Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Government The Internet Your Rights Online Politics Technology

WA Election To Try Online Voting 304

AuMatar writes "According to the Seattle Times, the King Conservation District is going to allow online voting to combat chronic low turnouts. You can already view the voting portal. As a citizen of WA seriously concerned with politics, anything that completely removes a paper trail like this scares me. Luckily, this is probably the least important election in the state. I wonder if anyone will hack the election so 300% of voters vote for Firefly or Stephen Colbert or something."
This discussion has been archived. No new comments can be posted.

WA Election To Try Online Voting

Comments Filter:
  • by Attila Dimedici ( 1036002 ) on Monday February 21, 2011 @04:07PM (#35271852)
    I have never understood the emphasis on voter turnout. It is more important to have voters who understand (and care about) the issues being voted on than it is to have a large number of voters. Making it easier to vote does not improve the responsiveness of government to the voters, it actually does the reverse.
    Of course if one examines the other policies supported by the "make it easier to vote" groups, one quickly realizes that they
    want a larger number of poorly informed voters.
    • by Hatta ( 162192 ) on Monday February 21, 2011 @04:46PM (#35272230) Journal

      High turnout means more people have consented to be ruled. Low turnout means they've withheld their consent. It has a direct bearing on the legitimacy of the government.

    • It depends on who you are trying to get elected. The higher the turnout, the harder it is to identify the dead ones.
    • When the average turn out for an election is less than 1% of the eligible voters, turn out becomes very important. Also, if the election were handled the way that typical elections were and mailed out to registered voters, I'd tend to agree with you. But in this case it's a sort of secret handshake deal because the state doesn't handle the election.

  • by masterwit ( 1800118 ) * on Monday February 21, 2011 @04:09PM (#35271876) Journal

    From the article:

    When Washington, D.C., tested an open-source electronic voting system intended for armed-forces members last year, a team of University of Michigan computer scientists hacked in and altered votes.

    Each time a vote was cast, the hackers left a "calling card" on the screen, played the Michigan fight song and secretly changed the latest vote — until election officials shut down the site after two days.

    "This obviously doesn't go a long way in building public confidence," Election Trust Managing Partner John Bodin said of the incident. But that shouldn't tarnish a "trusted" industry leader like Scytl, he said.

    On another note:
    Here is a Berkeley paper that looked at a voting system by Scytl used in Florida: http://www.eecs.berkeley.edu/~daw/papers/scytl-odbp.pdf [berkeley.edu]
    They we're mixed in their findings (jump to the conclusion if your just browsing...)

    I know fraud happens with paper, I know this saves money, but I'm still skeptical.

    From the FAQ after the second link in TFA:

    Q: How does the King CD eVoting platform provide end-to-end online balloting security?
    A: Secured by Scytl USA, this solution provides end-to-end security. Votes are encrypted and
    digitally signed by voters in the voters' voting devices (e.g., PCs) before they are cast. The private
    key to decrypt the votes is divided in shares which are distributed to the King CD Electoral Board
    (community stakeholders) before the election begins. The private key is destroyed in this process
    and do not exist during the election. At the end of the election, the King CD Electoral Board
    members have to meet to reconstruct the private key and decrypt the votes.

    Encryption is a good start... really I have mixed feelings about this too. Any thoughts on this encryption anyone? - I would love to hear from someone with industry experience.

    • by Thud457 ( 234763 )

      and it was all like beep! beep beep! and it at my vote.

      And it was a really good vote, too.

    • by RingDev ( 879105 )

      Question though... what happens if for what ever reason the private keys are lost/corrupted?

      If all it takes is disrupting the keys to prevent a specific district from having their votes counted, it could be quite damning. Even if they spread the key over every single district in the state, if any one of those key shares is corrupted/damaged/losts, would it not prevent the reading of any of the votes?

      Seems like given such a system, there is almost absolutely a back door of some kind. Having an entire state l

      • Generally the scheme is designed such that you don't need all of the pieces to recover a working private key. They use a variant of erasure code [wikipedia.org] to divide the key into N pieces, where any M pieces (M < N) are sufficient to reconstruct the original data. (This is the same technique used for forward error correction on CDs and DVDs, in the Tahoe-LAFS distributed filesystem, and in many other protocols. A ratio of eight pieces required out of ten generated is typical, allowing recovery with up to two lost p

    • Encryption is a good start... really I have mixed feelings about this too. Any thoughts on this encryption anyone? - I would love to hear from someone with industry experience.

      What does encryption accomplish, in this case, other than to help make sure the vote isn't altered in transit between the voter's machine and the server? That sort of vote-by-vote interference would be a very ineffective way to throw an election.

      It's like bank fraud - while obviously you want your banking sessions to go through SSL, people still manage to steal credit cards by the millions due to server hacks, idiot bankers who carry customer data home on unencrypted laptops, etc.

      When it comes to online vot

      • by vlm ( 69642 )

        What does encryption accomplish, in this case, other than to help make sure the vote isn't altered in transit between the voter's machine and the server? That sort of vote-by-vote interference would be a very ineffective way to throw an election.

        When I was young and poor the voting site at the dorm had multi-hour lines out the door and most of the machines don't work. Now that I'm old(er) and relatively wealthy and living in the appropriate area, oddly enough the voting site never has more than 5 minutes of waiting and all the machines work.

        Oldest trick in the book. So the T-1 to the poor persons voting district will have a BER of about 1e-3, awwww too bad, and the T-1 to my voting district will have a BER around 1e-12. What a surprise!

      • I have no doubt that the decision was made by non-technical people, and at best they were advised by IT managers (who, in my experience, tend to talk a good game with non-techie folks but really don't have anywhere near the requisite knowledge to be the sole technical advisors in this role).

        Thus is life. I mean once they had the risk tracking system in place and the 8 documents and signatures on it... well the plan is invincible. Plus with all the risk mitigation activities in place nothing could possibly go wrong. (And if it did well there is a plan to reduce the blame!)

    • My semi-educated guess here is that any sort of proprietary encryption protocol is more open to attack than a well-known publicly documented system, because once it becomes a valuable target (and public voting mechanisms are definitely high value targets) its security-by-obscurity goes away rapidly. Once security-by-obscurity is gone, then the proprietary algorithm is at a significant disadvantage just because there were fewer White Hats looking for bugs.

      In other words, I won't support any proprietary proto

  • by urbanriot ( 924981 ) on Monday February 21, 2011 @04:10PM (#35271884)
    So rather than politically engaged voters who care, travelling their voting station to cast a ballot, we can now encourage everyone to click vote, based on who has the best style, a trustworthy face and catchy slogans! Like. Comment. Vote.
    • So rather than politically engaged voters who care, travelling their voting station to cast a ballot, we can now encourage everyone to click vote, based on who has the best style, a trustworthy face and catchy slogans!

      Unfortunately 'those who care' don't necessarily make more informed decisions than those who don't. Probably the best solution is mandatory voting which other countries have implemented, which prevents highly motivated minoritys from dominating the political landscape.

      • I have mixed feelings about mandatory voting. On one hand, I know everyone should vote, on the other I don't believe it's governments role to force anyone. Plus, those who wouldn't have voted would be voting randomly and skew the results.
        Personally, I think that the right to vote should actually be a privilege to vote given to those who know what they are voting for/against. How about having a quick exam before the test to about out what candidates positions are, then allowing you to vote? (Ever listen to t
    • What you're failing to grasp is that this particular position isn't handled the way that literally every other issue is handled here. We're an absentee ballot state for 100% of the other issues, but in this case you have to go to one specific voting location on a particular day, and neither the day nor the election are advertised.

      I'm pretty politically engaged myself, but I was voting for a decade before I'd even heard of that particular election.

  • by ravenspear ( 756059 ) on Monday February 21, 2011 @04:11PM (#35271890)

    Seriously, this is just a horrible idea.

    You just cannot reliably determine anyone's identity online.

    There are some functions of government that can already be accessed online, like paying taxes. But that's not a problem since no one besides the taxpayer would want to voluntarily contribute money, so there is little incentive for someone to falsify their identity for that. There is huge incentive for people to participate in a free process (voting) that determines the policy course of states and nations.

    • But that's not a problem since no one besides the taxpayer would want to voluntarily contribute money, so there is little incentive for someone to falsify their identity for that.

      I could turn it into a problem easily enough: Electronically submit a tax return for somebody who's due for a large refund. Change the place the refund is sent to go to an account I own rather than an account my target owns. I'd actually be somewhat surprised if there haven't been crooked tax preparers trying this exact maneuver.

      The reasons that doesn't work have nothing to do with verifying identity online (which you correctly identify as being damn near impossible), and a lot to do with banks being smart

    • I wouldn't be so sure. The Electoral Reform Society in the UK conducts ballots for private organisations, for example leadership votes in political parties and unions. Twice now I've voted in one of their elections and they have a simple and elegant solution, based around snail mailing two login codes to anyone eligible to vote that can only be used once. The login codes both match, and one can't be used without the other, and it's only valid for the person casting the vote.

      Again, it works remarkably well.

      • It doesn't matter how secure the actual voting is, it's the counting that is the problem. In any democratic system, it has to be transparent, anyone has to be able to monitor the counting process, that's simply not possible with electronic voting. Not to mention that the process you mentioned makes it dead easy for votes to be bought, or stolen by threat of force. It makes it easy for a dominant life-partner to force his/her partner to vote for a specific party. It endangers the whole principle of electorat

    • by YA_Python_dev ( 885173 ) on Monday February 21, 2011 @05:46PM (#35272922) Journal

      What's even worse is if you actually can determine someone's identity online, even if it's not 100% reliable. Because then someone somewhere can determine how people voted and all kind of shit will hit the fan.

      But even a 100% perfect, secure, open source, pure gold, RMS-approved online voting system will have a fundamental flaw: people will be able to vote from a location (e.g. home) where others can see how they vote. This will enable criminal organizations to buy votes with money or threats and check that people actually vote the way they want.

      The only way to prevent this is to force people to vote in only one location, the fucking voting booth, where they can and must cast their vote in secret. So even if criminals pay someone to vote for a certain candidate, they will never be certain that he/she actually voted for that candidate.

      Any type of remote voting is fundamentally flawed. It's not about the implementation details, it's the basic concept that cannot work.

      And, yes, this is an actual and real problem: when Italy tried remote voting by mail for Italians abroad in 2008, criminals literally went home-to-home to bribe and threaten people and collect votes. Everyone knows this, but still the Mafia got their candidate elected (Nicola Di Girolamo, for the record). Yes politics in Italy are shitty for a number of other reasons, but that doesn't mean we shouldn't try to make life harder for criminals, here and elsewhere.

      The people that cast 300% of the votes for Colbert with high-tech hacks are the least of anyone's problem. The criminals that move 1% of the votes with low-tech bribes to voters will destroy your democracy.

  • Maybe I should change my name to Tom Dobbs and move to Washington.

    Of course I would never tell anyone to hack the voting system, no matter how many "hack me" stickers have been put on the back of the machines.

  • by jaymz666 ( 34050 ) on Monday February 21, 2011 @04:16PM (#35271944)

    How about a little context in the post about which WA we are talking about...

  • on line voteing can lead to you boss forcing you to vote his way and he can stand right over you as you vote.

    • Can anybody think of a good "duress code" style mechanism to address this? Being able to, for instance, cast your actual vote at time A and then being able to cast further ballots at later times that are silently discarded? Some way of signalling to the web form that the ballot you are "casting" should be discarded?

      Unfortunately, I can't think of anything that you couldn't also use quite efficiently for the various voter discouragement/vote misdirection tricks that are commonly deployed to suppress polli
  • Does online voting necessarily preclude a paper trail, or is there an electronic equivalent of a paper trail? What you want is something independent of the vote counting machines, which can be reasonably secured to prevent tampering, can be used at a later date to perform a recount if necessary, and which doesn't allow anyone to prove which way an individual voted (in order to ensure the secrecy of the ballot). I don't think you can do this with paper with an online ballot; you can't, for example, have peop

    • by mlts ( 1038732 ) *

      Even if users print out a paper copy, how can we know if it was intercepted and tampered with before it went to the printer?

      The ONLY solution to this that has any security whatsoever is a ZTIC-like device that unlocks with a PIN, and hooks up to the user's computer. This way, the user votes and confirms on the ZTIC, and the only thing the computer sees is the encryption transaction passing through.

  • The company's software appears to be from Scytl, a company based in Barcelona, Spain.

    Would anyone consider it a national security issue that public elections be held with technology either openly and freely available for review or at the very least, controlled by entities with not just a domestic presence, but a domestic registration?

    I don't think I'd be okay with the 2000 election "hanging chad" ballots being counted in India, because they might have been the more cost-effective solution. Isn't it okay to

    • I would be extremely concerned about the accuracy, auditability, and freedom from coercion/tampering of a voting system.

      However, I'm not at all sure that nationalism is an effective mechanism for advancing that concern(it might simply be orthogonal, it might actually be negative, by bringing the electoral system under the control of an entity with a strong domestic agenda.., or it might be positive; by bringing the electoral system under domestic scrutiny).

      Nationalism, per se, is (I think) irrelevant
  • by whoever57 ( 658626 ) on Monday February 21, 2011 @04:29PM (#35272078) Journal

    IMHO, postal votes should be reserved for those who can't get to the polling station because of some disability or travel. The problem with postal votes is that, for a family, or anywhere that has a shared postal address, you simply don't know who is completing the ballots and returning them.

    I expect that there are many households where the head of the household collects all the postal ballots, completes them, and then instructs the family member to sign (or simply forges a signature).

    Online voting has the same problem, plus many others.

    • IMHO, postal votes should be reserved for those who can't get to the polling station because of some disability or travel. The problem with postal votes is that, for a family, or anywhere that has a shared postal address, you simply don't know who is completing the ballots and returning them.

      Oregon has universal vote-by-mail, and it works very well. Plus, it pretty much eliminates vote caging.

  • ...control "conservation" in King County?

  • Very strong turnout in favor of "H3r5bal V14gR/\". I hear he really stands up for the voting man, if you know what I mean...
  • I just don't understand the paranoia.

    at the very least, let me vote at any polling location. that kind of convenience might require an ID check so they can pull up my ballot for my location, but if i don't want an ID check, i can go to my regular poling place.

  • I would wager that low voter turnout has very little to do with the ability to make it to the polls on time, and a whole hell of a lot to do with people just feeling like their vote doesn't matter anymore. At All. Period.

"jackpot: you may have an unneccessary change record" -- message from "diff"

Working...