I'm all for it.  Here is one of my favorite recent articles on this subject:

http://wildbee.org/2008/10/17/open-source-voting-transparent-cheap-and-you-get-to-read-your-ballot/

Public confidence in election results is fundamental to a democracy.  Consequently all aspects of an election must be completely transparent and auditable.  This translates to (1) a paper ballot that can, if all else fails, be inspected and counted by a human being.  If electronic counting is to be employed, transparency requires (2) software that can be reviewed and tested by any citizen (or by a computer expert designated by the citizen), (3) data that is in an open format that anyone can read given equipment conforming to publicly available standards, and (4) hardware that conforms to industry standards.

We have seen that proprietary software and data formats leave preclude any meaningful review by public officials, let alone by the public.  Consequently, after numerous elections with unlikely results, after numerous incidents where votes to revesrse an election suddenly from the electronic 4th dimension, and after expensive proprietary systems have proven themself not only vulnerable to hacking, but also prone to mechanical failure, it is time that the federal government support the development of open-source software to run our elections.

I urge the new adminstration to throw its support behind legislation that would fund development of open source software.
.

Absolutely necessary. I prefer a system of check-and-balance - two methods of counting same data.  Open source on any computer system; the other method must be by hand, publicly observable.  Impossible to game both to match. We had our change meetup in the neighborhood today and I made this my top priority. The rest of the group was nowhere near aware enough of election fraud and its methods as currently installed in our privatized digital vote-count.  Much educating to do - and transparency of the count is the fundamental value we are currently missing in our election system. Consequently we are at risk in every election that the intent of the voters will be secretly undone. Federal reforms have not addressed this yet.  Without addressing this, no other reforms restore our democracy - they only restore the appearance of democracy in which voters are mere actors without true impact on the manipulated results.  

Just going to paper ballots isn't enough. We need a system in which there is no doubt about voter intent, and no doubt whether votes get counted.

Hand-marked paper by itself is vulnerable to ballot-box stuffing, and has an error rate of 1% even with the best optical scanning systems. We have elections being decided by 0.0001% or less, as in Minnesota. Computer-printed ballots should be scannable at that level of confidence, and with continued refinements to the design, might achieve even better reliability.

Electronic voting by itself is inherently unauditable. But electronic ballot markers can eliminate overvoting and incorrect marking of ballots, and can be designed with security features that enhance auditability.

We also have to end the secrecy surrounding voting software and its certification. The great thing about using Open Source voting software is not just that programmers can read it. Anybody at all will be able to test it. We will be able to use the same software for party elections, school elections...

Without transparent elections, our democracy is a sham.  Citizens should be able to see EVERY aspect of the election process EXCEPT the actual act of anyone casting their individual votes. 

In California, Election Code Section 15004 states, in part "...citizens...may...check and review the preparation and operation of the tabulating devices, their programming and testing, and ... [be] in attendance at any and all phases of the election."

This needs to be a federal standard, giving citizens the right to be certain they are not shortchanged by unwitting or unsavory election officials whose election systems are compromised.  In 18 seconds a hacker can place a virus or Trojan horse into an election system and change the outcome without leaving a trace. 

It is essential that all votes be cast on paper and hand-counted in the precinct before the ballots ever leave the premises.  The outcome should then be posted at the precinct and on a Web site (ideally with a digital photo image) immediately upon completion of the hand-count.  That way citizens know how the votes came out in their precinct, and it gives a base to compare to when the Election officials aggregate all the votes for a given county or jurisdiction.

Don't allow secret software run on hackable electronic voting and counting (scanning) machines continue to be the way we learn the outcome of elections.  Demand hand-counted paper ballots counted in the precinct on Election night!

Obama probably won the election by more than 10 percent and not the 6 percent accepted by corporate media.  This election couldn't be rigged because Obama's popularity was beyond the "margin of fraud."  Voters may have given Dems a filibuster-proof senate, but we presently must accept the tally of the privatized, computerized voting systems.  Transparency is the only way we can have elections we can believe in. 

Yes!  We must open up the goverment - so we can see what our energy and our money is doing!  It is our responsibilty to keep the governements actions aligned with our goals/hopes/dreams/integrity for our Country.

We need sustainable election reform and transparency is the foundational step to get us there. At this time there is virtually no transparency in our elections and Observers have little ability to prevent problems, get key documents in a timely manner, or challenge operations when they see a problem. 
The key issues that all include transparency are Observer access & rights, timely reports, enforcement of the law, increase audit to 99% statistical reliability and chain of custody security. 1) Observer Access: Observers are being blocked from seeing and hearing what is going on, they don't get their questions answered thoroughly (the law doesn't necessarily require accurate or timely answers), therefore private vendors and election officials have complete control with no check and balance. 2) Timely reports: the chain of custody is constantly broken because the elections officials don't have to respond to public records requests for 10 days (laws vary across the states). By the time 10 days is up, if the report is received at all (many public records requests go unanswered), the data may have been adjusted so the chain of custody is not reliable. We need real time complete data, such as reports while they are being generated, and we need it delivered in electronic format that can be analyzed, such as Excel spreadsheets and posted to the internet. Too many times Observers get pdf files that are impossible to work with, and delivered too late to be useful or assure chain of custody. America deserves better. 3) Enforcement of the law: currently the weak and outdated election laws we have are not being enforced. New systems of enforcement are needed that cut across conflicts or interest or entrenched networks of decision makers that will not enforce the law. Stronger laws are needed and citizens should be able to press charges. The burden of proof should be on the shoulders of our public servants who are hired to provide reliable elections; Observers shouldn't have to "out guess" what is going on and find out very late that certain things were done or not done, or wonder what happened. 4) Increase the audit so it is robust enough to be meaningful. Go for a 99% statistical reliability, the equivalent of a 10% audit. In fact I believe our political parties need to staff and coordinate Observers who watch the entire election process including the audit.  This is a job that needs to be resourced because the complexity demands expertise and consistent vigilance. Without funding there is a flurry of activity when presidential elections occur every four years and the remainder of the time little is done to prevent or identify issues. 5) Chain of custody security: the paper ballots, paper trails and machine memory cards should all have a thorough, transparency chain of custody trail. At no time should these key items be hidden away where tampering could occur. Take a look at the Transparency Project in CA Humboldt county that recently identified missing votes from Diebold machines, this is the kind of independent verification that is needed statewide and nationally. http://www.times-standard.com/editorials/ci_11161384

I'm not sure how feasible this option would be, but another interesting idea would be to use some sort of visual cryptography (  http://people.csail.mit.edu/rivest/voting/papers/Chaum-SecretBallotReceiptsTrueVoterVerifiableElections.pdf ) where the ballot (one part/key of the visually encrypted ballot) is printed out and given to the voter who can then personally verify the ballot later by accessing a web portal where they can sign in and place the sheet in front of the screen to verify the ballot was counted as marked.

perhaps the web portal causes problems where votes are no longer private and could lead to bullying votes (as the mob used to do). perhaps the solution is to allow the voter to verify the ballot once printed locally at the voting station right before inserting the ballot into the ballot box. eh just some extra thoughts.

Vendor run "black box" elections are about as anti-democratic as you can get.  Vote counting in a real Democracy cannot be done in secret.  We have been poorly served for all the money spent on bad electronic voting systems.  We need a system that is transparent,
understandable, and reliable.  Nothing else will do.

We need security, transparency, voter-verified physically recountable ballots, and accessibility.  We have spent well over $3 billion on new machines and have gotten precious little of any of those.

Open source voting software. Anyone should be able to see what software is counting our votes. As it stands now, proprietary systems mean we DON'T KNOW if a bug has been put into the system to slant the voting results--we're not allowed to look. It's been demonstrated that your average college computer science student can hack the voting machines we have now. This is scary and wrong. 
Open source software and independent, nonpartisan and bipartisan review of vote counting will go a long way towards restoring confidence in our elections. (So will public funding of elections, but that's a different issue.)

I have seen the OVC ballot system demo and their work is way ahead of
any of the scanned ballot systems.  That, and I have a copy of the source
code (they were handing out CDs) to verify what it *really* does.

It is not that I don't trust the developers in an honesty sense, it is because I
have been in the industry for a long time and have read/debugged/fixed a lot
of code.  I know what marketing pressure can do to code quality, especially when the end user, in this case, the public cannot see it.  The nice quality control on open source is the scrutiny it receives by the user community in direct proportion to its critical use.

Until HAVA section 301 (a)(3)(A) gets changed, open source as it stands now is not an answer.  There's no open source system shown to be private and independent for accessibility.  The law needs to be either changed or an open source system must be developed.  This stops a lot of counties from purchasing any system that doesn't meet the requirements.  There are also State laws that make any identifyable mark (barcodes or the like) on a ballot or vote-specific receipts illegal.  That's to stop vote selling. 
There's also the problems already noted in paper-based systems.  We can't do away with paper.  40% of our county's ballots are now vote-by-mail.  We will always have ambiguous results with paper-based systems. 
What with the conflicting laws now in place, there is no common sense way to vote, count and audit elections. 

Open, transparent, and simple voting systems are essential to our democracy

how does one vote for an idea?

 In a democracy it is crucial for the citizenry to trust that their vote is counted.  Ever since we have gone to electronic means of voting, there has been a question as to the validity of the results. There must be a way to check and recheck results.  Reports of more votes tabulated for one candidate than there were voters only increases our insecurity with this system.  Open-source with hardcopy printouts is essential.  We should accept nothing less.

Open Source Voting is smart and necessary (along with universal, standard Vote-By-Mail ballots at both primary/caucus time & in the general) to actually having a democracy.

Kennedy beat Nixon by one vote per precinct. Webb beat Allan in VA by seven votes a precinct. We can all see how close it is in Minnesota in 2008. Remember the Diebold guy saying that George Bush didn't have to worry about Ohio? That's chilling.

It's Our voting system. How dare it be proprietary. Nonsense. Hurray for Open Source.

I'm an advocate for OVC's (http://openvotingconsortium.org/) efforts to bring auditable, open source elections to the general public.

Let's get this done, before the forces for closed, non-democratic voting systems take control again

Anthony, to vote for an idea, simply click on the green number at the top-left of the page, that says VOTE under it.  It will turn red after you vote for it.

As for open voting, this is a must.  If we are going to use computers in the voting process -- which we currently do and there are many good reasons to continue to do so -- it is imperative that all steps in the voting process, including the computer end, be fully transparent and understood by anyone wanting to take the time to look at it.  There have been ample studies showing the current computer voting systems to have large flaws allowing mistakes to happen which can not be detected or corrected.  Whether those mistakes are accidents or deliberate vote tampering is irrelevant -- the system needs to be able to cope with and eliminate all voting errors.

The system also needs to be completely auditable and allow complete recount ability.  Right now that means printing a paper ballot that the voter can view and verify it correctly reflects their vote, and which can be used for spot auditing, and in the event a recount is needed.

A voter's belief that the election can be hacked is a more insidious threat to our democracy than any particular hacked or stolen election.

If anything should be open source in a democracy, it's the election software.

BASICS:  Elections should be affordable for all districts to run as many machines as they need.  Easy to set up.  Easy to train volunteers to run.  Voting machines should be easy to use.  And elections should produce both paper and electronic audits that can be sample matched for accuracy at any time. 

This isn't rocket science.  It's just good business practices.  And it's about time we adopted them.  Open Voting Consortium has done all the hard work of making a system that does all this, so check it out.

No secret software.  No secret counting.  NO VENDORS.  Everything publicly owned, run and reviewed   We demand utter transparency where our votes are accurately cast and  accurately counted.    Certainly,  non-proprietary “open source” software  like that of  Open Voting Consortium (OVC)  with its paper ballot is a step in  that direction.    The need to ensure transparent and accurate voting, where our  PAPER ballots can be marked by each individual, then cast, then counted on ELECTION NIGHT—and always under the watchful eyes of the public—is a pressing issue for we Americans.  If optical scanners count our ballots rather than hand-counting, they should be PRECINCT-BASED only. (Both op-scans and DREs are hackable.)   AND, all “audits” and “recounts” must be predicated upon the results of  the paper ballots, not the results of the electronic machine. Strict chain-of-custody of the ballots—before, during, and after the election—must be enforced by multi-partisan observers. As it’s our  taxpayer money, we want voting done by the cheapest method and with utter transparency and security.  Not the  expensive, secretive, hackable, glitch-ridden, overpriced cash cow bonanza it is now for a few electronic voting machine companies. We are AGAINST the privatization of our voting—it’s antithetical to a democracy. We want voting systems and electoral tabulation that would be publicly owned and OPEN SOURCE, not proprietary software.  Diebold, ES&S—the whole lot of them—have consistently misinformed and lied to those who have bought their shoddy wares. Cost overruns and maintenance are mind boggling to those who’ve bought them. While there are honest election officials, there are also nefarious election officials.  Then, there are those officials who are just plain clueless.  Our whole electronic voting system under HAVA has become needlessly complex and compromised.  It needs immediate fixing.   

Half of us don't register.
Half of registered voters don't vote.
(We permit the 13% most rabid to choose our representatives?)
Those who choose not to participate are too often both highly educated and informed. ("--why encourage the bastards?")

It would help if our voting system was known to work. An open voting system is critical for our republic to function.
Take the process away from for profit business (Diebold, ES&S, etc.). Voting cannot be a for-profit business!

We must make the entire process visible, transparent, and continuously open to public scrutiny. OVC (http://openvotingconsortium.org/) seems the furthest along this path and warrant strong consideration, and support, for their on-going efforts to create a voting system that meets Professor Dill's criteria. Check them out.
 
And, while we're at it, why not remove controls for the Primary  Election process from the Dems an GOP?

See the Stephen Spoonamore videos on YouTube.

There have been a few good voting systems created specifically for this purpose - responding to HAVA and providing voter-verified paper records, high accessibility, security, low-cost operation, etc.  One by one each has been picked off the availability cart, in one way or another.

It must be recognized how many computer scientists of all stripes across the country eschew the machines we have bought.  David Dill is definitely not alone in his criticism.

And yes, Mr. Rogers, adding ballot access for all other parties would serve democracy well.  The citizens paying to hold primaries for the wealthiest two parties while we tell the other parties to go pound salt is not necessarily in the best image of the original intent of our republic.  Adding to that restrictive laws in so many states which force smaller parties to jump through impossible hoops benefits no one.There is no reason we cannot address that issue while we garner ourselves decent voting systems.

Dale Axelrod pointed out to me in an email that election system reform has fallen to a very low priority with MoveOn.  At some point, we need to move beyond attacking the crisis du jour and get to some of the fundamental faults in our society.  The crisis du jour is never a fundamental problem but is always symptomatic of deeper underlying faults.

The voting system is exactly the kind of fundamental problem that must be solved.  It happens that a real solution is at hand and can be implemented with funding already available left over from HAVA (Help America Vote Act of 2002).  We have to follow through.

So, please, dear readers, vote for this and get more people to vote!

 I do not trust Diebold.  I want a machine whose results cannot be changed by a hacker, and I want the machine to print out a PAPER BALLOT verifying the vote. 

I support open source software. I understand it can be made accessible to all. But even if it were not, that is not a reason to continue with secret, proprietary software. I  support open source software in either circumstance.The Americans with Disabilities Act requires a "reasonable accommodation." It does not require turning the world upside down. Insisting that a transparent election integrity be denied to all unless all can use identical voting systems is as absurd as requiring that all stairways and sidewalks be closed off unless and until they are all fitted with ramps or curb cuts. I used a wheelchair for years and most of the time I could not get even a reasonable accommodation. It took years of lobbying to get curbcuts by the local supermarket and by my health club. As frustrating as that struggle was, I would never demand that able-bodied pedestrians be denied appropriate access until and unless  I had appropriate access. And I would not demand that doors that are a problem for me be sealed so that no one can use them.
HAVA needs to be amended to require reasonable access for the disabled, not used as an excuse to deny voting integrity to the whole nation.It is crucial to protect democracy through open source software.

I'm all for paper ballots hand counted at the precinct. But if we really want representatives that are accountable to us, we need publicly funded campaigns. We pay, we vote, we count, we get someone who works for us. The money spent on this election cycle was beyond criminal. Right up there with the assault on Iraq, et al.

This is a good first step toward meeting Jimmy Carter's four components of what's missing from America's voting system (that prevents it from meeting international standards for free and fair elections):
1—free and equal access to the media,
2—nonpartisan election officials,
3—uniformly accurate voting methods within a district, and
4—when electronic technology is used, a physical archive of each vote (voter-verified paper ballot).

Hand Count Paper Ballots--Film it, Follow it, Do IT

There should always be a hand countable paper ballot -  and the technology necessary for volume should be open source- with no corporate secrecy allowed.  Yes We Can !!!

Do away with the "technology" altogether.  Have teams hand counting all paper ballots.  NO DRE'S and NO OPTICAL SCANS.  They're all vulnerable, and let's face it, audits are hard to get and they can be fraudulent, too.

Teams of CITIZENS (not party insiders) should be handcounting all of the paper ballots.

MAKE ELECTRONIC VOTING TRANSPARENT
Open source voting software. Anyone should be able to see what software is counting our votes. As it stands now, proprietary systems mean we DON'T KNOW if a bug has been put into the system to slant the voting results--we're not allowed to look. It's been demonstrated that your average college computer science student can hack the voting machines we have now. This is scary and wrong.


There should always be a hand countable paper ballot -  and the technology necessary for volume should be open source- with no corporate secrecy allowed.  Yes We Can !!!

Let's help to make the voting process simple, accessible, honest, and accountable. 

Invite us back to our own elections?  There's no "negotiation" here -- the elections are the sovereign acts of voting by we the People and there's no higher power than that. (When not voting we are subjects of the law, a much different capacity). 

Any government that doesn't have as a first priority respecting the will of We the People is neither a democratic government with a small d nor a republican government with a small r.  The test, the acid test in fact, for freedom and democracy is whether a voting system is not just transparent enough but allows VOTER CONTROL of elections so that our right to 'kick the bums out' is guaranteed -- most particularly when the bums are crooked or cheating bums -- which is when we need this right the most.

And by this I do NOT mean by a supermajority requirement as was effectively achieved by Obama/Biden.  I mean a guaranteed ability to kick out crooks at any level with 50.01% voting in a 2 way race.  In the final analysis, computers (other than for posting data) will not satisfy the needs of DEMOCRACY, so their features in the areas of speed or convenience, to the extent they actually even exist, are beside the point.  NOt prioritizing democracy's values is like hiring an unqualified person for a Ph.D. position and (most importantly) being oblivious to the risks and impacts of such a decision.

The minimum requirements for transparent elections:

1. Voter-marked paper ballots. (Either hand-marked or printed for the voter by a ballot-marking machine which does not count votes.) Ballots counted with optical scanners, ideally using open (non-proprietary--i.e. non-secret) software

2. Mandatory hand-counted audits of a random selection of precincts, the number of which is statistically-determined to generate a 99% probability that incorrect results (results in which the wrong candidate has been declared the winner) will be discovered. All criteria for drawing samples and the procedures for expanding an audit in the event of discrepancy between the machine count and hand counts announced well in advance of each election.

3. Maximum transparency of the entire election process: testing of the equipment; reporting of the results of each optical-scan machine; tabulation of the results at each level--typically county and then statewide totals; public observation of the establishment of the number of precincts to be audited and the drawing of audit samples; public observation of the hand-counting of audited precincts and the reporting of the results of these counts; and public observation of the process of extending an audit in the event that discrepancies are found between machine results and the results of hand-counting.

I believe this is the _most_ important activity we could focus on; without it we lose our ability to decide any OTHER endeavor!  And we lose our ability to "enforce" the will of the people by electing out those who fail to represent our interest and/or fail to do their jobs correctly. As such this is a "root" issue that goes to the core of the meaning of democracy.
     As a computer scientist on software security, I am thoroughly convinced that the quality of our touch screen and scan based voting systems is dangerously poor, and prone to accidental and intentional inaccuracies.  I feel strongly that these weaknesses have been leveraged. This is based on reading a wide variety of reports throughout the country by technical and non-technical folks alike.
     Obama's victory was decisive. If he had won with a closer margin of votes, we might be looking at a different president today. Citizens groups and new secretaries of state have done a tremendous job in stopping many attempts to suppress votes, but our work is definitely not complete until every legal vote counts.
      We need transparent voting systems. This is one case where proprietary software clearly does Not meet needs.  Voting software is not rocket science- there are public source code systems available TODAY (see http://openvoting.org/ ) that meet needs and cost a fraction of the proprietary systems we are using.  A simple move to these systems would have incredible consequences for our democracy and our future.  If you just sit down and think of what the world might look like today if Florida votes were counted correctly in 2000, you can begin to understand the incredible importance of this effort.

Even in 2008, with the thousands of election protectors out en masse, the exit polls were once again off by 6% in a MO scientific exit poll. In 200 they were off by 6.7% and the media STILL had the audacity to "adjust" the polls. HAVA was drafted by Bob Ney (convicted) and Jack Abramoff (convicted) and was not designed to help voters but the vendors. The "Hanging Chads' in FL in 2000 were enabled with poorly milled paper from Sequoia, it's own employees said wasn't sent to any other county but Dade. It was meant to create hanging chads and then the story was given a few good spins in the corporate media and VIOLA, a rationale for HAVA which greatly enriched Diebold, ES&S, Sequoia and others. Then they sold us election-rigging junk. Counties are now strapped to pay for the upkeep on this junk and our taxes are going through the roof. Enough with the junk, how about TRANSPARENCY with at least a 10% TRULY RANDOM hand-count audit against paper ballots, cast and counted in the precinct by those who cast them? This is not a NOVEL idea. It is what we did before we had machines, before elections were privatized and the count was counted by people we have not elected. With a standardized protocol, and enough good citizens longing to uphold democracy, it is an entirely efficable and verifiable way to count votes. If enough people participate, it can be done in 5 hours or less after the close of polls. Moving our counts to super precincts and out of the polling locations in neighborhoods destroys our political history. We need citizen involvement and honestly good training for our pollworkers. Recruiting fair-minded citizens IS important but with enough honest checks and balances with a bulletproof chain of custody in the system, fraud can be avoided. People, let's do this. Let's try transparency for a change. We can't trust the machines as they are.

I understand where people who want hand-counting are coming from, but it just isn't going to happen. Voting machines are, for good or ill, here to stay.
Given that, our duty is to make them as verifiable and transparent as possible.  openvoting.org has been thinking about this problem for years, and has the best solution I have seen.
If the voting infrastructure isn't reformed, there's no reason to bother with the other ideas on change.org

Several of the comments above argue for replacing all-electronic voting systems with paper ballots marked by hand, but hand-marking ballots brings its own set of problems. The latest example is the ordeal Minnesota is still undergoing, in the Franken-Coleman Senate race, as election judges try to infer voter intent from some truly weird ballots .
The best solution now available (http://openvoting.org/) uses computers not to count ballots, but only to print them. The voter indicates her choices as on today's all-electronic voting stations, but then she receives a printed summary of her selections. That summary ballot isn't actually cast until she drops it into a locked ballot box --presumably after comparing it with her intentions.
At this point, the voting station has served its purpose-- it plays no part in the actual counting of votes. What are counted are the printed paper ballots; they can be counted by hand, by an optical scanner entirely separate from the voting station, or both, as many times as necessary to convince all concerned that the count is valid.
Many more of the comments above argue that election software should be open source. Making election software open source is important, but not for its contribution to election security. Inspecting and testing software may make people feel better, but the good feeling is entirely illusory, because no finite amount inspection or testing of software can ever ensure that it is correct. From a security standpoint, open-source software is about as important for voting software as for banking software. In both cases, the software's correct functioning is ensured not by testing or examination of the software but by examination of the result--banking customers get receipts and statements, voters get printed ballots.
The great contribution of open-source to election software is to free county election boards from eternal bondage to a particular election-system vendor.  Systems like the prototype developed by the Open Voting Consortium, which runs open-source software on cheap commodity PCs, enable their owners to switch vendors in search of better performance or lower prices. The resulting competition, while greatly feared by the vendors of today's proprietary systems, will shift the balance of power from those vendors to their customers, i.e., to us citizens. 

There is no reason that the computer marking the ballot cannot also be tallying the votes within, simultaneously.  This is what computers are for - to do calculations quickly and presumably accurately for us.  Then we recount by hand a significant percentage - say 5% of randomly selected ballots - to help preserve integrity and certainty.  Manual recounts should be available to thos who challenge the election.  And the paper is considered to be the ballot, always.

See the AccuPoll system, which was offered in 2005.  Used with the SureCount option, it meets those standards.  Your vote cast within the computer is not counted until the large, legible, 8.5x11 paper ballot subsequently printed is reviewed by the voter and then scanned back in to the system.  AccuPoll was a public company - we all could have bought stock.  It offered the highest accessibility.  It did not run on any version of Windows (it ran on Linux).  It would have passed all the prior and coming HAVA standards.  The system ran on third-party parts - the counties would not be beholden to any one corporation for supplies and service.  It was designed to meet the requirements of HAVA.

I still cannot believe everyone did not buy it.  They based their east-coast operations out of the county and state in which I live, and still our county and state officials were not buying it.

Anyway, I liked it. 

We do not need to give up convenience and expediency offered by computers to gain security and accuracy.  As long as whatever it is produces a voter-verified paper ballot, we have the beginning of a good system.  Minimizing confusion - using precinct-count scanners rather than central count, e.g. - and developing accessibility follow directly after . . .  as long as we are not tied to using products by companies who are not developing what we really want and need.

There is no feasible way to ensure that results produced by e-voting systems of any type (including optical scan) are valid, i.e., have not been corrupted by fraud or error. Of course cheating is possible even in the absence of such machines, if elections are not conducted properly. However, procedures for running honest, publicly observable elections using hand-counted paper ballots (HCPB) are well understood. HCPB elections are the standard in most of the industrialized word outside the US, including Canada. In several US states, including New Hampshire and Maine, HCPB are used in many jurisdictions, and the parameters of elections there (number of races and candidates on the ballot, number of voters per precinct, etc.) are typical of US elections elsewhere. There have been no scandals involving the numbers generated in such elections.

Surprisingly, HCPB elections are less costly than elections using machines. The principal reason for this is that, unlike most systems using computers, voting systems are used for only a few days per year. It appears that the higher cost of e-voting elections is actually one of the reason why there is so much political pressure to install them. There is a lot of money to be made from the sale and use of e-voting systems. Vendors have been generous in using some of that money to persuade influential people to advocate their purchase and use. Nobody makes money from HCPB. The only beneficiaries are voters and taxpayers. 

A fuller discussion supporting the points made above can be found in
http://www1.cs.columbia.edu/~unger/articles/manualCount.html

Hand counting is a red herring and mis direction, as it will never work in high volume situations. Luddites will be waiting eternally for computerized systems to disappear. Back to reality, the open source element is a must- and the paper ballot requirement obvious. Why is this so difficult ? You would think some group of engineers would put together a system that was better security and less expensive. I'm getting tired of hearing about " Lizard people" and smiley faces drawn on the ballots. The systems should tender a perfect ballot using technology, then count the hand countable ballots publicly using bar code scans.

The calls to return voting in the U.S. to hand-counting are futile. It's just not going to happen. Similarly, the suggestion that all ballots be produced by computer-printer ballot-marking machines is going nowhere. The cost of providing sufficient numbers of machines is too great, especially after the fiasco of buying hundreds of thousands of defective DRE touchscreen machines.

Most national elections require approximately one machine for every 150 voters, to avoid disastrously long lines at peak voting hours. To serve 100 million voters would then require >660,000 ballot-marking machines, nationally. At a minimum unit cost of $3K, that would be another $2 Billion up-front. Plus around $60 million a year in annual storage, maintenance, programming, tech support and transportation expenses.

The only cost-effective, transparent, voter-verifiable way to go is with paper ballots, the vast majority of which are hand-marked by the voter. Every polling location would need to have only one ballot-marking machine, primarily for handicapped voters, but available for other voters who wished to use it, on a non-priority basis. All ballots will be counted by an optical scanner, in the precinct where they are marked. The precinct results would be posted publicly, at the precinct and online, before being transmitted to tabulation sites.

At the conclusion of voting, a publicly-observable process would be employed to select precincts for auditing by hand-count, sufficient in number to assure with 99% confidence that the true winner of each audited race has been correctly determined. All steps in the voting, tabulating and auditing processes would be maximally transparent, with the exception of steps necessary to preserve the privacy of each individual's choices.

No system can be guaranteed to be 100% error- and fraud-proof. But the system described above has the ability to minimize these occurences at a cost which the public--and, particularly, legislators and administrators--are like to find acceptable. It is a system which should be mandated by the U.S. Congress, leading to a uniform system of federal elections, at least, by no later than the 2012 elections.

I support open source and transparency in elections.  I have seen the Open Voting Consortium demo and its biggest strength is that it does not rely just on technology nor just on paper ballots.  It combines the two in a way that provides checks and balances for both the technical and manual processes.   

The posts here about the need to hand count paper ballots are off point.  Let the computers do what they do best (print ballots), and then verify the count with the physical paper trail.  The paper ballots then serve as the physical proof of every vote and can be re-counted if there is any discrepancy.  

When voters manually fill out ballots, they are subject to mistakes.  With computer printed ballots there is no chance of mis-interpreting votes.  We need a high degree of accuracy as demonstrated by many forms of accounting systems worldwide.  The Open Voting Consortium demonstrated a system at Linux World this year which showed that ballots could be printed with the same type of bar code used by Fedex and UPS.  This technology uses a high degree of redundancy to print bar codes for shipping labels so that when a label gets ripped or punctured, the bar code can still be scanned.  In voting this translates to ballots which cannot be spoiled so that all votes can be counted.  

This is what a democratic vote is all about.  Everyone should have the opportunity to have their votes counted, to know how they are being counted, and to watch over them being counted.  These are the checks and balances that ensure a proper vote count and that we remain a democratic form of government.

SOFTWARE  INDEPENDENCE is the key.  Voting systems need to work in a way that you don't HAVE TO TRUST the software.  That's done by opening the FORMATS at each step.  If anybody can read and verify any ballot, and any message from one tallying machine to the next, and everything is signed with strong cryptography, then the system can be audited end-to-end.  Open Voting Consortium's system is software independent.  Hand counted paper ballots aren't.

Open source helps, but it's not the whole solution.  Mainly it discourages vendors with something to hide from particupating.

yes - public confidence in elections is absolutely necessary in this day and age, even if we have to go back to very low-tech soltions like paper  ballots that can be counted by hand. 

I will repost here a comment that I sent in to the site in its first days:

[F]inally: the voting systems problem *has already been solved*. Even before 2004. It's just that no one in power was *listening*. This isn't even my idea; every piece of it was stolen from someone smarter than me.
Terminal does no counting at all, merely assists on low-vision and no-vision voters, and checks to avoid accidental undervoting, then prints a human-readable (but not touchable) OCR-A ballot card and drops it into a clear plastic validation hopper at the terminal. You inspect it, and move a lever to mechanically drop it into a ballot box or a discard box; if you discard it, the terminal knows, and you get to try again. The boxes are cheap, and locked, and can be easily swapped by the poll workers; they get unlocked only for counting, with a key that isn't in the precinct until T-30 at the end of the day. Recounts are a walk on, since humans and computers read the ballot card for counting in exactly the same way.
Each ballot is serial numbered and marked with which terminal printed it; that terminal number can be put on a sticker on the box so voters can see it; the serial numbers and capture of spoiled ballots provide a compete audit trail; the randomization of terminal selection by the voter makes tracking votes effectively impossible.
The terminals are off the critical path since a) they don't do any counting, and b) the voter reads the final ballot card for approval *and sees it go into the ballot box*. End of problems; everyone and anyone can manufacture the terminals and scanners. No sense in monkeying the code on the counters since recounts are trivial, and you can actually double-count the ballots through machines made by two different makers. Oh: *MANDATE* "None of the above" as a choice. I know the parties hate it, but it removes the Undervote Problem completely.
(I've expanded this a touch, and it doesn't coincide *exactly* with some of the other open-voting proposals out there, including OVC's, but it takes *all* of the tchnology out of the security-critical path, as nearly as I can see, and that seems to be the most important issue, to me.)

(My apologies; apparently blank lines are sometimes eaten by the comment machinery...)