Open Source Electronic Voting Progress Limited

An anonymous reader points us to a story about how the problems with electronic voting mostly stem from one source: the lack of mandated standardization. The LinuxInsider article goes on to suggest that once the issue of a universal voting platform is solved, the way is paved for open-source software to address concerns over accuracy and transparency. Though the article states that "no open source program for voting machines yet exists," it should be noted that such software was successfully tested earlier this month. Quoting: "People debate the merits of e-voting for a variety of reasons, including suspicion of new technologies and a general distrust of politics, according to Jamie McKown, Wiggins professor of government and polity at the College of the Atlantic. 'Reports on e-voting security often de-contextualize the history of voter fraud in this country, as if boxes were somehow assumed to be better. You constantly hear calls for paper trails, and open and free inspection of voting machine source code. But it's a very thorny issue and one that has a lot of facets,' McKown told LinuxInsider."

Comments

Standard is already set

[Romancer]

The standard is already being set by the people. Physical and electronic records verifiable by open process and contained in a completely sealed box with tamper detection.

Re:Standard is already set

[Teancum]

The United States got by for over 200 years without electronic voting. We should not switch to electronic voting simply because there is no functional problem with older methods of counting and there are no compelling benefits to justify switching. If it ain't broke, don't fix it.

I don't know how you can legitimately say such a thing. There clearly are some very compelling problems that come from the current voting process that the electronic voting methods are trying to address. Or more to the point, consider the current problem domain:

• Number of contests - Be able to provide an efficient method of allowing for up to hundreds of different contests that take place on multiple levels of governance. For the USA, this means federal, state, county, municipal, and other levels of government jurisdiction. Each office must be treated as a separate contest (in most cases) and allow voters to cast their ballots independently from one race to the next.
• Universal suffrage - Open up the voting process to allow everybody that can be considered a citizen to be able to cast their vote, without coersion, intimidation, or other methods of manipulating voters that would prevent them from being able to cast an honest ballot that genuinely represents their opinion at the voting booth
• Privacy - Going back to the universal sufferage, all votes cast must be kept confidential and ultimately untraceable in terms of being able to tie the votes cast by a specific citizen to a particular candidate.
• Restriction to Citizens - In spite of seeking universal suffrage, elections should be restricted to actual citizens and not non-citizen residents. Citizenship actually means something, including the right to vote.
• Elimination of Fraud - Each person who casts a vote ought to be able to cast a vote, but they are entitled to only vote once. Any means that allows a person to vote more than once ought to be considered fraud, as well as voting when you are not eligible. Tracking when a voter casts a ballot (to ensure they vote only once) and elimination of voter records due to loss of citizenship, death, or commission of a felony (constitutionally permitted means of removing eligibility to vote) are a part of any voting system.
• Voting Accuracy - Methods of tracking ballots cast as well as counting the votes for each canidate must be as accurate as reasonably possible. 100% accuracy is the ultimate goal based on legitimate ballots cast and reported results.
• Speedy Results - While not the most important goal and arguably the least important of all of the requirements I'm listing, results from the election must be reported in a timely fashion in order to provide overall confidence in the results. Voting methods that delay election results are considered less desirable and raise questions in regards to fraud.

I don't know what you have been taught in your history classes, but voting methods to reach these goals have changed considerably in the past 200 years. I mention this requirement domain as it is what seems to be current goals sought after by most election clerks and policy makers in the USA... not that I'm necessarily mentioning any specific law here.

Off the top of my head I can name over a dozen different voting methods that have been used and discontinued over the years. Some form of electronic voting is certainly new, but it isn't even the first mechanical voting system used. What newer voting methods (including electronic voting ideas) provide is a chance to get closer to a "perfect" voting system... even if that isn't necessarily possible.

Back when voting was done by landed gentry in a public meeting where every voter would "announce" their votes verbally, I'm not so sure that they would even understand the current set of voting problems facing the USA today. But that isn't the method used any more, even though it was the voting method over 200 years ago. Surprisingly it is still done in some cases in the USA in some special but limited circumstances, but isn't the situation for a typical general election.

Re:No open source voting?

[maxume]

Apparently too hard for you...

Or is demodratic in there to help fix the election?

Wrong thinking

[vertinox]

FTA:

People debate the merits of e-voting for a variety of reasons, including suspicion of new technologies and a general distrust of politics

I don't think its as much as a suspicion of new technologies as much as the objections of those familiar with it. Even those who works with computers at a basic level understand that its far easier to drag and drop a thousand doc files into a trash can on the desktop than it is to shred a thousand physical copies.

That is my biggest argument for paper ballots is not fear of new technology, but rather a safe guard of making it harder to destroy evidence of tampering. If you wanted to cheat and election, it is far easier to type an SQL command in a console than it is to dispose of or forge thousands of physical ballots without anyone noticing.

In a perfect world, electronic voting would be the obvious choice, but given human nature and politics there should be as many safeguards as possible against possible corruption.

Re:Wrong thinking

[Anonymous Coward]

This biggest problem w/electronic voting isn't the potential for fraud (though that's terrifyingly high), it's the perception of fraud. Given the polarized political climate today, with millions of people suspicious that the 2000 and 2004 elections were stolen, imagine the reaction to a close election "decided" by a completely-unaditable electronic process. Even if the process is nominally "auditable", with most current machines the audit trail can be forged as easily as the original votes.

Regardless of whether or not fraud occurred, huge numbers of people would believe that it possibly/probably did. The whole "he's not my president" meme would grow exponentially. I could easily forsee mass demonstrations (tens of millions of people), massive civil unrest, etc. And keep in mind that the potential for this outcome is completely independent of whether or not fraud actually occurred!

Not only is there no way to prove fraud, there's no way to prove a lack of fraud. That's what scares me.

Re:No open source voting?

[weak*]

while(1) {
if(lever==REPUBLICAN) republican = republican+2;
if(lever==DEMOCRATIC) democratic++;
assert(republican > democratic);
cout << "Republicans: " << republican << " " << "Godless, pussy liberals: " << democratic << end;
}
There, fixed that for you... oops... I may be in violation of my Diebold employee nondisclosure agreement

Nope

[zogger]

Sorry, no computers need be involved at all, open source or closed source or some hybrid. You shouldn't need to be a programmer to verify the count as a volunteer at the end of the day. Any scheme that uses even "open source" software that is "justified" by saying "you can look at the code yourself" is still flawed as most people are not able to read code and understand it, and you still have no idea what happens during and after the election, you would have to stop and analyze the code every single step of the ballot trail. Skip a step = opportunity for compromise with a follow up coverup to hide the tracks. That's two big fat flaws in the idea, and either one is enough to rule out using computerized voting. And if you say "well, this scheme a,b,c uses a paper trail so it is mo bettah!!", so what's the point again then? Just *use the paper trail* as the primary way to vote for the election in the first place, skip the thousand buck computers and rube goldberg nonsense in the first place, including those stupid punch cards with "chads", they aren't needed either. If it takes "too long to count", here's an idea, a full 24 hour voting period, and it can even be a mandated federal holiday for that matter, so no one needs to miss work to go vote, no matter what shift they work or any other excuse.

I love computers, like most folks here have owned them for years and owned quite a few of them, but for elections, I like a plain ballot box and normal paper ballots.

"Open source" with elections is, I am sure, being pushed by well meaning folks, but if falls exactly under the "if your main tool use is a hammer, everything looks like a nail to you" syndrome. It just ain't needed, tons of other projects out there could use the dev help instead.

Re:Nope

[vonPoonBurGer]

I completely, absolutely agree with you, mainly because paper ballots are what we use here in Canada. And just so you know, they don't have to take long to count. Here in Canada, if you want to run a candidate in a riding, you have to provide someone called a scrutineer to every polling station in that riding. The scrutineers are usually low-paid party staff, or completely unpaid volunteers. Each party/candidate has one scrutineer at each polling station, and once the polls close, they all count every ballot. Two important facets of the system prevent counting fraud. One, the counts from each scrutineer have to match, or they don't get to go home. And two, any interested voter is allowed to watch the counting process. The budget for Elections Canada is far, FAR less per capita than in the US, and we know who our new prime minister is on the night of the election. I shake my head and sigh every time I see one of these threads on /. advocating technology as the answer. You're barking up the wrong tree, folks, and wasting an assload of money in the process.

Re:Nope

[Teancum]

While I'm not this poor schmuck, my wife sometimes gets stuck with the onerous task of hand-counting votes for elections that aren't quite so complicated as a Presidential general election. Most of these tend to be a contest for a small-town mayor or some other similar contest where there is only one or two races on the ballot, or something like a school bond election.

Keep in mind that most of the people who are voting judges tend to be senior citizens, who frankly suffer slightly from dementia and other ailments common with advancing age. They tend to be people with free time, and generally manning a polling location isn't that difficult of a task. So are you really trusting the voting results to a bunch of people who can barely remember your name five minutes after you told it to them? In saying this I'm not saying my wife is a senior citizen, as she is in her 30's, but most of the people who she does work with are quite grey and tend to be retirees. This job does tend to be skewed for the elderly.

A true story is how my wife (since she is young enough and has done this for some time, is even head judge) sat down with her crew and counted out the ballots for one of these hand count races for nearly three hours after the polling location closed. All of the judges were given the full stack of ballots (they tried to count different stacks of ballots in a rotation to try and speed up the process) and then tried to compare the counts when they were done. Out of five judges, all five got completely different counts for all candidates. There was a "statistical average" for that precinct, but that is all it was. After dealing with this for yet another hour, (it was after midnight by this point in time), she simply got the other judges to sort of compromise between a rough average of what the counts were at... as they finally got the count within 10 votes for each candidate. The judges agreed and then signed the "official results" with the compromise tally. It didn't seem to impact the results of the election by being off by 10 votes, but that isn't always the case. That really inspired confidence in me that my vote really counted for anything. I guess it does in the long run, but I certainly understand the real need for recount laws after seeing this whole thing.

This is a tough problem. I asked my wife.... "How long would it take for you to process a Presidential general election with your crew by hand counting?" She looked at me real funny like I was trying to crack a joke, and then simply said "I'd just burn the ballots and pretend it never happened in the first place." She really didn't even want to think about the problems she would face in such a situation. After pressing her, she finally admitted that it would take several days or even weeks to count the results, and she wouldn't really trust the results either from her own precinct. She and I get into arguments over the reliability and accuracy of the Diebold machines she uses now, but openly admits that they are much better than what a hand count could ever possibly be. On this point and seeing her fellow poll workers, I'd have to agree.

How to do this right

[Animats]

It's really not that hard to do this right.

• Voting machines should have to meet the Nevada Gaming Commission Standards for Gaming Devices. Nevada has tough tamper-resistance standards (Immune to static shocks, 27KV sparks, 600V on the power input, and rapid turn on/off; must resist forced illegal entry, locked covers over circuit boards and program media), logging standards (counters that cannot be reset, non-erasable logs of program changes), and auditing standards ("Provide, as a minimum, a two-stage mechanism for validating all program components on demand via a communication port and protocol approved by the chairman.") There's no question those standards can be met; hundreds of thousands of slot machines are running right now in compliance with them. Those standards have been developed during decades of struggles against organized crime, employee theft, tax fraud, and attacks on slot machines, so they have serious real-world credibility.
• Use a minimal, published operating system, like Minix. Linux is too big to audit and changes too much.
• Use a paper trail within the machine, one that generates a printed copy of the voter's selections behind a window, along with a bar code representing the voter's choices. For recounts, run the paper log through a bar code scanner for a quick check, and if necessary, manually check votes against bar codes.
• Install two printers, and switch between them randomly, so that the paper trail doesn't provide enough information to tell who voted for whom. Use a printer that doesn't need ink or ribbon and makes a permanent record, like the old "silver printers" used in adding machines. Don't use a thermal printer; the print isn't permanent.

This really isn't that hard.

Maybe somethings aren't better...

[ducomputergeek]

Electronically...

Maybe paper and pencil might be the best tools for the job?

Anyone ever stop to consider that. I know it's blasphemous to say new technology isn't the solution to every problem at the High Citadel of Cowboy Neal, so burn me at the Karma steak...

Simplification

[QuickFox]

In the open-source solution that is linked in the summary, a touch-screen interface produces a printed paper with a barcode for scanning. I think the barcode is a mistake as it's an unnecessary abstraction.

Instead let the voter choose between manual forms and machine forms which both look exactly the same. The only difference is that if you fill in the manual form you make marks with a marker pen, but if you use the touch-screen interface the form comes out of the printer with the spots already marked the way you selected on the touch screen.

The scanner scans both types of forms in exactly the same way. In both cases it looks for the same human-readable ink-filled spots.

Electronic voting IS the problem

[ajs318]

Electronic voting IS the problem.

You can't trust what you don't understand, so any voting system needs to be Universally Comprehensible. An electronic system based on Open Source principles -- where the blueprints for the hardware and the listings of the software are available for all to examine -- is still really only comprehensible to a minority of the population. It doesn't satisfy the goal. (In the worst case, you could conceal a deliberate design defect by a combination of hardware and software techniques: anybody examining the hardware and not the software, or vice versa, will miss it.)

Just forget the whole thing as a failed experiment, and go back to pencil and paper and manual counting. Everybody knows what all the possible failure modes are, and how to minimise their effects.

Re:Electronic voting IS the problem

[arevos]

Actually, there is such thing as an secure electronic voting system. You can use cryptography to ensure that a voting process is at least as resistant to tampering as one done on paper, if not more so. There's some very interesting papers on it.

open source isn't the solution

[dannannan]

Even if the machines are based on open source software, how do you know what has actually been deployed on the machine you use to cast your vote? Someone has to set up those machines. Any public code review or testing, no matter how thorough, is completely nullified if that isn't the software that ends up on the machine on election day.

Why do geeky people (myself included) like to wipe a new machine before they use it? Why do corporate IT departments have policies about wiping new hardware, or machines that have been infected with a virus? Simply because when you are using a general purpose computer, it is complex enough that no human can have any confidence in what it is doing unless they had control over the entire installation process.

D

Re:The powers that be don't want the E- Vote to wo

[Sique]

The truth is those that benefits from the current systems do not want electronic voting to work. You caught me here. I profiteer from paper ballot voting. And I am convinced that for an elementary reason electronic voting will never work.

The whole idea behind electronic voting is to speed up the counting process to have the results early. And that's exactly the reason why I don't want any electronic voting. With paper ballots I (that's me personally. Not a rhetoric "I", but just me, the person registered as "Sique" on Slashdot) can make sure that at least in my voting district there is no tampering with the votes. I can watch the whole process, registering of the voters, printing the ballots, distributing the ballots, sealing of the voting boxes, checking the identity of the single voter, handing the ballots to the voters, putting of the ballots in the box, breaking the seal, counting and charting the results, then resealing the boxes and sending them to the central election office, and recounting them for the final results.

I don't need any special abilities. I don't need to understand code, I don't need to understand hardware, I don't need to know about chip card formats or sending protocols. But I can verify that my vote gets counted exactly as I cast it. Every speed up of the process means I lose the ability to watch what happens to my personal vote, or I have to give up the anonymity of my vote.

Where I come from this ability to be able to watch an election was the reason we caught the election board of a complete country rigging the election, and we had enough proof to put them in prison. I don't see how we would ever managed it without being able to watch the whole voting process.