8 elections to me. However, when the detailed final reporting module was run, the totals were correct. I don't know how that happens either. I had elections that when I ran the Summary Report and the Detail Report on the same election, the numbers were different, and the Detail Report ALWAYS matched the manually tallied results from each machine.
5) The canvas consists of four things. Rerunning the same cartridges into a new copy of the election database, as if election night never happened. Just redo the whole deal. Next you manually add in the absentee ballots, and the provisional ballots that have been accepted (all provisional and absentee ballots are paper ballots hand counted by BOE staff AFTER the election), examine each machine's paper tape (totals printed by each machine when the "CLOSE POLLS" button is hit) and crosscheck that paper to be sure that the totals on it match those in the election database for each machine, and then (I did this even though it was not required) enter those results from the tape into Excel spreadsheets for crosschecking the tabulation software. (These Excel spreadsheets became quite useful in reporting results in the proper formats to the the SoS office. The Danaher software did not export data in a usable format for the SoS.)
The remainder addresses misconceptions in your paragraphs above the questions.
1) My machines were kept not in "public storage" but in a county owned warehouse where ONLY the County Director of Facilities had the key. Even I had to get the key from him to deliver materials to the warehouse. These machines are so robust they do not even require controlled environmental conditions, unlike most DRE's. They have no disk drives, no modems, no networking, no wires of any kind to the outside other than a power cord. There is no precinct "zero-machine", no "hub", no "card activator", none of the garbage that other systems use. All results are transmitted by driving a cartridge (proprietary hardware format) and paper tape to election HQ after the polls close.
2) The machines have no "displays", (nor do they have touchscreens). All information presented to the voter is PRINTED on a paper overlay sheet affixed to the front of the machine. It is printed on a stock slightly translucent so that the round indicator LED's beneath it can be seen as lighting up when the corresponding elctromechanical switch (1 of 504 in a 12x42 array) corresponding to a candidate is chosen. Now before you have an AHA moment and suggest that what is on the sheet might be different than how the machine is programmed, you would be CORRECT if you were to make that point, except for this. Each ballot sheet, each machine, and each database programmed cartridge have a serial number, and part of the startup procedure is to be sure that all serial numbers match. If they do not, the machine may not be used. The ballot layout software forces the making of new ballot sheets and new cartridges if ANY change is made to the database after materials are produced.
3) You really don't understand the party issue. Parties are completely freeform string variables. I could use "Dave's Party" or "Kiss Me Party" or anything, and the software slavishly accepts it. You can run a Danaher with 6-10 parties that use similar names, if the law allowed it. The machine just doesn't know anything about parties, except that they exist and must have at least one charcter different from each other (otherwise they're the same). In fact, if I had "Republican", "REPUBLICAN", and "repubican", those would be three different parties on the system. In some years I used REPUBLICAN, and in others I used Republican, but never both, but nothing in the machine would have prevented it.
4) Again, I don't need the source code. I need to know that when I run a test election, no matter how complex, on a machine, that it gives the accurate result when I close the polls. And that has been done to my satisfaction. Now before you say that it could be programmed to do different things at different times, "How?" It has no internal clock. It has no internal calendar. There is no time stamp for anything. The date of the election is enetered as one of the first variables when the database is being programmed. It is entered by hand by the E.D. No Danaher ever knows what date or time it is. The tape in the machine just prints POLLS OPEN, not POLLS OPEN at 6:59AM 11/2/04. Nothing in the system would have prevented anyone from running the election on a machine at any time before or after the actual day.
You see, the Danaher really is a pretty "dumb" system. That is paradoxically its strength. |
mac Sperry
Frequent Voting Rights Forum Participant
Username: Littlemac
Post Number: 119
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 1:24 pm: |
|
Possibly Mr. Bellman you could elaborate on your being "fired" by the elections board less than 3 weeks before the last primary election raised in the following article from: http://www.readingeagle.com/blog/editorials/archives/2005/05/investigation_n.htm l
...
May 29, 2005
Investigation needed in election programs
The Issue: Because of errors in programming, 463 people in four city voting precincts will have to participate in a special election. Several races hang in the balance.
Our Opinion: Although no one is suggesting fraud, District Attorney Mark C. Baldwin should investigate the snafu and recommend how similar problems can be avoided.
With several primary election contests still undecided because of a mistake in the programming of some voting machines, it is time for the Berks County commissioners to answer a couple of questions.
Was this a direct result of the firing of V. Kurt Bellman?
How much is the programming glitch and the resulting special election going to cost the Berks County taxpayers?
Less than three weeks before the primary voting, the commissioners — at least the two Democratic commis-sioners, Judith L. Schwank and Thomas W. Gajewski Sr. — fired Bellman, elections director, for poor management.
Bellman, a Republican, claimed his dismissal was politically motivated. He said Schwank and Gajewski had a hit list that included many of the GOP-appointed officials working for the county.
Schwank and Gajewski dismissed the allegation, saying instead Bellman was behind in a lot of his work, including applying for state grants to upgrade elections equipment and train poll workers.
Whatever the motive behind the dismissal, subsequent events certainly suggest that the timing certainly wasn’t the best.
After Bellman was fired, Schwank and other election officials called in officials from Danaher Controls, a voting-machine software company from Gurnee, Ill., to com-plete the setup for the primary election.
When Danaher officials examined the work Bellman had done, they said the entire database for the election had to be rebuilt from scratch. It was four days before the election.
In the rush to get the work done, Danaher employees programmed 38 of the more than 380 data cartridges that record the votes as training cartridges, which reset voting totals to zero when the machines are opened.
We certainly are not questioning the analysis Danaher did of Bellman’s work, but during the four years Bellman was director of elections, no votes were lost.
We cannot help but wonder whether it would have been prudent to keep Bellman on the job until after the primary voting had been completed.
On election night, when the problems with the voting machines were coming to light, Schwank mused about how much worse things could have been had Bellman still been on the job.
The other side of that question is just as viable: Would this election have gone smoothly — as had the others Bellman supervised — had he not been fired.
Schwank has been asked how much this fiasco is going to cost the county, but all she said is that the expense will be negotiated. That is not good enough.
City Councilwoman Donna Reed, whose renomination in the 5th Council District remains in doubt because of the snafu, has called for an independent investigation into the programming glitch.
In light of some of the election irregularities that have taken place nationwide, Reed’s call for an independent investigation sounds like a good idea. And District Attorney Mark C. Baldwin would be an excellent person to conduct such an investigation.
Considering the complexity of the voting machines and their programming, it is somewhat surprising that these types of mistakes don’t occur more frequently. But when they do, it is important for the electorate to know that the error was inadvertent and not an attempt to tamper with the election.
We’re not suggesting that anyone attempted to do that, but an independent investigation would short-circuit any claims of impropriety that might arise and provide rec-ommendations to avoid a similar mistake.
Posted by readingeagle at May 29, 2005 01:00 AM
Comments
The big question seems to be...since Bellman did the programing, could he have deliberately caused the snafu to get even for his firing that was in the offing??...It's very likely he knew something was brewing.
I think a thorough investigation should be done by an independent source. Baldwin is too close to the situation.
Posted by: I'm Ms222 at May 26, 2005 09:20 PM
Why was he the only person in the county who knew how to program these machines? What if he hadn't been fired but had been kiled in a car accident or something? Why is the actual voting process so complicated that we need people with college degrees to conduct elections? Can we please go back to paper ballots counted by hand?
Posted by: Jim Keller at May 31, 2005 09:47 AM
The concerned voting citizens of Pennsylvania have submitted a complete petition to the Secretary of State of Pennsylvania to have the Danaher voting system used in Berks County re-examined.
Posted by: The American Voter at June 19, 2005
(Message edited by littlemac on April 20, 2006) |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 37
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 1:49 pm: |
|
Sure mac,
What elaboration would you like?
Kurt |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 38
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 2:25 pm: |
|
mac,
I'll tell you this. The decision to have only one person who was involved with programming the machines (me) was the Commissioners'. They would not add staff despite repeated requests from me to do so. The county was in a fiscal crisis wherein departments were being cut back, not expanded.
I asked for the proverbial "second pair of eyes" for database programming and was denied.
When I was fired, the 2005 primary was 98% programmed. Only the last few single precinct offices on the ballot remained, and the production of ballots and cartridges. Sample ballots were already being printed for precincts without the last few offices.
The Commissioners lost two weeks going to the wrong vendor until they contacted Danaher. The first vendor (not Danaher) thought they could finish my programming. I suspect, but have no way to know, that during those two weeks the database may have gotten corrupted. The May 2005 database was built by me from a stripped down copy of the May 2003 one, since its basic structure (number and types of offices) was similar. The basic ballot layout was identical.
When I was fired, I begged the County Administrator to allow me to finish the election because I knew how complicated the 2005 database was (>900 candidates). He refused, saying they didn't think they'd have a problem. Well, they did.
What is accurate is the fact that I was behind on administrative work, and on filing my county HAVA plan. My assistant, who incidentally was fired later because he would not cooperate in blaming me for the election debacle, was threatened at the time if he even spoke to me for help in setting up the election.
The Danaher software is EXTREMELY cumbersome and difficult to set up in an election where every precinct has a unique ballot. I told the Commissioners and the Administrator that. They refused to listen, and they got what happened.
Danaher put all their people on redoing the database from scratch. They apparently worked 24 hours a day. When they went to make cartridges, they programmed 38 of them as "test cartridges" rather than "production cartridges". Several of those "test cartridges" lost votes.
The difference is this. In a production cartridge, if the machine is turned off during the day and rebooted, the cartridge retains all votes and picks up where it left off. All votes are retained. A "test cartridge" is purposely designed so that when the machine is turned off, the machine resets everything to zero, and all votes are discarded. The "test cartridge" would be better called a "training cartride" because its main use is in pollworker classes, that need the ability to repeatedly demonstrate the whole cycle of an election.
Well, on the May primary 4 of those 38 machines with test cartridges were rebooted at some point during the day. One was rebooted at the end of the day, and all votes on it were lost. The revote never occurred in the end. Those votes ended up lost forever.
If you finish your research, mac, you'll find that Danaher admitted they mistakenly programmed the test cartridges. I had not created any cartridges yet for the 2005 primary. At least one key Danaher staffer was fired for their role.
As for Commissioner Schwank and Gajewski's comments, they are at considerable risk for a lawsuit and have since gotten real quiet. Oh, and you'll love this part. Who replaced me? The same woman I replaced, who in the interim sold the Advanced Voting Solutions WinVote and later Diebolds all over the country. Kinda makes you all warm and fuzzy, doesn't it? |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 40
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 3:15 pm: |
|
mac,
One other elaboration, The District Attorney did indeed do his investigation. I was questioned. I was exonerated of all involvement in the snafu. In fact, the D.A.'s report included the suggestions for avoiding a recurrence that I provided them.
The best thing that came of it is the changes that Danaher put in the software that allegedly prevent a training cartridge from being used.
I say allegedly because I haven't seen the new version. It gets its shakedown here in my county on May 16th, run by my successor. I wish her luck. She may need it. The first version of anything new is a big risk. |
Mike Myhre
Voting Rights Forum Participant
Username: Mike_myhre
Post Number: 17
Registered: 02-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 3:54 pm: |
|
Kurt,
Have you considered the fact that someone could have intentionally placed the 'test' cartriges in machines they wanted to lose morning votes on? |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 42
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 4:09 pm: |
|
Mike,
I thought about it a lot, yes. But I dismissed it because the 38 machines were in numerical order by precinct number. They were the last 38 machines in the City of Reading, comprising the last 19 precincts in Reading. That was the reasonable sized batch of carts that would be "burned" in one batch, and in a logical order.
The Wards involved were 15, 16, 17, 18, & 19, Not all wards have the same number of precincts.
15 is the near northwest, 16 the extreme southeast, 17 the far northeast, 18 across the river to the south, and 19 across the river to the northwest. They have little in common with each other. If there was a method to it, it escapes me. And I grew up in downtown. I know its politics well.
If I had to come up with a pattern it would be that compared to the rest of the city, the affected wards are, on average, more affluent than the rest of the city, but not by a lot.
Good thinking, though, Mike. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 43
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 4:18 pm: |
|
BTW Mike,
I DO think that 2000 was "stolen", but not the way you probably think.
The 2000 election would have come out completely differently if the faulty Florida felon purge hadn't happened, even if all else had gone down as it did (butterfly ballot, chads, the woiks).
I don't personally believe 2004 was stolen, unless one considers that Bush wouldn't have even been on the ballot had his "loss" in 2000 actually happened. |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 9
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 8:25 pm: |
|
Ref. posts April 12, 10:39PM and April 13, 4:58AM
>A phone line is not output-only. No ports that can implement error checking or retransmission are output only. You need some form of signal going to the sender to indicate what he's sent doesn't check out and must be retransmitted.<
I have tried to contact experts to verify your contention that data cannot be "output only". I've been told "Yes, "no" and maybe. Are there any geeks out there that want to weigh in? One said how about using a "live CD" where the total program is on read only memory to prevent hacking?
To me it's a moot point, If there is no such thing as "output only" lines, then you need to add another computer at the voting place and swap floppy disk files between booth computers. |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 10
Registered: 03-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, April 20, 2006 - 8:30 pm: |
|
Ref. post April 18, 2006 - 05:03 am
>As you have printed the ballots 'randomly' on the roll, there isn't any immediacy to seeing an error, you have to go back and count every vote on the roll to find the problem, and the evidence is in the hands of the elections workers and subsequently the Board of Elections, not the common voting citizens.<
Ballots would not be randomly placed on the entire roll. An arbitrary sized group of approximatly say 40 contiguous ballot spaces on the continuous-form is completely randomly filled and the subtotal and total is calculated by the computer. The ballots, subtotals, and totals find their way to the internet so that the "trusted people" (judges) can compare the internet on their own laptops with the continuous-form ballot at the voting place. When verified they signal the central office where the internet is controlled and the internet copy is changed from "not yet verified" to "verified". This is the point in time, roughly an hour after voting, where a voter can verify that his vote has not changed from the time of voting where he verified his vote on the continuous-form ballot. The ballots, totals, and subtotals have been verified and bundled for easy recounts if necessary. If any ballot is changed on the official internet copy causing the subtotals to change by even one vote, there is no reason to panic as the previous ( verified ) internet copy takes precedence. My website(see first post) explains the bundling process completely. |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 11
Registered: 03-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Thursday, April 20, 2006 - 8:35 pm: |
|
Ref. post April 18, 9:23
>You're never going to see running subtotals during the day either. The future trend is to delay results until later. There is no apetite for partial counts or early counts<
The total process described is not to satisfy peoples apetite for early counts, but to create a duplicate of the official record, the continuous-form plural cast ballot, on the internet. This can be done within roughly an hour of each person actually voting. There is no reason for this internet record to ever change. The "trusted people(judges, from every major political party) at the voting place have verified that the continuous-form printout is the same as the internet as read on their laptops. Now if a counterfeit continouous form ballot is presented, it is suspect rather than the verified internet copy.
If it's a crime in some states, for instance PA, to insert the internet, during election there, so be it, then the internet will have to be dispensed with there and the "official record" will need a few more layers of guarding |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 514
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Friday, April 21, 2006 - 10:05 am: |
|
My points are these:
First, you can't easily check the votes on the roles against cast votes in a way that is very timely. Unless the voter is allowed to view that role as his vote is being cast (and before it's accepted) what's going to keep this verified?
Second, if your void printing procedure isn't done in a way that is exclusive of the presentation of a valid ballot printing, how can the voter be sure that his ballot won't be voided and re-voted? I don't think that he can. Since you're trying to keep the vote untraceable, the only time you can verify a 'block' of votes is after the 'block' has been written to the paper, these leaves a significant margin of time for it to be adulterated, if the adulteration is done by the voting system.
The bit about ports is this, and you can confirm it from common sense: You're writing a program for a PC that sends to another device, it's a significant amount of(or valuable) datat. After it's sent, you want some form of confirmation that it was received at all/received correctly. Either data needs to be exchanged (validating the transfer by checksums, etc.(preferable) or a signal coming back to the PC that indicates the device 'got it'. If there aren't any lines that are input to the PC, you neither have a way of knowing that the other device is connected or taking the data, or validating its correctness. See?
As to string variables and their testability, Mr. Bellman: I have programmed in assemblers, various incarnations of C and unless you write executable programming, then I probably know a great deal more about variable comparison in it's nuts and bolts than you do. Resident functions in any reasonable decent compiler that has strings involved in it is easily promoting all characters in a string to upper or lower case. If not, they are easily written in a few lines of code. This fact renders your argument for string irregularity as a protection as very weak.
As to your not argument about 'not open to the public' you've just indicated a very powerful possible 'single point of failure'.
Also Mr. Bellman, since your machine nows nothing about time, what's to stop someone from putting in a 'production cartridge' putting on several votes and putting the cartridge back in the stack? It seems you haven't thought of that. But then we all have assumptions that make some things hard to see, don't we? |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 515
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Friday, April 21, 2006 - 10:08 am: |
|
My points are these:
First, you can't easily check the votes on the roles against cast votes in a way that is very timely. Unless the voter is allowed to view that role as his vote is being cast (and before it's accepted) what's going to keep this verified?
Second, if your void printing procedure isn't done in a way that is exclusive of the presentation of a valid ballot printing, how can the voter be sure that his ballot won't be voided and re-voted? I don't think that he can. Since you're trying to keep the vote untraceable, the only time you can verify a 'block' of votes is after the 'block' has been written to the paper, these leaves a significant margin of time for it to be adulterated, if the adulteration is done by the voting system.
The bit about ports is this, and you can confirm it from common sense: You're writing a program for a PC that sends to another device, it's a significant amount of(or valuable) datat. After it's sent, you want some form of confirmation that it was received at all/received correctly. Either data needs to be exchanged (validating the transfer by checksums, etc.(preferable) or a signal coming back to the PC that indicates the device 'got it'. If there aren't any lines that are input to the PC, you neither have a way of knowing that the other device is connected or taking the data, or validating its correctness. See?
As to string variables and their testability, Mr. Bellman: I have programmed in assemblers, various incarnations of C and unless you write executable programming, then I probably know a great deal more about variable comparison in it's nuts and bolts than you do. Resident functions in any reasonable decent compiler that has strings involved in it is easily promoting all characters in a string to upper or lower case. If not, they are easily written in a few lines of code. This fact renders your argument for string irregularity as a protection as very weak.
As to your argument about 'not open to the public' you've just indicated a very powerful possible 'single point of failure'.
Also Mr. Bellman, since your machine knows nothing about time, what's to stop someone from putting in a 'production cartridge' putting on several votes and putting the cartridge back in the stack? It seems you haven't thought of that. But then we all have assumptions that make some things hard to see, don't we? |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 55
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Friday, April 21, 2006 - 10:43 am: |
|
Brant,
"what's to stop someone from putting in a 'production cartridge' putting on several votes and putting the cartridge back in the stack?"
Great question. If people fail to do their jobs correctly, nothing. If, however, they make sure that the opening count is zero, as they are required to do and sign on it, and be sure that as each voter votes the public counter increments, as they are supposed to do, and see that the numbered list of voters contains the same number of voters as the total votes cast on the machine, as they're supposed to do, then all will be fine.
Heck, if people fail to do what they're supposed to be doing, the whole system breaks down, regardless of technology.
As for your point on 'not open to the public', I missed it. I'm guessing it has to do with storage of the machines. Correct me if I'm wrong. You're right. While the machines are in storage, no one has access. Even I did not. While they are being prepped, the public theoretically has access. Theoretically, because in my four years, no one ever once availed themselves of that access.
That is the fly in the ointment of the "public oversight" argument, as I see it. I've never witnessed any desire of the public to do any oversight. Now that may soon be changing rapidly. And that would be great. No one I know in elections would stonewall that, except out of ignorance. The amount of ignorance on election law among election administrators is quite disturbing, especially in "backwater" areas. |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 56
Registered: 04-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Friday, April 21, 2006 - 10:48 am: |
|
Re: inorance of election law among those who should know better.
It ain't what they don't know that scares ya', it's what they know for sure that just ain't so.
I wish that old saw weren't applicable, but it unfortunately is. Many can't put two words together without asking their solicitor. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 2153
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Friday, April 21, 2006 - 11:16 am: |
|
And knowing for sure is really bad re: HAVA. Lots of election officials believe that HAVA requires them to switch to electronic machines, even though it doesn't. (It requires some form of accessible voting for some disabled voters but not all, and this doesn't require electronics.) |
V. Kurt Bellman
Voting Rights Forum Participant
Username: Formerelecdir
Post Number: 58
Registered: 04-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Friday, April 21, 2006 - 11:37 am: |
|
Catherine:
Of course, HAVA doesn't exist in a vacuum. It is supplemented here by state enabling legislation and constrained by the content of each state's HAVA State Plan.
In SOME states, HAVA compliance does mean electronics, because the alternatives aren't certified. Even in some places where alternatives are certified, the more radical accessibilty organizations are mobilizing against anything but DRE's. I know that in PA, several areas that were talking about the AutoMark were threatened with lawsuits by the radical accessibility folks. And those folks have already demonstrated their litigiousness. The VotePAD failed certs here.
But in the abstract, Catherine, you are dead on correct. |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 12
Registered: 03-2006
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Friday, April 21, 2006 - 5:38 pm: |
|
Brant
>My points are these:
First, you can't easily check the votes on the roles against cast votes in a way that is very timely. Unless the voter is allowed to view that role as his vote is being cast (and before it's accepted) what's going to keep this verified? <
The voter is allowed to view printed ballot after making selections. The printer has moved the continuous form ballot to a random position for voting(one of say 40 positions). If voter accepts printed (conditional) ballot, the word "Accepted" is added to the ballot. If voter rejects, the word "Rejected" is added. The voter sees whichever word( Accepted or Rejected) is being printed. If "Reject" is the selection, voter gets to vote again.
>Second, if your void printing procedure isn't done in a way that is exclusive of the presentation<
The printing is done in view of voter
>Since you're trying to keep the vote untraceable<
A number known only to the voter is printed on the ballot. This identification is no more subject to vote selling than the present absentee voting option.
> the only time you can verify a 'block' of votes is after the 'block' has been written to the paper, these leaves a significant margin of time for it to be adulterated<
The individual ballots are written in view of the voter. After the last voter views and accepts ballot, the whole block is complete.
>The bit about ports is this, and you can confirm it from common sense: You're writing a program for a PC that sends to another device, it's a significant amount of(or valuable) datat. After it's sent, you want some form of confirmation that it was received<
no I don"t
> Either data needs to be exchanged (validating the transfer by checksums, etc.(preferable) or a signal coming back to the PC that indicates the device 'got it'. If there aren't any lines that are input to the PC, you neither have a way of knowing that the other device is connected or taking the data, or validating its correctness. See? <
I'm only interested in somehow getting information to the internet. Once the subtotals and ballots appear on the internet, The "trusted group" (judges) determines, from their laptops, if it concurs with the printed continuous form ballot. They don't "verify" the internet conditional pural cast ballot the until the transfer "gets it right". After it does, the internet presentation is changed from "not yet verified" to "verified". |
Brant Lamb
Frequent Voting Rights Forum Participant
Username: Brantl
Post Number: 516
Registered: 01-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, April 24, 2006 - 6:39 am: |
|
Let me see if I've gotten this right; you don't care whether transmission of the ballots was received correctly? Because if you have no way for the receiving device to indicate that its received them to the sending device, you have no way for the sender to be certain that what was sent was even received, thereby allowing ballots to be put into what we in the hardware business call the 'bit bucket' (irrevocably lost).
By assuming this kind of system, you are making the paper roll the only reliable evidence of voting, because nothing transmitted is safe from corruption or having any surety that all communications were received (in your system). |
robert nordberg
Voting Rights Forum Participant
Username: Bobn
Post Number: 13
Registered: 03-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Monday, April 24, 2006 - 7:45 pm: |
|
Here is the process. An official record of an election is created on a continuous form printer. The printing of each ballot is randomly placed in group of contiguous ballot spaces and observed and verified by each voter while voting. After an arbitrary group of ballots, say 40, is printed out on the continuous form, the next ballot space is used to sumarize the subtotals and totals. This means that the group of ballots together with the summary are naturally bundled together in a group for easy identifition by election judges during the election day. The official total count never resides in computer memory to be contaminated, but resides on paper, the storage medium of choice.
If any particular voting commission chooses to use the internet to make exact copies of the continuous-form ballot then we can attain additional security and also allow voters to verify that their votes were counted by including a unique number on each ballot as it is being printed.
How can we accomplish making exact copies on the internet?
I am not going to allow any input lines to the computer controlling the continuous-form ballot in order to prevent hacking. If "output only" lines don't exist, then transfer the group subtotals and totals via floppy disks to another computer. Let's say somehow a transfer takes place to the internet and that it happens to be corrupted. That's O.K. because it has a header saying "NOT YET VERIFIED". The judges at the polling place send a message to the people controlling the internet that the internet as viewed on their laptops is incorrect. The booth computer controlling the continuous-form ballot continues to output the official record on a repetitive basis, and another attempt to get it right on the internet is made. This is repeated until finally an attempt to get the file to the internet is successful and the judges agree that the internet copy is correct. They then inform the people controlling the internet that the latest attempt was successful, and the header on the latest internet file is changed from "NOT YET VERIFIED" to "VERIFIED".
In effect, the verification of the files on the internet is accomplished by the judges at the polling place, rather than a file redundancy check by the communication computers.
You refer the the official record as a paper roll, I call it a paper stack (of folded sheets) as it is more easily handled by people. |
|
|
