Are E-Voting Paper Trails A Bad Idea?

from the depends-on-your-point-of-view dept is running an interview with Carnegie Mellon CS professor Michael Shamos, who is rather opinionated about everyone who keeps pushing for a voter-verified paper trail in e-voting machines. He thinks that a paper trail is a bad idea, but his reasoning is a bit problematic. Most of his complaints are around the fact that paper ballots aren't particularly reliable either (which no one denies). But that, alone, isn't a reason to get rid of a paper trail. After all, the idea is that the paper trail is a backup -- a way to check and verify that the electronic votes are counted correctly.

From there, Shamos complains that various voting laws say that if the e-voting totals and the paper totals don't match up, you have to use the paper totals. To him, that's a problem because (as he pointed out) the paper totals are not particularly reliable. But, again, the problem here isn't actually with the paper trail, but the laws that automatically ditch the e-voting result, rather than specify a way to verify what the real vote total is.

Towards the end, Shamos does make some good points about how the real focus should be on building a system that does end-to-end verification for reliability, so that people can feel confident that their vote was counted correctly. However, he falsely accuses supporters of paper trails of not being interested in this. That's simply inaccurate for the most part. Almost every expert on the subject that you talk to wants a better overall e-voting system that is more secure, accurate and reliable. Having a verified paper trail isn't meant to solve all the problems, but as a temporary solution for an obvious problem with current e-voting machines. Yes, we need a better overall system, and, no, paper isn't perfect -- but Shamos seems to be accusing the paper backup supporters of stuff they're not saying.

Reader Comments (rss)

(Flattened / Threaded)

  1. identicon
    Anonymous Coward, Apr 22nd, 2008 @ 1:41am

    The taxpayers spent billions on voting machines that couldn't be certified in a casino.

    Now Shamos is complaining there's no more money for voting research.

    A lot of people come home broke from the casino. But at least they had fun, yeah? Saw a few shows, drank some free alcohol..

    reply to this | link to this | view in thread ]

  2. identicon
    Anonymous Coward, Apr 22nd, 2008 @ 3:04am

    The answer to the question: How can we WASTE the next billion dollars of TAX payers $$!!!!!!!!!!

    reply to this | link to this | view in thread ]

  3. identicon
    Sterge, Apr 22nd, 2008 @ 5:05am

    I'm a computer programmer by both trade and hobby. I have been baffled by this 'issue' since I started hearing about it years ago. I guess my confusion comes in the difficulty in writing a computer program that 'tallies votes'. This is a project for an undergraduate student. A simple project... one that doesn't require difficult new algorithms or fantastic framework design. This isn't HARD to do...

    And what exactly can't be 'verified' in either the 'archiving logs' of whatever database that is storing this 'complicated' (lol) data or some simple 'coded logging' system that logs each users selections (securely and encrypted of course)? I must be missing some kind of 'legalese' that is preventing the genius coders from perfecting this simple system I suppose?

    I could always use the extra cash if they'd like to reassign this project lol.

    reply to this | link to this | view in thread ]

  4. identicon
    Mr. PaperTrail, Apr 22nd, 2008 @ 5:31am

    When you say the paper trail is....

    unreliable you're saying that the individual looking at the paper and submitting it didn't read it prior to the submittal.

    What the paper back up should do is give the voter a chance to self verify their vote. I would hope that any voter who found the paper ballot not match their cast vote would sound an alarm that the machine did not function as designed.

    The support from paper ballots comes from the fact that there is a one to one correlation between what is in the hand and what ends up in the box. With the E-voting what you have is a image on a screen and a counter in memory. You are dependent on the link in between to be reliable in order to trust the relation between the two.

    reply to this | link to this | view in thread ]

  5. identicon
    Haywood, Apr 22nd, 2008 @ 6:06am

    Re: When you say the paper trail is....

    That is the only sensible solution. A paper trail that only is printed to verify by officials could, and likely would, be as flawed as the electronic result. I could see having it spit a paper at you on the way out that clearly indicated your choices working. Maybe not in Floriduh, though.

    reply to this | link to this | view in thread ]

  6. identicon
    TheTraveler, Apr 22nd, 2008 @ 6:11am

    Re: When you say the paper trail is....

    The problem with paper ballots is not the one to one comparison that’s the issue. Yes, you vote, you look at your ballot and it’s correct you put it in the box and have your self a wonderful day! The problem is the counting! This seems to be the hard part if you take the paper trail as trump then someone that lets say likes one candidate, can “OOPS” I miscounted, but you don’t know, because there are tons of paper votes especially in a big city. Human error on the counting is what is about an accurate as a drunken midget trying to fire a broken arrow with a long bow! At least the computer can keep count when you press the button.

    reply to this | link to this | view in thread ]

  7. identicon
    Anonymous Coward, Apr 22nd, 2008 @ 6:13am

    Indeed, this is a very simple application, if you TRUST the programmer. Suppose some obscure conspiration, which leads someone (an insider) to corrupt the algorithm to transfer some votes from A to B. Suppose this change is made BEFORE the data is stored in the database. Suppose, also, that no one corrupted the algorithm, but someone wants to know who voted in who by getting access to the (encrypted) log files. In all of these cases, the system is vulnerable from INSIDE.

    The danger in this scenario is that only one person, with the right privileges, can jeopardize an entire, conutry-wide, election.

    I still prefer the semi-electronic system, used in some countries, where the voter assigns his vote in traditional paper form, then put it a scanner. He receives an electronic confirmation of the scanned vote (in display), then press OK, and the form goes to the box. If something goes wrong, the box can be openned, and the paper votes can be counted again, either by scanner, or by hand.

    Yes, paper forms can be changed when manipulated, but this will be a local data corruption, not a systemic, country-wide corruption.

    reply to this | link to this | view in thread ]

  8. identicon
    David, Apr 22nd, 2008 @ 6:30am

    EVoting not just a tech problem

    The real problem with E-Voting isn't just technical.

    While writing a simple counting program isn't exactly rocket science there is an entire work-flow built around the process from the actual vote capture to the final tally. This process requires many steps, across disparate communications channel, in an inherently exposed environment.

    While in principle designing the actual capture interface shouldn't be that hard (and even there you'd be surprised!) trusting the entire flow with the requirement that there be no potential fraud, misreporting, vote corruption, etc., given enormous stakes, is a stretch.

    Adding to this is "vendor risk" the fact that the system's vendor is itself subject to potential corruption and may have substantial stakes in the game (e.g. tax breaks, new contracts, personal agenda of senior executives etc.).

    Clearly a paper trail, subject to inherently different errors and modes of corruption adds to the overall security of the process. A visual check by the voter is indeed a very good idea.

    No method is fool-proof, and errors will arise, whether by intent (fraud) or as a result of technology or process failure.

    The point of a paper trail is precisely to make such failures less likely or more difficult to orchestrate willingly, and in the case of a close vote to make it easier to identify large-scale fraud, technological failure or other impediments to the vote process, potentially strengthening the case for a re-vote.

    Clearly there is no silver bullet. But blindly trusting technology or its agents is asking for trouble.

    Given the stakes it seems reasonable to try to make life tough for the bad guy, easy for the good one, and more transparent for all.

    reply to this | link to this | view in thread ]

  9. identicon
    Anonymous, Apr 22nd, 2008 @ 6:33am


    Going to an electronic voting system is such a stupid idea, I don't even want to contemplate it.

    Do you really think someone wont tamper with electronic data? Even if it wasn't internet accessible, I doubt anyone in the world can create a system secure enough to stop a determined electronics "mastermind" from skewing the votes via some sort of device.

    reply to this | link to this | view in thread ]

  10. identicon
    Overcast, Apr 22nd, 2008 @ 6:38am

    It's just a matter of convenience for the politicians. With new technology and such, they feel they should no longer need to take the time and trouble to have to tamper with the paper or punch card votes. It's much easier if they can just have someone change the fields in a database now.

    reply to this | link to this | view in thread ]

  11. identicon
    Powerkor, Apr 22nd, 2008 @ 6:38am

    Re: Re: When you say the paper trail is....

    I couldn't agree more. There will never be a secure system because then the politicans couldn't cheat. Duh.

    If it was 100% secure, then the vote would actually be for the people, by the people, but it never has been and never will be.

    If uncle sam would let our vote count it would be like giving free unicycle lessons to blind elbino monkeys in the rain.

    In other words, they don't think we are smart enough to vote... they know what is best for us... but they still offer the free lessons anyway.

    reply to this | link to this | view in thread ]

  12. identicon
    Powerkor, Apr 22nd, 2008 @ 6:44am

    What the need is to combind analog with digital technologies....because everyone likes the old-school mile counters, but instead used for voting!

    A hybrid of technologies for sure is needed, this is just one idea.

    May you contribute.

    reply to this | link to this | view in thread ]

  13. identicon
    Anonymous Coward, Apr 22nd, 2008 @ 7:06am

    who am I?

    reply to this | link to this | view in thread ]

  14. identicon
    TheTraveler, Apr 22nd, 2008 @ 7:19am

    Re: wow...

    We trust everything else to technology on a massive scale; you have bank transactions crossing the internet on secure lines to little boxes that dispense money for anyone with a little plastic card and a 4 digit number. This complex system is extremely more complicated for the mass volume of different financial institutions, types of bank accounts, languages, card types, transaction types and User interfacing! We can easily make these secure so that its damn near impossible to even tip one over let alone try and hack it. And yes you do however have the issues of implementing different candidates and different user inputs, but you do have to realize that banking ATMs follow under the same constraint as voting machines, you have one person that wants to tamper with them on the code level and you’re done for. And for that reason they have them tested, and debugged, and looked over, and secured, over and over again. If they didn’t millions of dollars would be easily stolen every day with little effort. E-voting is essential due to the advancements in technology and you have about a good of chance as stopping it as a heroin overdosed rattlesnake has of stopping a space shuttle from landing! Again I agree that a total E-system is NOT the answer but I feel that we have the technology to secure both the programming and transmission of this information so why not use it.

    You obviously are NOT the “mastermind” behind anything!

    reply to this | link to this | view in thread ]

  15. identicon
    Peter, Apr 22nd, 2008 @ 7:22am

    Voting papertrail is BAD

    ... Actually the concept of matching a voter to his vote is terrible...
    ... Just imagine in poorer areas where people swap food for votes... The candidate would demand the proof of the vote before delivering the food... It would give this politician an extra tool

    reply to this | link to this | view in thread ]

  16. identicon
    Eileen, Apr 22nd, 2008 @ 8:09am

    Re: Voting papertrail is BAD

    This is a stupid argument. Any politician caught doing this (and how would they not?) would be committing political SUICIDE. People hazard votes for people that *might* be cheaters (i.e., all politicians), but you're not going to waste your time on someone you KNOW is cheating (plus, it is illegal...)

    reply to this | link to this | view in thread ]

  17. identicon
    peter is a retard, Apr 22nd, 2008 @ 8:23am

    yeah peter your just a fucking idiot

    reply to this | link to this | view in thread ]

  18. identicon
    Jason, Apr 22nd, 2008 @ 8:36am


    How stupid can the manufacturers of these systems be?
    First you vote electronically.
    Then you view your results.
    Approving your votes prints out a paper copy of your votes.
    Then you deposit said paper copy of votes into ballot box.
    This way there is a paper copy for every electronic vote submitted.
    You can give me a $100,000,000 for my idea. Thanks.

    reply to this | link to this | view in thread ]

  19. identicon
    jason is a retard, Apr 22nd, 2008 @ 9:24am


    jason your just a fucking idiot

    reply to this | link to this | view in thread ]

  20. identicon
    TheTraveler, Apr 22nd, 2008 @ 9:27am


    I think that is pretty much the way they do it… I don’t remember from last year when I voted, and I do not vote in the primaries, but I thought that WAS the way it happens.
    You E-Vote, It prints a piece of paper and then you turn that in! I don’t understand really why this is a big deal. I don’t think the hand count should supersede the electronic count when it comes down to it. I think Washington is just to scared to allow a computer to handle this.

    reply to this | link to this | view in thread ]

  21. identicon
    Chris Brand, Apr 22nd, 2008 @ 9:56am

    Re: wow

    The big difference between a banking system and a voting system is that the banking system provides a way for the everyday user to check what happened - it's called a statement (they're also much more concerned about making sure all the numbers add up the same in different parts of the system - read Ed Felten's account of how that's not the case for the people who supervise elections).

    If the voting system produced a statement for each voter saying "this is how your vote got counted", then they could check that it was correct. Amazingly enough some now do - it's called the paper trail.

    reply to this | link to this | view in thread ]

  22. identicon
    anon, Apr 22nd, 2008 @ 10:33am

    Shamos and others are doofuses!

    The question isn't Electronic vs. Paper. It's (Electronic + Paper) vs. Electronic.

    All of Shamos' arguments are complete straw men.

    reply to this | link to this | view in thread ]

  23. identicon
    Anonymous Coward, Apr 22nd, 2008 @ 10:58am

    Optical Scan ...

    ... I said it in the original article's comments and I'll say it again. The simple solution, one that has been in use for almost 20 years, it the optical scan system.

    You mark up a ballot and the machine scans it. If there is a dispute, you can go back and have humans look at the voter marked ballots.

    Simple. Easy. Not foolproof (what is), but it achieves all the supposed design criteria without breaking everything.

    But, fundamentally, the only problem with a traditional paper ballot (the kind you mark with an X in a box) is that it takes forever to count them and Americans want instant gratification (or at least the media does since then they can have their election night circus).

    reply to this | link to this | view in thread ]

  24. icon
    Mike (profile), Apr 22nd, 2008 @ 11:59am

    Re: Voting papertrail is BAD

    ... Actually the concept of matching a voter to his vote is terrible...
    ... Just imagine in poorer areas where people swap food for votes... The candidate would demand the proof of the vote before delivering the food... It would give this politician an extra tool

    Peter, I don't think you understand how the voting paper trail works. It doesn't give the voter a receipt to hand out or show to anyone. It just lets them confirm -- in the machine -- that his or her vote was accurate.

    reply to this | link to this | view in thread ]

  25. identicon
    Anonymous Coward, Apr 22nd, 2008 @ 1:30pm


    The way this professor twisted things around it makes me wonder if he has some hidden agenda.

    reply to this | link to this | view in thread ]

  26. identicon
    Ms research e-voting vs paper, Apr 22nd, 2008 @ 9:33pm

    e-voting paper trails

    Why does anyone think that machines secretly programmed by private corporations with agendas can be trusted over your neighbors in multi-partisan teams hand counting your votes with security enhancements, closed circuit TV monitoring the counting and simplified easy to mark and read ballot styles?! An average precinct has 900-1200 people vote in a heavy turn-out. For 100 years citizens at the precinct level have easily hand-counted votes. Did fraud occur? YES, however not anything like the systematic exponentially explosive nationwide fraud we have today. No amount of pre-testing will ever detect fraud, come on computer folks you KNOW this! The computer behaves differently in test mode than how it is programmed on election day to count your "verified" vote. Google the Hrusti hacks and get with the real program here...these machines can record the vote differently than the paper you "verify". Go read the newly released Singer vs. HartInterCivic lawsuit and do your homework folks. Get it right on election night. Audits are completely inadequate and can be easily gamed. Folks, we have not had real elections for a long time since the advent of the op-scan, we have had "selections". The whole florida hanging chad show was by design-Dan Rather reported that story as the ballots were ordered at the printers to fail in Democratic leaning counties. Senator Bob Ney who is now in jail brought you the HAVA bill that allocated 63 billion dollars of our taxpayer dollars to purchase these hunks of junk.
    MANY computer experts will tell you that computers have no place in elections. I know techies do not want to hear that.
    Yes computers can be designed to do almost anything, but every citizen provides the check and balance on auditing ttheir own bank account. Not so with elections. DRE's have antique operating systems and the federal certifications are paid for by the vendors as reported by Bev Harris and Lou Dobbs. Admit it-we do not have safe elections-sad to say this is a disaster. My hope and wish is that every citizen read up on this, do your research and help restore integrity so we can restore our stolen democracy. Yours in hope...

    reply to this | link to this | view in thread ]

  27. identicon
    AThayer, Apr 23rd, 2008 @ 10:18am

    E-voting paper trails

    Hear, Hear!!! I heartily agree with the comments by Ms Research.

    I've been an election official in Vermont for 18 years, in two different towns, both of which use hand-counted paper ballots for all elections. As our presidential elections are also the time for state and county elections, our ballots can be very complex and involve over 15 offices. The same is true for town elections.

    Rather than being a burden, hand-counting of the ballots is an opportunity for community. We get to know people of opposing parties as we hand-count together. AND our results are ALWAYS verifiable. Paper ballots are used for federal elections in Canada and hand-counted. Surely the US can do as well as Canada in our own elections!

    What Shamos misses entirely is that the ONLY reliable way to verify a voter's intent is to use paper ballots and hand count them. A reasonably short amount of training of counters (less than 15 minutes) can assure that the votes recorded by pairs of individuals hand-counting paper ballots are accurate and reliable. When a recount is required, the paper ballots themselves are proof of the voter's intent, not a punch card, or some electronic notation on a computer card or system.

    All the computer programmers I know, including my husband, assure me that programming an electronic machine, or electronic card, to flip votes is an easy task. Computer code can be inserted to start the process after the election starts, and end just before the election ends, and the coding can be done in a manner the isn't observable by "testing" the electronic machine or electronic cards for a particular election. That code can assure that the preferred candidate always wins by a certain percentage point, regardless of how the votes were actually cast.

    I'm fascinated that there isn't more discussion about how to reclaim this essential part of democracy, and to ensure that voting on hand-counted paper ballots is used even in our largest cities. It is always possible to get public participation if the right approach is used. It seems likely that, as interested as people are in who gets elected in local and national elections, it is possible to get as many people to volunteer to count ballots as the nubmer of people who pay to go to a basketball or football game, or a Nascar race.

    In using electronic systems, we are ceding the most important part of democracy to electronics and machines without analyzing whether this is really going to preserve the integrity of elections.

    reply to this | link to this | view in thread ]

Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Use markdown for basic formatting. HTML is no longer supported.
  Save me a cookie
Follow Techdirt
Insider Shop - Show Your Support!

Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Report this ad  |  Hide Techdirt ads
Recent Stories
Report this ad  |  Hide Techdirt ads


Email This

This feature is only available to registered users. Register or sign in to use it.