Wired News: More Diebold Machine Flaws
- Thread starter Ladewig
- Start date
Well, the ATM you're going to hit this evening in preparation for your weekend may well be a Diebold machine; you trust that, I hope?Ladewig said:
That having been said, I don't understand why they can't use the optical scanning methods that have worked so well for the SAT exams over the years. I don't recall ever hearing anyone challenging the accuracy of scoring those "fill in the circles" sheets (I'm talking about the scoring process itself, not challenges to whether certain answers intended to be counted as incorrect should have been counted as correct).
My voting district works exactly that way. You fill in the circles you want on the paper ballot, insert the ballot into a folder that blocks from prying eyes who you voted for, then bring it to a ballot box that sucks the paper ballot out of the binder, scans it, and records your votes, keeping the paper ballot for possible recounts if necessary. I believe the ballot box machine will reject your ballot if you screw up a vote, like voting for two presidential candidates.
What problem can anyone have with that system?
Rob Lister
Unregistered
- Joined
- Apr 1, 2004
- Messages
- 8,504
Re: Re: Wired News: More Diebold Machine Flaws
It's too accurate.
BPSCG said:
It's too accurate.
Ladewig
I lost an avatar bet.
- Joined
- Dec 4, 2001
- Messages
- 28,828
Re: Re: Wired News: More Diebold Machine Flaws
False comparison. I've used MicroSoft software that was reliable and secure, but that does not mean all MicroSoft programs are reliable and secure. And as Thompson pointed out, a system is only as secure as the people allowed to access it. If a Diebold employee is called in to fix a machine on election day, the average polling worker is not computer literate enough to stop that employee from switching votes in such a way that cannot be detected.
The people criticisizing the security are not conspiracy kooks off the street.
--Herbert Thompson, director of security technology at Security Innovation and a teacher of computer security at the Florida Institute of Technology. Thompson has authored several nonfiction books on computer security.
--David Jefferson, a computer scientist at Lawrence Livermore National Laboratory and a member of the California secretary of state's voting systems panel
BPSCG said:
False comparison. I've used MicroSoft software that was reliable and secure, but that does not mean all MicroSoft programs are reliable and secure. And as Thompson pointed out, a system is only as secure as the people allowed to access it. If a Diebold employee is called in to fix a machine on election day, the average polling worker is not computer literate enough to stop that employee from switching votes in such a way that cannot be detected.
The people criticisizing the security are not conspiracy kooks off the street.
--Herbert Thompson, director of security technology at Security Innovation and a teacher of computer security at the Florida Institute of Technology. Thompson has authored several nonfiction books on computer security.
--David Jefferson, a computer scientist at Lawrence Livermore National Laboratory and a member of the California secretary of state's voting systems panel
A touch screen machine should be virtually 100% accurate. ATMs do not make mistakes.
Any voting method that relies on paper in anyway will not be as accurate. Erasures, dust and light writing on optical scans will cause problems.
The idea of a paper trail is nice if it is used as an audit method or in case of some unexpected failure in an individual machine - e.g. power failure. But many people want to use the paper for manual recounts which is a joke. It replaces an accurate first count with an unreliable count due to human error (deliberate or not.)
I realize that we are in a transitional period towards electronic voting and that may cause problems but it is not difficult to get to an almost perfect system:
1) Touch screen voting.
2) All voting machines need their software verified by NIST.
3) All voting machine software should be available to the public.
4) Random audits of the mechanical performance of machines should be done before and after elections.
5) Their should be a paper trail for random audits.
6) There should be strict rules laid out that specify action in case of any failure e.g. What do we do when there is a power failure? What do we do if an audit finds an error? How do we count questionable ballots?
6) Absentee ballots should be minimize because they are inherently inaccurate and fraud prone.
CBL
Any voting method that relies on paper in anyway will not be as accurate. Erasures, dust and light writing on optical scans will cause problems.
The idea of a paper trail is nice if it is used as an audit method or in case of some unexpected failure in an individual machine - e.g. power failure. But many people want to use the paper for manual recounts which is a joke. It replaces an accurate first count with an unreliable count due to human error (deliberate or not.)
I realize that we are in a transitional period towards electronic voting and that may cause problems but it is not difficult to get to an almost perfect system:
1) Touch screen voting.
2) All voting machines need their software verified by NIST.
3) All voting machine software should be available to the public.
4) Random audits of the mechanical performance of machines should be done before and after elections.
5) Their should be a paper trail for random audits.
6) There should be strict rules laid out that specify action in case of any failure e.g. What do we do when there is a power failure? What do we do if an audit finds an error? How do we count questionable ballots?
6) Absentee ballots should be minimize because they are inherently inaccurate and fraud prone.
CBL
Are we confusing apples and oranges? To say that an electronic ATM-style voting machine is potentially open to tampering is not the same thing as saying it is inaccurate. So putting fraud aside for a minute, how accurate is a touchscreen voting machine compared to an optical reader machine?
Next, the accuracy of a machine does not tell you the user-friendliness of it. So while an optical reader machine may read a pencilled ballot very accurately, there is no way to tell if the voter understood the procedure and voted the way they intended. Whereas with a touchscreen machine, the screens may be such that even an elderly Floridian can understand it, especially if it is large font capable. You can even put a freaking photo of the candidate if you so desire, so you can eliminate the need for a voter to even be literate.
I think touchscreens are the way to go.
Next, the accuracy of a machine does not tell you the user-friendliness of it. So while an optical reader machine may read a pencilled ballot very accurately, there is no way to tell if the voter understood the procedure and voted the way they intended. Whereas with a touchscreen machine, the screens may be such that even an elderly Floridian can understand it, especially if it is large font capable. You can even put a freaking photo of the candidate if you so desire, so you can eliminate the need for a voter to even be literate.
I think touchscreens are the way to go.
Rob Lister
Unregistered
- Joined
- Apr 1, 2004
- Messages
- 8,504
Luke T. said:
Putting fraud aside makes the whole debate moot. The potential for fraud, pro and con, is what drives this debate.
Put yourself in the shoes of a very powerful but more than slightly crooked politician (as if there were any other kind). What system would you prefer if you were slightly ahead in the polls? If slightly behind?
If slightly behind, I would prefer the punch cards. If slightly ahead the computer.
So the computer is the better system?
My opponent thinks so because the company that maintains them is a big supporter of his.
My mother has been on the League of Women Voter committee to recommend voting machines. They try out many competing machines with real voters and make recommendations. I know one of the big issues is what the voters are comfortable with - e.g. touch screens vs. optical readers.
I do not know which one the League liked. I know there was one that enabled the election officer to see the votes made. Obviously that one was unpopular.
I should note that I spent 30 minutes with her arguing about manual recounts. She insists it is the way to go even though it is inaccurate. People prefer what they have had to superior new stuff.
CBL
I do not know which one the League liked. I know there was one that enabled the election officer to see the votes made. Obviously that one was unpopular.
I should note that I spent 30 minutes with her arguing about manual recounts. She insists it is the way to go even though it is inaccurate. People prefer what they have had to superior new stuff.
CBL
Rob Lister said:
So what is the potential for fraud with the other methods? I did ask about the computer in optical scanning machines.
It seems any system is potentially subject to fraud. To pick on the touchscreens alone seems like maybe there is a competitor out there driving this.
CBL4 said:
I would be interested in knowing which machine the test subjects prefered. I would also be interested in knowing what kinds of screens a voter sees on the touchscreens. It seems to me that you could make the displays as user-friendly or as user-unfriendly as you want and that someone will one day make that very argument in court. "The screens were too complicated to understand."
Re: Re: Wired News: More Diebold Machine Flaws
I once knew someone who got $50's when $20's should have been dispensed. I guess you could blame that on human error, if it was loaded wrong.
I also understand you can be charged with a crime for not correcting such a mistake when you see it. Lucky for this lady, it wasn't a large sum of money so I doubt the bank really cared.
BPSCG said:
I once knew someone who got $50's when $20's should have been dispensed. I guess you could blame that on human error, if it was loaded wrong.
I also understand you can be charged with a crime for not correcting such a mistake when you see it. Lucky for this lady, it wasn't a large sum of money so I doubt the bank really cared.
I don't think that claim would get far. People in nursing homes vote in higher numbers than the population at large, even in nursing homes catering to Alzheimer's patients. How do you claim the screen is too complicated if Alzheimer's patients can figure it out?Luke T. said:
Of course, the Alzheimer's patients are getting a little "help" in filling out the ballots - from family or from nursing home attendants. Speaking of fraud.
Unfortunately, my mother lives 2500 miles away. I will ask her the next time I talk.
I should note that she is on a local board not the national one. Her experience may be atypical. The LoWV does have some articles on their website about voting machines - http://www.lwv.org/
CBL
Luke T. said:
There is a possibility of fraud in all systems, but the biggest complaint of the touch screen systems is the lack of a paper trail in case an audit (unlikely as that may be) is necessary. Proposals to "improve" the reliability of the touch screen systems usually have a paper print-out of the votes (not a tally of all votes entered) by each person though those with the votes encoded in a bar code (i.e. non-human readable) are questioned as being tamperable by altering the print-out while you are altering the vote recorded. A paper trail that the person voting sees and oks as his/her/its vote is a good way to prove the accuracy of the system.
One good thing with paperless touch screen systems is that the Florida debacle with endless recounts and discussions of what constitutes a valid vote (dimpled chads) cannot happen as the only record will be an electronic one, on removable cards, hard drives, or transmitted over networks such as the internet.
The problem is that if a person can touch the paper ballot, it can get lost or replaced. This could lead to a false implication of fraud. Allowing the voter to see but not touch seems like it would be expensive and he still might not believe its accuracy.
CBL