Now that I've upgraded to Europa (Eclipse 3.3), trying Eclipse with the Cygwin toolchain is on my todo list. I'd like to develop in one environment for all my target platforms and then cross-compile.
I started using Eclipse 3.2 and CDT for writing some 'C'/JNI stuff earlier this year. It was okay. I've got Visual Studio, but I don't use it any more if I don't have to. Mostly because I dev on both Windows and Mac OS X.
Two weeks ago, I downloaded "Europa" Eclipse 3.3 and the updated CDT. It's a huge step forward.
I love doing my Java/JNI/'C' work in a single environment.
In brief, if your poll site doesn't somehow shuffle (mix) the order of the voters using a touchscreen, you can infer how everyone voted. Without using timestamps.
Ballot boxes are the physical equivalent of a secure one-way hash. After the ballot goes into the box, there's no way to tie it back to the voter. You achieve end-to-end traceability and the public vote count by ensuring the physical security of the ballots.
At this time, there is no computerized voting system that ensures both the secret ballot and the public vote count. Some academics think it can be done with novel voting systems (e.g. PunchScan). I remain unconvinced.
The rule is election day for the presidential election will be the first tuesday after the first monday in November. This uniform day was set federally to prevent individual states from gaming the system by voting earlier or later.
The date was chosen when horses were the dominate form of transportation. Sunday was the sabbath, so no can do. One travel day was factored in, so Saturday and Monday are out too. Ensuring the voting happened in the second week of the month was done for business/accounting reasons (less disruptive). So it was decided to vote on Tuesday.
I encourage you to call your local elections office and ask to become a poll worker. You'll learn more in a day working an election than you can read online in a year.
The "second chance" feature of my county's Diebold AccuVote OS (optical scanner) units is covered in our poll worker guides. Alas, I can't find an online link for that guide.
Black Box Voting's doc cache has a copy of the AccuVote OS manual online. I just did a quick scan. The ballot rejection feature is mentioned on page 35. Here's the entry.
I'm reasonably certain that all the other vendor's precinct-based optical scanners would have the same functionality.
Note that disabling this feature is a very, very bad idea. Part of how Florida 2000 was gamed was turning off the second chance feature in heavily Democratic areas. So if the ballot had an error, the machine just accepted the ballot and didn't count the votes. (Florida isn't a "voter intent" state. I don't know all their rules. But I think it's an all or nothing system. So any error on the ballot invalidates all the votes.) Meanwhile, the second chance feature was left on in heavily Republican areas. This is one of the reasons why different jurisdictions had different rates of undervoting and rejected ballots.
If you're really curious, email me at zappini gmail com and when I find my electronic copy of our poll worker guide, I'll send it to you.
I don't have a dog in the HR 811 fight. Having argued both sides, I've been ambivalent. (I've even given some money to the good folks at verifiedvoting.org. You should too.)
I can, however, state with absolute certainty that the voter verified paper audit trail (VVPAT) is a placebo and is worse than wrong. The VVPAT is the toilet paper roll attached to the computerized voting machine. Here's why:
1) most voters don't verify their votes
2) there's no reason to believe the votes printed and the votes recorded are the same
3) because these machines are crap, there's no way to meaningfully audit them
I'm an election integrity activist in King Co WA. I attended the "audit" of our Diebold AccuVote TSx machines after the Nov 2006 general election. 21 of the sealed toilet paper cannisters were pulled from the pile of some 400. 3 county-wide races were chosen to recount.
Humans manually recounted the paper trail. One reads, another records.
When the VVPAT wasn't readable, due to rip, jamming, smudge, or whatever, the entire paper trail is REPRINTED from the memory card. So it wasn't clear to me what was being "audited".
This happened 3 times. Meaning 10% of the VVPATs were spoiled.
After the Cuyahoga County OH election meltdown in 2006, an outfit named ESI did a post-mortem. They reported the same failure rate, plus a lot more of other failures.
Advocates for HR 811 make the very valid point that some jurisdictions will use touchscreens without paper trails in 2008. I get that. But the solution isn't to throw good money after bad money, to go through the charade of election integrity. The correct answer is to switch back to paper ballots tabulated on precinct-based optical scanners. It's the best available solution (today).
Any way, back to HR 811. Voters Unite has a great summary of its history and why it's such a bad idea: More Harm Than Good.
I have a counter example. If a poll site has a single computerized voting machine with a "voter verified paper audit trail", then yes, you can infer who voted for what.
That is the big problem with paper ballots - they can be incorrectly filled out.
Nice strawman.
With precinct-based optical ballot scanners, the machine detects undervotes, overvotes, and stray marks. The voter is then given a "second chance", the ballot is spoiled, the voter gets a new ballot. These procedures are part of why voting on precinct-based optical scanners is the most accurate and reliable way to vote today.
Me too. I thought you were arguing for security through obscurity. I was getting my geek mojo warmed up to thrash you. Rereading your comment, specifically the title, I realized that you weren't ignorant/crazy.
There is exactly no one to create a fully computerized voting machine which simultaneously protects the secret ballot (voter's privacy) and ensures a public vote count.
Ballot marking devices, like the AutoMark, are conceptually legitimate, but overwrought in practice.
I have similar experience as a computer person. Plus I've been an election integrity activists for about 2 years now.
My number one reason for distrusting computerised systems is that they enable "wholesale fraud" with a single point attack,
That's a pretty good reason. I have two more.
Most of the gear we're buying, especially the computerized voting machines, are crap. All the comparative studies have shown that they're unreliable. (I don't have cites handy. I'm thinking of the GOA report, the ESI report covering Ohio 2006, and the metric ton of data Voters Unite has compiled and analyzed.
Computerized voting and counting systems fail silently. With voting systems, it's a problem that can't be fixed. With counting systems (e.g. optical scanners), you need robust auditing and testing to verify the systems in use.
I'm a poll inspector in King County WA. Voting on our Diebold AccuVote TSx touchscreens actually takes longer than using a paper ballot. Further, issuing and casting virtual ballots on the touchscreens has more administrivia overhead than just using paper ballots.
Additionally, with paper ballots, there's much less of a bottleneck. Have you seen the documentary "American Blackout"? The under provisioning of voting machines created hours long queues, with people outside in the rain waiting to vote. That doesn't happen with paper ballots.
My poll site has just 5 precincts (very modest in size). We have a dozen privacy booths set up. Last November, during the peak voting time (after work, before polls close) we had 23 people voting at the same time. We just set up more tables and people found space to vote. Not a problem.
Now, if you're talking about tabulation (counting the votes), paper ballots are also not a speed or capacity bottleneck. With precinct-based optical scanners, there's no bottle neck (and they catch under- and overvotes). At the end of election day, results are just uploaded (memory cards and sneakernet).
Due to our laws and rules, my jurisdiction also has a mandatory manual recount every general election to resolve close elections. Computers don't (can't) help there either.
My observation is that the introduction of computers (digitization) made everything more burdensome and complicated, not less. At the same time, all processes and procedures became less transparent.
My conclusion is that computerized (nee electronic) voting does not serve the interests of democracy.
The majority of jurisdictions in the United States use the Australian Ballot system. That means secret ballot, uniform ballots, and a public vote count. This has been shown to be the best compromise for our form of elections.
Providing a voting receipt destroys the secret ballot. Opening the door for coercion, vote buying, and other forms of voter and election fraud.
Receipts are not necessary, or even desirable, for ensuring election integrity.
How can you comment on this issue and not be aware of the problems with computerized voting? If you spent more than 30 seconds googling, you'd find acres of information which completely contradict your statements.
and the only sensible system is electronic with paper trail which removes small errors (and provides accessibility) but allows for backup (albeit with aforementioned small errors) in the case of an electronic failure.
Computerized voting is much more error prone than using paper ballots.
Perhaps you recall the 16,000 under votes in Sarasota, Florida?
I've served as a poll inspector in King County WA, am a geek ninja, and an election integrity activists. You can read my erudite postings on WashBlog.
Oh wait, you just wanted ONE issue. Hm. Hard too choose just one.
It's easy to make a long list if you're allowed to just make stuff up.
Chads
I assume you're referring to the problems with chads in the Florida 2000 presidential race. The specific cause was the machines hadn't been cleaned out in 10 years, so the chads couldn't fall through.
Long history of people cheating them (While the current system sucks, a combo of electtronic + paper if properly done, can double our chances of catching fraud)
Um, no. There is no way to mitigate the problems with computerized voting. A voter verified paper audit trail (VVPAT) is a placebo.
Unless you were referring to the ballot marking devices, like the AutoMark. That's an okay solution.
As for cheating with paper ballot. Um, yea. Your point? In relative terms, are you trying to say that fraud and errors with paper ballots are a greater risk than with computerized voting machines? If so, please explain.
Takes too long to count.
Huh? My county, like many, uses precinct based optical scanners for poll site ballots. Not a problem.
Takes up a lot of space.
Huh? Have you see a warehouse filled with computerized voting equipment? The +900k paper ballots (and misc) from our elections are stored on a few pallets.
Costs a lot more money.
This is completely false. Cost per ballot for computerized voting is much greater than with paper poll site ballots. You may have been (inadvertently) referring to paper mail ballots. In that case, computerized voting and mail ballots are about the same.
If someone is removed from the ballot, we have to reprint, which may not happen in time
That's a straw man argument.
Delivery must be assured with enough to all, which means a lot of waste
With mail ballots, true. With poll site ballots, the elections office estimates turnout and orders 10% more than projected.
Blind people have issues
With computerized voting, true.
With paper ballots, not necessarily. There are many HAVA compliant solutions for assisting disabled voters. Tactile ballots, protective sleeves like the Vote-PAD and EqualiVote, ballot marking devices like the AutoMark, some goofy phone user interface used in Connecticut, the InkaVote used in LA County, etc, etc.
People that don't read english have issues
Reading English, yea. Our paper ballots are printed in Chinese and English. No problem. Soon, I expect will also have Spanish. No problem.
Ballot design for large number of possible candidates - people seriously want to be the guy on the top of the list, it gives a small, but real boost to their numbers
Huh? Candidate order is already shuffled in many jurisdictions, by law, on paper ballots.
So, other than being completely wrong on all points, you did pretty good. Thanks for posting.
Yes, Sherlock, you're absolutely correct. I oppose HR 811's legitimization of insecure, unreliable computerized voting systems. Therefore I support the continued use of insecure, unreliable computerized voting systems.
So, your argument, briefly, is that if we all clap loud enough, Tinker Bell will be able to fly?
I stated that computerized voting machines cannot ensure both the secret ballot and the public vote count.
Your counter arguments were non sequiturs about accuracy and idiot proofing.
We're talking past each other here.
But since you brought it up...
I really should have the cite handy, but anyway, comparatively, with current technology, the most accurate way to count is with paper ballots on precinct-based optical scanners. If you have a better idea, please, by all means, I encourage you to pursue it. God knows we need the help. Meanwhile, I'd humbly ask that we limit the debate to today's world.
You are completely and utterly wrong. Making bold statements with absolute certainty does not make them more true.
The problems with chads in Florida 2000 were because the machines hadn't been cleaned in ten years. The wells were filled with chads from past elections, preventing punch throughs.
There is no means existent to have a computerized voting machines which ensure both the secret ballot and the public vote count. And none of the currently available systems do either.
For giggles, and get some much needed exposure to the reality-based community, please serve as a poll worker or observe the "audits" of your county's computerized voting machines.
EFF is a great org and historically been very good on this issue.
Alas, on HR 811, they're wrong.
The toilet paper trail, what you referred to as the voter verified paper audit trail, is a placebo. I've witnessed the mandatory partial random audit of our VVPATs (King County WA, Nov 2006). A complete sham. When the paper trail is damaged, it was reprinted from the memory card.
Election audits are similiarly wrong-headed. If you work in quality assurance, you'll already know why: You can't test your way to quality. Audits *may* tell you if your election was conducted as expected. Audits *cannot* rule out fraud.
Besides, detecting errors *after the fact* in elections is too late. The correct strategy, approach, priority, whatever is to prevent mistakes.
Other parts of HR 811 is equally wrong-headed. But these two aspects are the ones I know best, so feel qualified to comment on.
Alas, reality does not support your worldview. Sorry. Your post does not merit either "insightful" or a rating above your karma.
Nothing can mitigate the utter failure of computerized voting to both ensure the secret ballot and the public vote count. Can not be done. HR 811, with its toilet paper roll and audits, is just a placebo.
Lack of identification is not a problem. You're parroting the debunked theories of voter fraud.
Election rigging, recently, has been in the form of mass voter disenfranchisement
- caging lists
- intimidation
- misinformation
- too few voting machines available
- throwing away absentee (mail) ballots
- shunting voters to provisional ballots, which are not counted
- etc.
and misc problems with computerized voting machines
- alarming rates of undervoting
- not recording minority languages ballots
- using uncertified software
- suspicious, unauthorized access to voting machines
- etc.
I'm a poll inspector in King County WA and an election integrity activist. You're completely wrong about the advantages of electronic ballots over paper ballots.
The (correctly configured) precinct-based optical scanner rejects ballots with overvoters, undervotes, stray marks, etc.
If you spoil your ballot, you get a new one.
Tactile ballots (e.g. Vote-PAD) have better usability for more different types of disabilities.
Candidate order is already shuffled on paper ballots (where required).
Just scanned the replies (3 or higher, nested). Lots of uncritical praise, very little skepticism. If everyone here can take a breather from the mutual admiration and basking in the collective wisdom, as it were, I'd like to set everyone here straight.
First, and please remember this, the ideal is private voting and public counting. Aka "The Australian Ballot".
Computerized voting machines are fatally flawed. Unredeemable. There is no way to have a fully electronic system which protects the secret ballot as well as ensures the public vote count. Can't be done. Cannot. Be. Done. Period. Despite what all the electronic voting enthusiasts tell you. (I'm looking at you, Avi Rubin.) If you don't understand this, then please stop kibitzing, figure out how our voting systems should work (historically) and get up to speed. Thank you.
Second, this bill relies on "auditing" to ensure the integrity of our elections.
That never works. You cannot test your way to quality. Any one working software knows this. If you're in software and don't, please stop pretending and resign your job.
Additionally, by the time the mistake happens in an election, it's already too late. Too late. Because there's no recourse.
Timothy B. Lee, and other electronic voting apologists, like to mischaracterize the opposition of informed and experienced election integrity activists and experts. I can't guess why. My pet theory is unbounded technolophilia. Others suspect darker motives. Who's to say.
Slashdot Mirror
Nice.
Now that I've upgraded to Europa (Eclipse 3.3), trying Eclipse with the Cygwin toolchain is on my todo list. I'd like to develop in one environment for all my target platforms and then cross-compile.
I started using Eclipse 3.2 and CDT for writing some 'C'/JNI stuff earlier this year. It was okay. I've got Visual Studio, but I don't use it any more if I don't have to. Mostly because I dev on both Windows and Mac OS X.
Two weeks ago, I downloaded "Europa" Eclipse 3.3 and the updated CDT. It's a huge step forward.
I love doing my Java/JNI/'C' work in a single environment.
I wrote about similar attack late last year:
No Privacy on King County's Touchscreens
In brief, if your poll site doesn't somehow shuffle (mix) the order of the voters using a touchscreen, you can infer how everyone voted. Without using timestamps.
Ballot boxes are the physical equivalent of a secure one-way hash. After the ballot goes into the box, there's no way to tie it back to the voter. You achieve end-to-end traceability and the public vote count by ensuring the physical security of the ballots.
At this time, there is no computerized voting system that ensures both the secret ballot and the public vote count. Some academics think it can be done with novel voting systems (e.g. PunchScan). I remain unconvinced.
The rule is election day for the presidential election will be the first tuesday after the first monday in November. This uniform day was set federally to prevent individual states from gaming the system by voting earlier or later.
http://en.wikipedia.org/wiki/Election_Day_(United
The date was chosen when horses were the dominate form of transportation. Sunday was the sabbath, so no can do. One travel day was factored in, so Saturday and Monday are out too. Ensuring the voting happened in the second week of the month was done for business/accounting reasons (less disruptive). So it was decided to vote on Tuesday.
I encourage you to call your local elections office and ask to become a poll worker. You'll learn more in a day working an election than you can read online in a year.
The "second chance" feature of my county's Diebold AccuVote OS (optical scanner) units is covered in our poll worker guides. Alas, I can't find an online link for that guide.
Black Box Voting's doc cache has a copy of the AccuVote OS manual online. I just did a quick scan. The ballot rejection feature is mentioned on page 35. Here's the entry.
I'm reasonably certain that all the other vendor's precinct-based optical scanners would have the same functionality.
Note that disabling this feature is a very, very bad idea. Part of how Florida 2000 was gamed was turning off the second chance feature in heavily Democratic areas. So if the ballot had an error, the machine just accepted the ballot and didn't count the votes. (Florida isn't a "voter intent" state. I don't know all their rules. But I think it's an all or nothing system. So any error on the ballot invalidates all the votes.) Meanwhile, the second chance feature was left on in heavily Republican areas. This is one of the reasons why different jurisdictions had different rates of undervoting and rejected ballots.
If you're really curious, email me at zappini gmail com and when I find my electronic copy of our poll worker guide, I'll send it to you.
I don't have a dog in the HR 811 fight. Having argued both sides, I've been ambivalent. (I've even given some money to the good folks at verifiedvoting.org. You should too.)
I can, however, state with absolute certainty that the voter verified paper audit trail (VVPAT) is a placebo and is worse than wrong. The VVPAT is the toilet paper roll attached to the computerized voting machine. Here's why:
1) most voters don't verify their votes
2) there's no reason to believe the votes printed and the votes recorded are the same
3) because these machines are crap, there's no way to meaningfully audit them
I'm an election integrity activist in King Co WA. I attended the "audit" of our Diebold AccuVote TSx machines after the Nov 2006 general election. 21 of the sealed toilet paper cannisters were pulled from the pile of some 400. 3 county-wide races were chosen to recount.
Humans manually recounted the paper trail. One reads, another records.
When the VVPAT wasn't readable, due to rip, jamming, smudge, or whatever, the entire paper trail is REPRINTED from the memory card. So it wasn't clear to me what was being "audited".
This happened 3 times. Meaning 10% of the VVPATs were spoiled.
After the Cuyahoga County OH election meltdown in 2006, an outfit named ESI did a post-mortem. They reported the same failure rate, plus a lot more of other failures.
Advocates for HR 811 make the very valid point that some jurisdictions will use touchscreens without paper trails in 2008. I get that. But the solution isn't to throw good money after bad money, to go through the charade of election integrity. The correct answer is to switch back to paper ballots tabulated on precinct-based optical scanners. It's the best available solution (today).
Any way, back to HR 811. Voters Unite has a great summary of its history and why it's such a bad idea: More Harm Than Good.
I have a counter example. If a poll site has a single computerized voting machine with a "voter verified paper audit trail", then yes, you can infer who voted for what.
No Privacy on KC's Touchscreens
I live in King County WA. We use Diebold's AccuVote TSx. Just one per poll site (for now, they're trying to eliminate paper poll ballots).
The devil is in the details. Each jurisdiction has it's own arcana and administrivia. So your local mileage may vary.
Nice strawman.
With precinct-based optical ballot scanners, the machine detects undervotes, overvotes, and stray marks. The voter is then given a "second chance", the ballot is spoiled, the voter gets a new ballot. These procedures are part of why voting on precinct-based optical scanners is the most accurate and reliable way to vote today.
Me too. I thought you were arguing for security through obscurity. I was getting my geek mojo warmed up to thrash you. Rereading your comment, specifically the title, I realized that you weren't ignorant/crazy.
Incorrect.
There is exactly no one to create a fully computerized voting machine which simultaneously protects the secret ballot (voter's privacy) and ensures a public vote count.
Ballot marking devices, like the AutoMark, are conceptually legitimate, but overwrought in practice.
That's a pretty good reason. I have two more.
Most of the gear we're buying, especially the computerized voting machines, are crap. All the comparative studies have shown that they're unreliable. (I don't have cites handy. I'm thinking of the GOA report, the ESI report covering Ohio 2006, and the metric ton of data Voters Unite has compiled and analyzed.
Computerized voting and counting systems fail silently. With voting systems, it's a problem that can't be fixed. With counting systems (e.g. optical scanners), you need robust auditing and testing to verify the systems in use.
What part of using paper ballots is slow?
I'm a poll inspector in King County WA. Voting on our Diebold AccuVote TSx touchscreens actually takes longer than using a paper ballot. Further, issuing and casting virtual ballots on the touchscreens has more administrivia overhead than just using paper ballots.
Additionally, with paper ballots, there's much less of a bottleneck. Have you seen the documentary "American Blackout"? The under provisioning of voting machines created hours long queues, with people outside in the rain waiting to vote. That doesn't happen with paper ballots.
My poll site has just 5 precincts (very modest in size). We have a dozen privacy booths set up. Last November, during the peak voting time (after work, before polls close) we had 23 people voting at the same time. We just set up more tables and people found space to vote. Not a problem.
Now, if you're talking about tabulation (counting the votes), paper ballots are also not a speed or capacity bottleneck. With precinct-based optical scanners, there's no bottle neck (and they catch under- and overvotes). At the end of election day, results are just uploaded (memory cards and sneakernet).
Due to our laws and rules, my jurisdiction also has a mandatory manual recount every general election to resolve close elections. Computers don't (can't) help there either.
My observation is that the introduction of computers (digitization) made everything more burdensome and complicated, not less. At the same time, all processes and procedures became less transparent.
My conclusion is that computerized (nee electronic) voting does not serve the interests of democracy.
The majority of jurisdictions in the United States use the Australian Ballot system. That means secret ballot, uniform ballots, and a public vote count. This has been shown to be the best compromise for our form of elections.
Providing a voting receipt destroys the secret ballot. Opening the door for coercion, vote buying, and other forms of voter and election fraud.
Receipts are not necessary, or even desirable, for ensuring election integrity.
How can you comment on this issue and not be aware of the problems with computerized voting? If you spent more than 30 seconds googling, you'd find acres of information which completely contradict your statements.
and the only sensible system is electronic with paper trail which removes small errors (and provides accessibility) but allows for backup (albeit with aforementioned small errors) in the case of an electronic failure.
Computerized voting is much more error prone than using paper ballots.
Perhaps you recall the 16,000 under votes in Sarasota, Florida?
ESI report of the Cuyahoga County Ohio's election meltdown details all sorts of errors with computerized voting.
No intellectually honest person who has looked into this agrees with you.
You either didn't bother to look or stubbornly remain willfully ignorant. Either way, thanks for posting.
I've served as a poll inspector in King County WA, am a geek ninja, and an election integrity activists. You can read my erudite postings on WashBlog.
Oh wait, you just wanted ONE issue. Hm. Hard too choose just one.
It's easy to make a long list if you're allowed to just make stuff up.
Chads
I assume you're referring to the problems with chads in the Florida 2000 presidential race. The specific cause was the machines hadn't been cleaned out in 10 years, so the chads couldn't fall through.
Long history of people cheating them (While the current system sucks, a combo of electtronic + paper if properly done, can double our chances of catching fraud)
Um, no. There is no way to mitigate the problems with computerized voting. A voter verified paper audit trail (VVPAT) is a placebo.
Unless you were referring to the ballot marking devices, like the AutoMark. That's an okay solution.
As for cheating with paper ballot. Um, yea. Your point? In relative terms, are you trying to say that fraud and errors with paper ballots are a greater risk than with computerized voting machines? If so, please explain.
Takes too long to count.
Huh? My county, like many, uses precinct based optical scanners for poll site ballots. Not a problem.
Takes up a lot of space.
Huh? Have you see a warehouse filled with computerized voting equipment? The +900k paper ballots (and misc) from our elections are stored on a few pallets.
Costs a lot more money.
This is completely false. Cost per ballot for computerized voting is much greater than with paper poll site ballots. You may have been (inadvertently) referring to paper mail ballots. In that case, computerized voting and mail ballots are about the same.
If someone is removed from the ballot, we have to reprint, which may not happen in time
That's a straw man argument.
Delivery must be assured with enough to all, which means a lot of waste
With mail ballots, true. With poll site ballots, the elections office estimates turnout and orders 10% more than projected.
Blind people have issues
With computerized voting, true.
With paper ballots, not necessarily. There are many HAVA compliant solutions for assisting disabled voters. Tactile ballots, protective sleeves like the Vote-PAD and EqualiVote, ballot marking devices like the AutoMark, some goofy phone user interface used in Connecticut, the InkaVote used in LA County, etc, etc.
People that don't read english have issues
Reading English, yea. Our paper ballots are printed in Chinese and English. No problem. Soon, I expect will also have Spanish. No problem.
Ballot design for large number of possible candidates - people seriously want to be the guy on the top of the list, it gives a small, but real boost to their numbers
Huh? Candidate order is already shuffled in many jurisdictions, by law, on paper ballots.
So, other than being completely wrong on all points, you did pretty good. Thanks for posting.
Yes, Sherlock, you're absolutely correct. I oppose HR 811's legitimization of insecure, unreliable computerized voting systems. Therefore I support the continued use of insecure, unreliable computerized voting systems.
Award yourself a cookie.
And thanks for posting.
So, your argument, briefly, is that if we all clap loud enough, Tinker Bell will be able to fly?
I stated that computerized voting machines cannot ensure both the secret ballot and the public vote count.
Your counter arguments were non sequiturs about accuracy and idiot proofing.
We're talking past each other here.
But since you brought it up...
I really should have the cite handy, but anyway, comparatively, with current technology, the most accurate way to count is with paper ballots on precinct-based optical scanners. If you have a better idea, please, by all means, I encourage you to pursue it. God knows we need the help. Meanwhile, I'd humbly ask that we limit the debate to today's world.
Again, thank you for posting.
You are completely correct in your assessment of HR 811. Thank you for your efforts.
You are completely and utterly wrong. Making bold statements with absolute certainty does not make them more true.
The problems with chads in Florida 2000 were because the machines hadn't been cleaned in ten years. The wells were filled with chads from past elections, preventing punch throughs.
There is no means existent to have a computerized voting machines which ensure both the secret ballot and the public vote count. And none of the currently available systems do either.
For giggles, and get some much needed exposure to the reality-based community, please serve as a poll worker or observe the "audits" of your county's computerized voting machines.
So, other than that, thanks for posting.
EFF is a great org and historically been very good on this issue.
Alas, on HR 811, they're wrong.
The toilet paper trail, what you referred to as the voter verified paper audit trail, is a placebo. I've witnessed the mandatory partial random audit of our VVPATs (King County WA, Nov 2006). A complete sham. When the paper trail is damaged, it was reprinted from the memory card.
Election audits are similiarly wrong-headed. If you work in quality assurance, you'll already know why: You can't test your way to quality. Audits *may* tell you if your election was conducted as expected. Audits *cannot* rule out fraud.
Besides, detecting errors *after the fact* in elections is too late. The correct strategy, approach, priority, whatever is to prevent mistakes.
Other parts of HR 811 is equally wrong-headed. But these two aspects are the ones I know best, so feel qualified to comment on.
Alas, reality does not support your worldview. Sorry. Your post does not merit either "insightful" or a rating above your karma.
Nothing can mitigate the utter failure of computerized voting to both ensure the secret ballot and the public vote count. Can not be done. HR 811, with its toilet paper roll and audits, is just a placebo.
Lack of identification is not a problem. You're parroting the debunked theories of voter fraud.
Election rigging, recently, has been in the form of mass voter disenfranchisement
- caging lists
- intimidation
- misinformation
- too few voting machines available
- throwing away absentee (mail) ballots
- shunting voters to provisional ballots, which are not counted
- etc.
and misc problems with computerized voting machines
- alarming rates of undervoting
- not recording minority languages ballots
- using uncertified software
- suspicious, unauthorized access to voting machines
- etc.
Other than that, thanks for posting.
I'm a poll inspector in King County WA and an election integrity activist. You're completely wrong about the advantages of electronic ballots over paper ballots.
The (correctly configured) precinct-based optical scanner rejects ballots with overvoters, undervotes, stray marks, etc.
If you spoil your ballot, you get a new one.
Tactile ballots (e.g. Vote-PAD) have better usability for more different types of disabilities.
Candidate order is already shuffled on paper ballots (where required).
Otherwise, thanks for your post.
Just scanned the replies (3 or higher, nested). Lots of uncritical praise, very little skepticism. If everyone here can take a breather from the mutual admiration and basking in the collective wisdom, as it were, I'd like to set everyone here straight.
First, and please remember this, the ideal is private voting and public counting. Aka "The Australian Ballot".
Computerized voting machines are fatally flawed. Unredeemable. There is no way to have a fully electronic system which protects the secret ballot as well as ensures the public vote count. Can't be done. Cannot. Be. Done. Period. Despite what all the electronic voting enthusiasts tell you. (I'm looking at you, Avi Rubin.) If you don't understand this, then please stop kibitzing, figure out how our voting systems should work (historically) and get up to speed. Thank you.
Second, this bill relies on "auditing" to ensure the integrity of our elections.
That never works. You cannot test your way to quality. Any one working software knows this. If you're in software and don't, please stop pretending and resign your job.
Additionally, by the time the mistake happens in an election, it's already too late. Too late. Because there's no recourse.
Timothy B. Lee, and other electronic voting apologists, like to mischaracterize the opposition of informed and experienced election integrity activists and experts. I can't guess why. My pet theory is unbounded technolophilia. Others suspect darker motives. Who's to say.
Anyone actually concerned about the health of our democracy would do well to read the criticisms of Holt's HR 811. Here's Beware of the Bandwagon -- A concise list of problems with Holt Bill HR 811 and Help Amend HR 811 to prohibit "electronic ballots."
That is all. I'm happy to answer anyone's questions. I'll check back later.
Exactly.
That's a poor comparison.
The democracies of India and the United States are very different beasts. Ours (USA) is significantly more complicated.
India has proportional representation and their ballots are very simple.
In the USA, we vote on everything, have more complicated ballots (in comparison), have multiple elections per year, and winner-takes-all elections.
Here's my comment from yesterday sporting greater detail.