Skip to main content

I'm an IT professional, specializing in system and network security.  While I have to admit that constructing a perfectly accurate and 100% tamper-proof e-voting system is hypothetically possible (though highly unlikely), this whole argument misses the point.

The absolute number one requirement for any voting system is that the entire process must be transparent to virtually every citizen.  It is this transparency that ensures fairness, accuracy and security.  Since, with any e-voting system, the entire recording/counting process is conducted by program code that only a miniscule percentage of US citizens can understand, e-voting systems are inherently NOT transparent.  

I'll repeat that, in case you missed it:  E-voting systems are inherently NOT TRANSPARENT!!

You might suggest that those of us who CAN read code can attest to a system's accuracy and security, but what if other experts disagree?  How is it decided which experts are correct, are telling the truth?  Who does this deciding?  Someone who can read code?  How does one who cannot read code trust THEIR decision?  How does one trust the count?

Open Source doesn't address the transparency issue except among people who can already read code.  That still leaves the process opaque to the other 98% of citizens/voters.

The "paper trail" solution is also specious.  I personally deal with, on average, five malware-infected PCs a week that are doing one thing while displaying or printing something completely different.  So, a voter votes and the machine prints a receipt that verifies his input, but the system still records his vote as something else.  Yes, you can hand-count all the receipts in full public view as the official vote count, but then you've simply had the voter use a really fucking expensive pencil to mark his ballot.  

And, of course, all of this ignores a prime IT directive: "If it ain't broke, don't fix it."

So, what precisely is "broken" with a voting system that uses paper ballots, hand-marked and hand-counted in full public view?

If you want to take action to ensure that future elections produce a fair and accurate count, you can start by clamoring for your local election officials to reject ALL e-voting systems in favor of paper ballots, hand-marked in private and then hand-counted in full public view.

Originally posted to sxwarren on Wed Sep 20, 2006 at 01:08 PM PDT.


Take your pick

6%4 votes
93%57 votes

| 61 votes | Vote | Results

Your Email has been sent.
You must add at least one tag to this diary before publishing it.

Add keywords that describe this diary. Separate multiple keywords with commas.
Tagging tips - Search For Tags - Browse For Tags


More Tagging tips:

A tag is a way to search for this diary. If someone is searching for "Barack Obama," is this a diary they'd be trying to find?

Use a person's full name, without any title. Senator Obama may become President Obama, and Michelle Obama might run for office.

If your diary covers an election or elected official, use election tags, which are generally the state abbreviation followed by the office. CA-01 is the first district House seat. CA-Sen covers both senate races. NY-GOV covers the New York governor's race.

Tags do not compound: that is, "education reform" is a completely different tag from "education". A tag like "reform" alone is probably not meaningful.

Consider if one or more of these tags fits your diary: Civil Rights, Community, Congress, Culture, Economy, Education, Elections, Energy, Environment, Health Care, International, Labor, Law, Media, Meta, National Security, Science, Transportation, or White House. If your diary is specific to a state, consider adding the state (California, Texas, etc). Keep in mind, though, that there are many wonderful and important diaries that don't fit in any of these tags. Don't worry if yours doesn't.

You can add a private note to this diary when hotlisting it:
Are you sure you want to remove this diary from your hotlist?
Are you sure you want to remove your recommendation? You can only recommend a diary once, so you will not be able to re-recommend it afterwards.
Rescue this diary, and add a note:
Are you sure you want to remove this diary from Rescue?
Choose where to republish this diary. The diary will be added to the queue for that group. Publish it from the queue to make it appear.

You must be a member of a group to use this feature.

Add a quick update to your diary without changing the diary itself:
Are you sure you want to remove this diary?
(The diary will be removed from the site and returned to your drafts for further editing.)
(The diary will be removed.)
Are you sure you want to save these changes to the published diary?

Comment Preferences

  •  You Think We Could Borrow These? (1+ / 0-)
    Recommended by:

    "[T]hat I have no remedy for all the sorrows of the world is no reason for my accepting yours. It simply supports the strong probability that yours is a fake."

    by Heronymous Cowherd on Wed Sep 20, 2006 at 01:12:08 PM PDT

    •  I don't see a solution to the transparency issue (3+ / 0-)
      Recommended by:
      baxxor, corvo, kraant

      in these.  The opposite, if anything.

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 01:16:03 PM PDT

      [ Parent ]

      •  Transparency Isn't Always the Answer (1+ / 0-)
        Recommended by:

        but I basically agree with you. It may be the initial answer for us. I don't think the Diebold machines would be much more secure, even if the source were completely open. I think that the contract needs to be stripped from Diebold, as the company is way too compromised.

        If you read the article, you'll see it's Linux-based (was CE, but they changed it), and there seem to be no fraud questions.

        Transparency will only get you so far. It needs to be accompanied by protocol and enforcement.

        "[T]hat I have no remedy for all the sorrows of the world is no reason for my accepting yours. It simply supports the strong probability that yours is a fake."

        by Heronymous Cowherd on Wed Sep 20, 2006 at 01:24:42 PM PDT

        [ Parent ]

    •  Awful (2+ / 0-)
      Recommended by:
      sxwarren, kraant

      Just read that entry on the Brazilian voting machines.

      They suck.  Proprietary software.  No paper trail for auditing.  Might as well be Diebold.

  •  What ARE you thinking? (4+ / 0-)

    Without forced e-voting, it would be tougher for neo-cons to steal elections, divert taxpayer dollars to their e-machine buddies, and, ultimately, impose a police state.


    Whom the Gods would destroy they first make Republican.

    by tovan on Wed Sep 20, 2006 at 01:15:19 PM PDT

  •  Transparent is as transparent as you make it (3+ / 0-)
    Recommended by:
    sxwarren, kraant, sfRenter

    My first question in considering voting transparency is, "How can you have secrecy and transparency?"

    Meaning: if we continue to prize the ability to vote in secret we will never know for sure whether our vote is counted correctly.

    I used to think that it would be great if, when you voted, you took a piece of paper and stuck it on a wall so that everyone could see the vote being cast and there would be no way to question it or for it to be "changed" later. I still think this is a pretty good idea.

    I also think that something similar can be done in an e-voting environment. I'm a database programmer. I would like to be able to view the database to verify how my vote was cast. The whole database. And I want the code open so we know if there are any events occuring to change the database for other users.

    •  I like the sticking it on the wall idea, too! (3+ / 0-)
      Recommended by:
      leevank, corvo, kraant

      There are still two things about conducting elections in bits.  

      Most citizens, even if they were shown the whole database, still wouldn't have any idea what they were looking at - "seeing" without comprehending is still not transparent.

      Second, the whole point to "computers" for me is that they provide the ability to manipulate data on a scale and with a speed that would be physically impossible with hard copy - and there's the rub.  Recording counting and storing votes digitally inherently makes the votes vulnerable to manipulation on a scale and with a speed that would be impossible with hard copy.

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 01:46:11 PM PDT

      [ Parent ]

    •  Would be perfect (3+ / 0-)
      Recommended by:
      sxwarren, simon551, kraant

      And the database should be available, minus the identification, to anyone.  If an individual voter checks, and finds that their vote has been recorded incorrectly,  that's a good enough reason to decertify the results. (Although, they would need a printout at vote time, to demonstrate how they actually voted).  Meanwhile, any group or individual could verify the results.

      Unfortunately,  the concern about votes being exposed is very important to some people.  If faced with the choice, however, I would rather have public voting than secret election theft.

    •  It's very easy with paper ballots (4+ / 0-)
      Recommended by:
      ovals49, sxwarren, corvo, kraant
      1. Invite anybody who wants to do so to look at the empty boxes in the morning.  Then seal the boxes, except for the slit at the top.
      1. Hand each voter a paper ballot which they mark in secret and publicly place in the box.  Have an observer present from each party to assure that each voter only gets one ballot.
      1. Count them at the end of the voting day, with counters from both parties having to put the ballot aside if they can't agree on how a particular vote should be counted -- and in full view of any registered voter in that precinct who wants to observe the process.
      1. Since many jurisidictions have lots of minor offices and ballot questions which it might be impossible to get counted on election night, we might take a tip from the way at least some Scandinavian contries do it -- which is to have separate, color-coded ballots and ballot boxes for federal, state, and local candidates, and possibly yet another one for the ballots on which there are bond issues, charter amendments, state constitutional amendments, and that sort of thing.  Then you could seal the ballot boxes for the local offices and ballot questions, for example, and come back the next day and count those votes.  Frankly, there aren't a whole lot of people who are waiting breathlessly on election night to find out who got elected to the town council or county recorder, or whether some local bond issue passed or failed, and it's no major sacrifice to wait an extra day to get some of those elections decided.  In Maryland last week, we didn't get the results recorded and called in, and the voting machines broken down and ready for pickup, until a good three hours after the polls closed.  In that amount of time, or only a little more, we could EASILY have hand counted paper ballots for all of the federak and state offices.
      •  That's exactly the way we do elections (1+ / 0-)
        Recommended by:

        in Waldoboro, Maine.  AFAIK, no one here wants the system to change one iota.

        Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

        by sxwarren on Wed Sep 20, 2006 at 02:59:17 PM PDT

        [ Parent ]

        •  Get ready to change (0+ / 0-)

          One of the unanticipated consequences of the Supreme Court's decision giving the 2000 election to George W. Bush (which I predict will someday be looked upon much the way we now look upon the Dred Scott decision) is that they were forced to use language that talked about people's votes needing to be counted the same way throughout a state, which led to the argument that we need the same voting systems throughout a state (although the Supreme Court never said that).

          And the arguments of the high tech proponents is that in big cities with big precincts, there are simply too many ballots to be counted by hand.  That's utterly falacious, since there are some pretty big cities in Canada and Europe where paper ballots are hand-counted, but a lot of people believe it.  And so the pressure is on to have e-voting even in small localities where it's obvious that hand-counted paper ballots work just fine.

          •  Over my frickin' dead body! (1+ / 0-)
            Recommended by:

            No, seriously.  If somebody attempts to force voters in small towns in Maine like Waldoboro to switch to e-voting, they are going to be in a world of hurt.

            Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

            by sxwarren on Wed Sep 20, 2006 at 03:31:48 PM PDT

            [ Parent ]

  •  User interface versus counting (6+ / 0-)

    The real flaw in e-voting is the idea that computerized counting and an easier-to-use user interface are intertwined. They are two separate problems to be solved, and should be kept separate to provide the necessary transparency.

    If it's determined that punch ballots and marked ballots are too difficult for people to use, come up with a nice touch-screen system that prints out your ballot for you. Tinker with it all you want. Make it work like an ATM, people (mostly) know how to operate those.

    But the voting machine should spit out a paper ballot (not a receipt!), which is then optically scanned by a separate machine. And the ballots must be human-readable for hand counts.

    Anything else is a failed application of technology.

    Are you shaking or biting the invisible hand?

    by puppethead on Wed Sep 20, 2006 at 01:34:28 PM PDT

    •  Nah (1+ / 0-)
      Recommended by:

      That's way too obvious.  (In other words, "Stop making sense".)

    •  Once you have the voter's intentions recorded (0+ / 0-)

      Printing it out and hoping you scan it right or that it is what is scanned is introducing a source of error.

      Everyone who thinks that paper ballots are a fail safe voting mechanism is fooling themselves.  Paper ballots are subject to manual/automatic counting error, fraud, loss, introduction of extra ballots etc.  People are always finding an extra box of ballots in someones car in a close election.

      While the current electronic voting machines seem to be pretty sloppy.  I believe that we can develop machines with open code and a verifiable audit trail.

      Transparancy sounds nice, but there is no way I can personally verify that the 2008 presidential election is counted properly.  I am going to have to take someones word for it.

      If experts disagree about the security of published code, there are enough people out there who are qualified that the truth can be determined.

      •  You can verify it in YOUR precinct! (1+ / 0-)
        Recommended by:

        You say, "Transparancy sounds nice, but there is no way I can personally verify that the 2008 presidential election is counted properly.  I am going to have to take someones word for it."

        You can verify that the ballots were counted right in YOUR precinct, and others all across the country can verify it in THEIR precincts.

        You also say, "If experts disagree about the security of published code, there are enough people out there who are qualified that the truth can be determined."  And what happens when the "experts" disagree among themselves about whether cheating is possible and how easy it would be (as I can almost assure you they would)?  How does anyone KNOW which experts to believe, since most of us don't know enough to be able to figure out which ones are most credible?

        Finally, with respect to the proverbial "extra box of ballots" in somebody's car, that's not going to matter except in an unbelievably close statewide election.  But with e-voting, it's easy to flip the results in any case where the candidates are within 5 or 10 percent of each other, and the patterns wouldn't be obvious enough to be detected.  Let me describe to you the simplest system, one in which every 10th vote (or whatever other rate of cheating you want) for Candidate A is switched to Candidate B.  Let's further look at what would happen in the precincts (Strong A, Strong B, and Swing), if you flipped every 10th vote from Candidate A to Candidate B.

        Precinct #1:

        Actual Vote:  Counted Vote:
        A - 750       A - 675
        B - 250       B - 325

        Precinct #2:

        Actual Vote:  Counted Vote:
        A - 500       A - 450
        B - 500       B - 550

        Precinct #3:

        Actual Vote   Counted Vote:
        A - 250       A - 225
        B - 750       B - 775

        Candidate A would still handily win where he should run strongly, and the precincts that are swing precincts would still be close.  But if you could switch the results of any election where the actual vote for the candidate you wanted to cheat was 55% of the total or less.

      •  No one is saying 'paper ballots are failsafe'. (0+ / 0-)

        But the paper ballot process is, or can more easily be made, virtually 100% transparent to every voter.  Transparency doesn't just "sound nice" to me, it's my number one fucking requirement and it simply CAN NOT BE ACCOMPLISHED with computer voting systems.

        Also, it's much easier to intentionally screw up election results on a huge scale without being observed when the process depends on digital systems.


        If experts disagree about the security of published code, there are enough people out there who are qualified that the truth can be determined.

        Who decides who, among people who can actually read the published code, is qualified to "determine the truth"?  Perhaps we should have a vote.  Personally, I would rather take the word of a bunch of my fellow voters who have watched the ballots being placed in the box, watched the box like hawks and then watched as other voters counted the pieces of paper.

        Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

        by sxwarren on Wed Sep 20, 2006 at 04:17:16 PM PDT

        [ Parent ]

        •  Your number one priority (0+ / 0-)

          Have you ever watched the counting at your local poll?  I haven't.  I think the process isn't nearly as transparent as you think because a lot of stuff is going on at the same time and no one can see the whole process.

          The ability for experts to critique published code and come to consensus is not as hard as you seem to think.  If they disagree, they will have to say why.  There are lots of people who can read code and a consensus will emerge.

          •  I refuse to give any small group (0+ / 0-)

            of "experts" that much power.  The entire process must be open to inspection and verification by as near as possible to 100% of the voting population.  What you're suggesting limits that opportunity to as close as possible to ZERO% of the voting population.

            That is just not fucking acceptable.

            Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

            by sxwarren on Wed Sep 20, 2006 at 08:41:22 PM PDT

            [ Parent ]

  •  Audits (2+ / 0-)
    Recommended by:
    kraant, sfRenter

    Audits of the paper ballots would be random.  The paper count would be compared to the electronic count.  A significantly large discrepancy would trigger a full-scale, PAPER, recount.

    Now how large a sample do you need?  That is a good question but it's probably between 10-20%.

    •  Sample size depends on the margin. (3+ / 0-)
      Recommended by:
      corvo, kraant, sfRenter

      What's an adequate sample size to verify an election that's apparently to be decided by a difference of less than 1%?

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 01:49:25 PM PDT

      [ Parent ]

    •  Not totally random (3+ / 0-)
      Recommended by:
      corvo, kraant, sfRenter

      If I had my druthers, I'd audit a small randomly-selected sample of precincts, and I'd also give each political party the right to designate a small number of additional precincts (perhaps 1%), so that if something just seemed "off" about the automated count in a particular precinct, the party who thought it might have been jobbed could verify the accuracy of the count in that precinct.

      How to decide whether the audit would mandate a complete recount?  Very simply, if the magnitude of the difference in the hand-counted precincts would be sufficient to change the outcome of the election if duplicated in every precinct, then all the precincts get hand-counted.

    •  Any discrepancy (0+ / 0-)

      The thing about electronic voting with a paper trail is that if you pick a machine and manually audit the paper trail and check it against the stored number it should exactly match.

      If it doesn't match, you count the paper over again very carefully and find where you miscounted.

      If you become confident that the paper was counted correctly and does not match the stored value even by one vote it is time for a criminal investigation, recount of the entire precinct manually etc.

      If this happens in more than one or two precincts it's time to look at a new election.

      We expect a little slop and lack of repeatability in our paper ballot processing.  I suspect that the Florida 2000 vote margin was smaller than the expected counting error of handling that many ballots.  With a properly built electronic system, you should not have to tolerate any random variation.

  •  Vote absentee (1+ / 0-)
    Recommended by:

    If you are in a county that's forced to use Diebold machines, vote absentee.

    "Everything's shiny, Captain. Not to fret."

    by rmwarnick on Wed Sep 20, 2006 at 01:46:09 PM PDT

    •  Then you only need faith that (3+ / 0-)
      Recommended by:
      corvo, kraant, sfRenter

      your ballot actually gets to the polling place - and on time.

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 01:50:33 PM PDT

      [ Parent ]

      •  Actually, it's worse than that (3+ / 0-)
        Recommended by:
        sxwarren, kraant, sfRenter

        I've been told that, here in Florida,  the Elections office can throw out your vote if they think there's anything wrong with your signature.  And they don't have to ever notify you of that.

        Think about it.  What should be a felony is SOP.

        •  yeah, e-voting only exposes us... (2+ / 0-)
          Recommended by:
          sxwarren, kraant

          to the next generation of election fraud.

          Technology alone generally can't solve our problems.  But technology alone is not the enemy, either.

          •  Not everything is 'improved' by (1+ / 0-)
            Recommended by:

            implementing a technological solution and, more often than most technology-types (like me) would care to admit, the technological replacement for a manual system that has been working well actually screws things up.  Some things just need to be left alone.  Voting is one of those things.

            Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

            by sxwarren on Wed Sep 20, 2006 at 03:03:36 PM PDT

            [ Parent ]

            •  Our voting system works well (0+ / 0-)

              As long as one side wins by at least 2-3%.  You can have a lot of errors and slop and still not affect the election.  It's just when it gets really close that you need to be more precise than we have been in the past.

              For some reason we are getting a lot of close elections.  I think it is because each party is trying to start from it's base and pick up enough of the center to get over 50%.  They don't want to go too far toward the center to avoid alienating the base.  So they go just enough -- and the vote winds up close to 50-50

              •  The problem with e-voting (0+ / 0-)

                The problem with e-voting is that an election doesn't even have to be all that close to be stolen, as the mock election between "George Washington" and "Bendedict Arnold" run by the Princeton group amply demonstrated.

                •  Which is why e-voting (0+ / 0-)

                  must be auditable and must have spot audits.  Clearly an unauditable process is unacceptable.  Plus you actually have to do the audits for them to mean anything.

                  The problem with e-voting is that the current technology is pretty poor and is closed to protect the interests of the company rather than the interests of the people.

            •  the technology is promising here in two respects. (0+ / 0-)

              now, don't get me wrong, Diebold particularly -- but many other major voting technology companies, too -- are suspect.  The lack of common-sense which is readily apparent in how these systems are sold and in how legislating bodies around the country ignore obvious risks and don't insist on proper safeguards -- this state of apparent collusion, putting business or efficiency ahead of the sanctity of individual votes -- all of this is profoundly troubling.

              Anyone who votes on a Diebold machine with no voter-verified and hand-recountable paper backup this November should be worried, as should all Americans.

              But while the current practice of e-voting is an abomination, so too were Ohio punchcards, so too was the myriad of disenfranchisement in Florida.  There are people who want to put their interests ahead of the will of the voting public.  People like Jeb Bush, Katherine Harris, but there are many others.  And they have thousands of schemes and techniques, ideas, to make this happen.

              With that being said, I endorse the comments of others here, for example, this by puppethead and this discussion.  e-voting has the promise of more accurate votes, less voter confusion, and efficient and expidited public results for all but the closet or contested elections.  For those close and controversial ones though, any good system would have to include multiple checks including voter-verified, optically scannable and hand-recountable paper backups; and also including reporting from individual precincts as well as regionally and centrally, with verification by reps from all the qualifying parties, so that the numbers are publicly cross-checkable.

              We don't need to be afraid to use this stuff, but we need people who actually care about improving our elections rather than self-interested crony-politicians and corporate shills to think through the implications and design a system which is reasonably safe and secure.  With the proper measures in place, this could be more secure than any current methods or even hand-written-affidavit voting.

              •  Please note, I am a 'computer guy.' (0+ / 0-)

                I'm not "afraid to use this stuff".  For voting systems (the simplest of all math and counting problems), computers are unnecessary and do not provide better accuracy.  And "efficient and expedited public results" are not high on my list of requirements for voting systems.

                Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                by sxwarren on Wed Sep 20, 2006 at 08:48:14 PM PDT

                [ Parent ]

                •  possibly more accurate (1+ / 0-)
                  Recommended by:

                  than either punchcard or optical-scan (fill-in-the-bubble) systems, which already involve machines.

                  Again, I believe those who've certified current lines of Diebold e-vote machines for this election have (at least) negligently infringed on the voting rights of their constituents.

                  But it's not really that hard to see how these tools can be helpful.

                  Clearly, speed of election returns is not even in the same ballpark, value-wise, as issues such as accessability, accuracy, and security of the voting system.  But those risks need to be addressed no matter what media is chosen for taking votes.

                  I only advocate that e-voting, if ever to be used, must not be so until we can adequately address those issues in light of the higher magnitude of risk involved.  And it seems to me that some physical ballot, probably on some kind of paper, which would be an indisputable record of a single vote and couldn't be transmuted into a vote for another candidate, is an important part of the system and one should be collected for every electronic vote the machine records.  A voter could check the printout for accuracy, compare it to the cast vote on the machine's screen, and then leave one copy in a collect box and take one with her.  Samplings of the physical ballots could be evaluated to uncover errors in the electronic totals, and if anything is uncovered, or in close races, or whenever there's a contest -- then all of the physical ballots could be scan-counted by other machines, or hand-counted.

                  But having the system which performs the initial count also involved in aiding the voter to vote correctly, so the system can be sure to understand the voter's intent, can help reduce a source of error which current punchcard or optical-scan ballots have yet to resolve.

                  And, IF we can make e-voting adequately safe, secure, and trusted; the accuracy, efficiency, and speed of election results would be nice.  Not worth giving up democracy for, no, but if safe, nice.

                  And Urban local, State and Federal voting is hardly a simple counting problem.  Indeed it's enormous and can be complicated (we've got ranked choice for local SF elections, which really can make a mess of things), which is partly why we can't get clear and undisputed results.  I hope more thoughtful and careful minds make progress over this issue, like, now.

                  But I don't know if our positions are so different to keep arguing over.  I want to make clear that no voting is acceptable unless hand counting of physical ballots is a possible avenue to settle ambiguities or contests.  Your position seems to be that no voting is acceptable unless automated electronic processes have no role in counting the votes or determining the totals.  That position I would find simplistic and not completely rational.  Perhaps you instead make a smaller point: that since we need a verifiable physical (paper) backup anyway (and I argue we do), that it is unnecessary and maybe even iniffecient to introduce machinery and automation into the process.  While e-voting clearly is NOT necessary, I also don't think AVOIDING computers in all parts of the voting and counting system is necessary (theoretically, of course.)  Which is why I allude to the fear-of-technology.

                  Since I don't fear computers, I am open to and willing to explore ideas about how they can be further utilized to aid us in voting and counting elections, so long as a physical paper vote is produced for each voter, so that the internal computer numbers would not be the final word.  Your arguments here seem to have gone beyond my insistance that a physical ballot is a necessary component of voting.  Instead, I feel you've been trying to argue that we should not use computers at all in this process -- that they offer no advantages and only and always bring unacceptable and un-securable risks.  Forgive me if I've misread your position, but if I have it right, it does seem to come from a ground-point of technology-avoidance.

                  •  You have my position mostly correct, but (1+ / 0-)
                    Recommended by:

                    you've missed the mark on how I've come to it.

                    I am no technophobe.  I began hands-on work with computers (and some of the first optical scanners) in 1968.  Since then, I've studied procssor and operating system design, dabbled with writing and fixing code in several programming languages, held positions in technical support and database administration, managed enterprise-wide rollouts of new applications and new hardware for major corporations, and participated in the development of new applications from the requirement statement phase.  Lately, I've been using some very sophisticated software tools to reveal, defeat and remove malicious software that employs operating system kernel-level "rootkit" techniques to successfully hide itself from even the best available antivirus programs.

                    Through my professional experiences, I've developed a deep appreciation for how the power of computer systems has vastly augmented our ability to quickly perform complex calculations on huge data sets, whether it be rendering the full-motion graphics of computer games or confirming the existence of Global Warming.  I've also developed an understanding that computer technology is just now on the cusp of adolescence.  There are wonders yet to come so incredible that even Sci-Fi writers have yet to explore them thoroughly.  Computer-enabled telekinesis of a sort is already here, though in its infancy, and computer-enabled telepathy is now conceivable.  But the advent of each new wave of computer technology wonders, just as with, for instance, advances in chemical technologies such as pesticides and drugs, is accompanied by crucial ethical and pragmatic issues that must be addressed, preferably before such wonders are wantonly applied.  And, with each new technology advance, what actually transpires at the core of systems becomes more obscure and less-well-understood by an increasing percentage of people, even among those who deal with the technology professionally on a daily basis.

                    Through my experiences, I have also witnessed countless instances of computer technology inappropriately applied - as a solution in search of a problem - by well-intentioned folks who, dazzled by the technology and eager to apply it to everything, haven't spent sufficient time and effort on outlining the existing process and properly enumerating and prioritizing the constraints and requirements of the system that they are attempting to duplicate with a computer system.  This is the "requirement statement phase" that is all-to-often given short shrift.

                    I firmly believe that the very concept of e-voting is one of these instances.

                    On the issue of transparency, please consider one more practical, social example.  Even a simple desktop computer can, hypothetically, select a subset of numbers from a given set of numbers with a much greater "degree of randomness" than can be achieved by a system that selects that subset using, say, numbered ping-pong balls bouncing around on air currents in a closed plexiglass chamber.  However, people who play Powerball would scream bloody murder over the potential for fraud if it was suddenly decided that the process for selecting the winning numbers were to henceforth be conducted in the "black box" of an RNG on a computer.  Because the process would then not be transparent to them.  And yet, were are apparently more than willing to set aside a similar requirement for transparency in the system we use to select the next person to take possession of the nuclear football or the next set of people who will decide whether or not to authorize our next President to use military force against Iran.

                    How irrational is that?

                    Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                    by sxwarren on Thu Sep 21, 2006 at 06:57:40 AM PDT

                    [ Parent ]

                    •  my roomate is on Party Poker right now... (1+ / 0-)
                      Recommended by:

                      with 60,000 others, who, at this very moment are at the mercy of a private corporation located in havens in Gibralter and Native Tribal Canada.  They are all playing poker with real money, even though there are no cards, no chips, no tables, no proof that your opponants are actual real people. There's no transparency, just trust that fair RNG programming is running it all, and yet they all come to gamble, hundreds of dollars or more.  And that's in the middle of the afternoon, on a workday, on just one site.

                      So I'm not sure your powerball illustration is all the way right.

                      Sorry, I didn't mean to imply any doubt as to your credentials, history of working with computers.  Please don't take offense.  You clearly care alot about this issue, and it's one of our most important battles today.  We've lost all hope when the will of the American people isn't even measured, maybe not even approximated in good-faith!

                      Remember, voting is a black-box exercise no matter how it's done.  Most people leterally put their votes into opaque boxes, sometimes into mailboxes, and hope they get counted.  Ultimately there will have to be a measure of secrecy because voting is anonymous, and anonymous voting I believe IS essential.

                      What I fear with e-voting is the potential for fewer people to make larger frauds on the count, with less of a means to uncover their fraud.

                      But I do think, for example, that the Optical-Scan systems we use in SF now are pretty good.  The paper ballots can be scanned again and again by different machines, and can be fairly easily hand-counted as well.

                      You haven't spoken specifically about optical-scan or punch-card voting, although I've been using them for comparison to a possible electronically-assisted voting system.  I assume though that you prefer only human-counting of all ballots.  

                      I however don't see the advantage of trying to tabulate millions of votes only by people.  There is every reason to assume that totals would change from count to recount to additional recount, and huge amounts of time and resources would be expended as a matter of course.

                      One nice thing about optical-scan is that the ballots can be quickly counted and counted again, in theory even by different sets of machines created and programmed by entirely different people, and hand counting is still available to get yet another alternative look at the data.

                      If you were to convince me that only handwritten and handcounted ballots must be prefered to optical-scan (smartly applied... which may not be how currently applied in some Jurisdictions), then I would withdraw any support in the potential of electronically-assisted voting.

                      But I see with optical-scan clear advantages, and so long as manual recount is used effectively to double-check, the risks are not then out-of-bounds or even much out-of-line with risks always associated with anonymous voting and difuse vote-counting.

                      And If I'm correct that optical-scan may be a fair and useful tool to aid us in voting, then it's not hard to see the potential for electronically-assisted voting.  After all, the voting machine could verify and re-verify with the voter that the vote recorded (thus far) matches the voters intent.  It could warn or prevent a voter from overvoting.  It could note and warn to prevent inadvertant undervoting.  The printed ballot could be designed for easy and clear scanability (not relying on an indivdual voter to do the bubbles correctly themselves) as well as obvious readibility (voted-for candidates appearing in large bold text next to each race.)

                      With electronically-assisted voting, we could have more secure data because we'd have more crosschecks on how it is viewed.  The machines would electronically tally votes, the ballots could be scanned for a different count, and hand-counts would allow a third look.  The law would allow the handcounts to be the ultimate arbiter in the event of problems or discrepencies or challenges, and the feedback loop would make for improvements to the system with each succeeding election.

                      In such a system my faith in having my vote counted would increase.  

                      The advocates for paperless voting are creating a monster, no doubt.  It is only one sad factor that they've so perverted what may be an excellent excellent idea.

                      •  No offense taken (1+ / 0-)
                        Recommended by:

                        regarding the credentials thing.  I only state them to illustrate that I'm no more a technophobe than are the nuclear physicists who campaign against the continued existence of nuclear weapons.

                        I disagree that voting with paper ballots is as much a "black box excercise" as computer voting, except metaphorically.  In our little town, voters get handed their ballot(s) printed simply and unambiguously on sheets of paper and go into little private enclosures to mark them and fold them before depositing them in a large locked box (get this - many folks don't even bother to use the booths).  When voting is completed, the box (constantly monitored by a number of my fellow voters to preserve the chain of evidence, so to speak) is opened and the ballots counted by hand (several times by different people) in full view of anyone who wants to watch.  This recording and counting process is completely transparent, meaning comprehensible, to everyone who participates. The primary objection I hear to this system is "Well, that's fine for a small town with a thousand or so voters, but it doesn't work in urban areas where there may be ten thousand voter in a precinct."  That only tells me that we need smaller, more numerous precincts.

                        And the final, agreed upon, local tallies are posted in full public view at the town office before being transmitted to an aggregation point.  The aggregate totals, as well as the individual precint totals, are published in a public record.  If these published totals for a precint differ from the reported totals posted locally, the local precint election officials (actually, anyone in town) can raise a red flag.  There is no one small group of people attempting to tabulate millions of votes.  

                        Now, I have no problem with using optical scanners to generate redundant local counts that are immediately verified locally by hand counts, as long as all of this is performed in full public view and as long as the formatting requirements of the scanners don't make the ballots themselves more ambiguous or reduce human readibility or complicate the process.  And I appreciate your suggestions regarding "electronically-assisted voting" although, to some extent, this still strikes me as getting into that "extremely expensive writing implement" area.

                        What I object to, specifically, is any system in which an individual's selection is recorded and tallied within computer software (an invisible process no matter how it's characterized or it's putative accuracy proclaimed*) and becomes the official count, with public hand-counts relegated to after-thought or invoked only if there are suspected problems.

                        • If you're so inclined, please review my discussion threads with "william shipley", someone who apparently knows much more than the average citizen about computer systems, regarding the assumed "magical" accuracy of computer systems, even those not intentionally perverted.

                        As far as your roommate gambling on a poker site, all I can say is "YIKES!" if for no other reason than I have yet to see a computer that's been used on one of those sites that wasn't covered in adware, if not spyware.  It's certainly not a gamble I would make and it's trivial compared to gambling the outcome of a national election.  In any case, people who throw half their weekly paycheck at a one-time draw against 1:146,107,962 odds are a completely different breed from those who play poker where at least some skill is involved.

                        The thing about all gambling is that the house always wins.  And that's the real gamble we make when we allow computer system recording and tallying of our votes to be the official count.

                        Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                        by sxwarren on Fri Sep 22, 2006 at 05:53:43 AM PDT

                        [ Parent ]

                        •  Could I possibly ask you to read this? (1+ / 0-)
                          Recommended by:

                          Now that I'm done, having come back to this post time and again today, can I really post such a monster?  I hope you find it worth your while...

                          • all who campaign against nuclear weapons are right.
                          • some who campaign against nuclear weapons are technophobes.
                          • even some nuclear physicists may be technophobes.
                          • perhaps some nuclear physicists have MORE reason to be technophobes than us laypeople. (laypeople regarding nuclear-physics, of course.)

                          Perhaps I need to flesh out a bit of my background here, because I'm reading alot into what you've said about yours, coupled with your thoughts on this topic, and perhaps it's not fair unless I open myself up to the same.  I think it's pretty interesting to see how our experiences shape our perspectives, anyway, so...

                          I go way back with computers and computer-programming as well, although in ways very different than you.  I was not around when those first punch-card and magnetic tape systems came into use.  I'm 29.  Computers have been in my life as far back as I can remember.  I could program in BASIC on an Apple IIe in 1st grade, and won academic awards for computer programming by 4th.  As a child I was in high-demand around my family and older family friends, always shuttled around so I could program VCRs, Home-Security systems, Phones.

                          In college, first semester, I took a random seminar about philosophical questions and issues related to artificial inteligence, which sparked in me a fascination that has taken me on a life-course I never expected.  I majored in philosophy, studying a wide range of historical writers of course, but with a specialization in Logic.  I particularly studied symbolic-logic systems and manipulation, including Turing's concepts which led to our first computing tapes.  Indeed, I had exhausted all the 300-level symbolic-logic related philosophy courses by my junior year.  So to further my studies, I undertook a minor in Computer Science, during which I learned circuit-logic, programming, software architecture, and algorhithym analysis at multiple levels.  By the end of college, I had come full circle, only now I was studying AI from the inside, actually designing systems which could learn or adapt, wheras before I had studied only the abstract theory and philosophical paradoxes behind the field.

                          After a bit of post-college professional training, I embarked on a six-year career as a web-designer/developer/programmer/webmaster (you know the type).

                          My whole life I've been in crowded urban environments, growing up in the Urban/Suburban mix of north-shore Chicagoland, college in central Illinois, back to Chicago, and now SF for several years.  

                          My first voting experience was in college in Champaign, IL.  Incidentally, this was also my first experience with vote suppression.  Many of the city and county elected officials had been on record for years, expressing their dissatisfaction at a system that requires them to allow College students to vote in their community.  According to these officials, the college students are not permanent community members, not invested enough in Champaign to deserve local representation.  After all, they go home every summer, and after four years or so they're gone!  Never mind that the entire economy of Champaign depends on the University and its Students; never mind that while individual students come and go, student concerns and interests persist; never mind that the student population of Champaign equals the non-student population; never mind the law.  What really was happenning was, the business class, which lived on the outskirts of town, and profitted off the university and it's students -- well, they were all conservative republicans.  The student population, of course, included a very large and vocal liberal contingent, and generally voted Democratic.  The politicians were smearing the students as being idiots who don't care when they vote -- but really they were just trying to protect their own local power.  (Who would've guessed, right?)

                          Anyway, these politicians didn't just speak out against the students, they did all that bad stuff we always hear about -- misleading registration information (info cards telling students they must vote absentee from where their parents live); strawman phonecalls misstating the date of the election; and insufficient precincts, booths, and ballots around campus, causing me and hundreds of others to wait for 3 hours or longer in line on election day, skipping all our classes so we could vote (some kind of punch-card system if I remember correctly).

                          By my last election in Champaign things were better, but only because the Feds had been tipped off, and Clinton's election commission and FBI were in town, monitoring everything, and a federally-brought suit against our chief elections officer was already under way.

                          More recently, as a campaign volunteer in a local race in SF, I noticed how threadbare the monitoring process can be.  I attribute this to a lack of interest in our elections -- not enough volunteers.  I was working GOTV, periodically checking the precinct list to see who was registered but hadn't voted yet, then trying to track down those people (at least, those among them who we thought would vote on our side!) and giving them a friendly reminder and some encouragement to vote.  

                          Our campaign, like the main competition, put most of their meager volunteer resources on this kind of GOTV task.  But neither side had enough volunteers to make full use of their legal allowance to monitor the polling places.  I believe local election law allows one or two partisian monitors from either side to sit at the polling places and make sure things are being done properly, but absent them, the place really relies on the poll-workers who volunteer with the county to actually conduct the polling-place activities.  These people are officially non-partisian, but they could be anyone.  Our campaign had one volunteer who'd drive from location to location, checking things out more or less at random, and taking calls from the GOTV volunteers who may have noticed something when they checked-in at their precinct location.  The only partisian monitor I came across at the precinct I was working was with the Republicans, which was ironic since their candidate got about 3% of the vote (the Supervisor elections in SF are non-partisan, with multiple candidates playing for the Democrats, which is basically everybody in town).  He looked bored and was reading a book every time I came in.

                          The past several elections I've voted absentee.  This is great because you can sit down on a convenient evening with a glass of good tequila and all the Voter Information Guides and newspaper clippings and the internet at your fingertips, and really try to make some informed decisions about the myriad of State and Local Ballot Initiatives and all the races and make an informed choice in a familiar and comfortable place.  And by the time you finish voting you've got a good little buzz going and you meander off to the mailbox and feel like a good little citizen.

                          Of course, it seems many times out here the election is followed several days later by boxes of uncounted ballots washing ashore along the bay somewhere.

                          To finish my bio quickly -- I gotta get to my substantive points soon and hope this post won't crash Kos's servers :) -- after the Dot-Com bubble burst I reassessed everything and went to law-school, then took the bar.  Now I'm an unenployed lawyer with too much time on my hands for blogging.  

                          My techie-circle of friends includes several programmers and systems-architects more advanced than I ever was, although I generally have been comfortable talking most kinds of shop with them relating to programming and security issues.  My best friend for the past 10 years (he posts around here rarely, or at least has in the past, under the name misinformed) has been a big player in computer security from many diferent angles, having gone so far as to architiect and run systems for Sony, securely delivering popular music online.

                          Okay, here's the point.  We have different perspectives on technology, you and I, and we have different perspectives on elections.  I certainly don't mean to brand you as a shack-in-the-woods technophobe (sometimes I long for that kind of life myself) but I do think, on the index of technology-faith/technology-fear that you are demonstrating more fear and I more faith.  

                          Of course, I believe my position is in the right place in the middle of the spectrum, and that your position further down towards the anti-tech side is unwarranted.  I likewise assume you find my position to be off-kilter -- that my faith is akin to believing in magic -- and that your position is actually the more balanced and accurate one.

                          Let me highlight a few specific things you've said which I think are incorrect or unfair.  I'll start with what was perhaps the central point of your diary:

                          E-voting systems are inherently NOT TRANSPARENT!!

                          Actually, while E-voting systems currently marketed are not transparent, they're succeptible to fraud, and they're controlled by untrustworthy people -- these facts have little to do with anything inherent in computers or technology.  It's the implementation which is so flawed.

                          Imagine the hypothetical voting system I outlined in my previous post, with electronic voting machines assisting the voter in making choices, then printing two copies of ballots, which are verified with data displayed onscreen, before one copy is left in a ballot box and the other kept by the voter.  

                          Now lets take it a step further.  We can digitally scan all the ballots in the ballot box, creating a digital image of each.  After each election the electronically created totals would be available immediately, and soon thereafter a database of each scanned ballot would be available online, and searchable by state, township, precinct, down to the individual ballot box.  Any member of the media or the public, anywhere in the world, at any time, could scan and search the database and research whether the scanned ballot images are matching the internally-created numbers for that box/precinct/city/etc.  Additionally, any member of the public may (if they choose to lose their anonymity) come forward with their receipt if it doesn't match the scanned image available online.

                          Now I'd call that transparancy.  Not just are the votes counted in a big room in a small town for the town to see, but everyone in the country, indeed the world, could count for themselves.  We would call the ballots in the boxes (locked away after scanning) the OFFICIAL ballots and rely on handcounts of them when necessary, but with the right redundancy and transparency and security and (very important) with the right LEGAL FRAMEWORK to handle close contests and challenges, we'd be able to rely on the efficiency of the machines most of the time.


                          ...but then you've simply had the voter use a really fucking expensive pencil to mark his ballot.

                          It's not hard to see this statement as expressing an anti-technology sentiment, and you've returned to this point in your post above.

                          A computer IS just a really fucking expensive pencil.  A fancy pencil.  A pencil which is better than a pencil.  

                          A pencil is just a tool and a computer only a newer tool, developped and used for the advantages it presents.  A pencil put to paper can write correspondence, account for money, organize appointments, draw pictures, and perform calculations of all sorts -- but not the way you can with computers.

                          So you're right, in a sense, but your disdain is unwarranted.  I've outlined already some advantages which can be utilized with proper application of technology to voting (transparancy, redundancy, efficiency, multiple-step voter validation to reduce incorrect or confused votes and to reduce uncountable votes due to voter-error) and further illustration follows.  These advantages are worth pursuing, imo.


                          You might suggest that those of us who CAN read code can attest to a system's accuracy and security, but what if other experts disagree?  How is it decided which experts are correct, are telling the truth?  Who does this deciding?

                          This argument has some merit when arguing against paperless voting, but is hardly an issue in arguing for machineless voting, against the kinds of systems I'd like to see.  In my system, any problems with the code would be reflected in discrepencies with the results.

                          And anyway, a team of intelligent academics and scientists creating a simple open-source operating system and designing counting and aggregating algorithyms -- teammembers may come to differences, but it would take a grand conspiracy to get a faction large enough to credibly argue they were creating a non-fraudulent system when in-fact they were.  It's much easier for a private corp like Diebold to spin bullshit like that, but if we put the best and the brightest on it, manhattan-project-style, there is little doubt they could develop sound, simple, workable code.  And make it open to outside critique.

                          Your argument over that point seems to tap into liberal disgust at those few academics and scientists who proclaim publicly there's no evolution or global warming or whatever.  Unlikely such b.s. could effect a project such as this when properly conceived, but that's why we have all the redundancies, and the paper, just in case.


                          If it ain't broke, don't fix it.

                          Some things just need to be left alone.  Voting is one of those things.

                          The primary objection I hear to this system is "Well, that's fine for a small town with a thousand or so voters, but it doesn't work in urban areas where there may be ten thousand voter in a precinct."  That only tells me that we need smaller, more numerous precincts.

                          I put these quotes together because, while your small-town voting system "ain't broke", my experiences have shown me broken elections time and time again.  So I don't wish your town to jump to a worse system, but something needs to be done in our major cities, from Cleveland to Pittsburgh to Miami to even our little left-coast metropolis, San Francisco.  And dropping the current systems for a all human-powered hand-eye-paper system could be a disaster.

                          You speak of more precincts, but it's not as simple as all that.  Where are all the volunteers?  Who will open their house and sit all day at the polling-place, where are all the partisian volunteers protecting the integrity of their party's voters' votes at each polling place?  Parties send volunteers into crowded, contentious cities, sometimes people coming from different states, because the political infrastructure within the city is insufficient.  But then who's monitoring the elections in the places those people left behind?

                          The solutions to this may require real change.  We can talk about a national holiday, so that all business close and nobody has to go to work on election day.  We can talk about mandatory electoral service, getting a notice every few elections like we get jury-duty notices.  But we also can and should talk about using better tools, ie. technology, to ease the human burden -- and perhaps increase accuracy and efficiency at the same time.

                          I see in these points of yours a bit of a nostalgia and a cry-out for classic, small-town values.  That's nice, but it doesn't really solve the election problems I've seen in the city.


                          ...please review my discussion threads with "william shipley", someone who apparently knows much more than the average citizen about computer systems, regarding the assumed "magical" accuracy of computer systems, even those not intentionally perverted.

                          I have been following the threads here closely, and I agree with much of what william shipley has said, particularly with respect to the human potential of error in voting and vote-counting, and also regarding the importance of auditing, tracking, and comparing the datasets in an election to root out errors and fraud.

                          I do appreciate the point you've made about banks still counting the drawers manually.

                          From my own studies and experience, I am convinced that, abstractly, philosophically, a computer simply does not make mistakes in doing what it's told.  It may sometimes appear to make mistakes, but this is only because the machine is physically not working, or because the machine has been programmed to operate differently than some have been led to beleive.  So machines themselves can't steal elections, although people may certainly try, by sabotoging the machines but also by other methods.

                          I knew already in first grade, that if I told my Apple IIe to count up the numbers from one to a million and display them on the screen as it counted, that that is what it would do.  Maybe the electricity would go out and cut into the process, maybe the program contained an error leading to no action, or jumbled results (printing absolute gibberish instead of numbers), maybe the cable to the video card would short and the screen would go blank.

                          But the machine would never skip 315,005 on it's way to a million, it wouldn't count 555,555 twice, it wouldn't try to trick me by jumping to the end of the count before counting all the early numbers.  The machine's ability to count, when compared to my own, is like magic.  It's people telling machines to act in incorrect or even sneaky-like ways which makes them appear to screw-up tasks like counting, but from the machine's perspective, it is doing everything exactly as it is told all the time.

                          Which is why I hate your using as an illustration the fact that our Windows boxes are of poor quality and crash all the time.  The software packages involved are enormous and rather poorly designed, but the machines are still exactly executing the code they are given.  It's the human error which makes the machines crash, not anything to do with inherent tendancies of computers to "make mistakes"

                          misinformed (the buddy I spoke of earlier) has worked with actual songs, whose every note and nuance can be digitally copied and distributed thousands of times around the world, with each copy being exactly identical to every other copy.  The files even include their own map to self-check themselves and ensure the copy is perfect.  No human has that kind of preciseness.


                          So, in summary,
                          if you came to computer technology when it was in its infancy and you were already mature, you may indeed have less comfort with computer technology than some members of younger generations for whom computer technology is second nature.

                          This is especially true since, as a professional in development, we all know how none of this stuff EVER seems to be working.  Indeed, we've kept jobs for years just trying to constantly fix stuff as we come to notice how badly it's broken.  This frustration may shake the faith, especially if one can remember a happy life when none of those hassles existed.

                          But it's human error (or perhaps malicious human mischef) that causes every one of those mistakes.  The computer didn't do it on its own.  And human ingenuity works to solve the problems, too, just as always.

                          Also, I find you have every right to speak protectively about the voting system in your small town, but your advice isn't practically applicable to many big-city urban voting problems of today.

                          Third, WE MUST FIGHT AND SPEAK OUT AGAINST THE STEALING OF OUR ELECTIONS.  WE MUST HOLD ACCOUNTABLE, SEND TO JAIL, THOSE PEOPLE WHO WOULD STEAL OUR ELECTIONS.  And punish anyone who wrecklessly or negligently makes or allows it to happen.  

                          This includes elected officials who allow nontransparent, privately-controlled systems to produce untestable and unverifyable results. Any acceptable system must include a physical paper trail of ballots and receipts. And a proper legal framework for reasonable challenges to be investigated and settled, with hand-counts when appropriate, before final vote counts are certified. This is the most primary practical point right now.

                          Fourth, don't discount the potential to use computers to make voting better.  

                          Stepping bravely into the future is really all any of us can do.

                          •  Wow! Very well stated. (1+ / 0-)
                            Recommended by:

                            While I don't cmpletely agree with all of your points, I really appreciate the time and careful thought you've spent on your "monster."  ;-)  I'm obligated to spend s similar effort on digesting this before I respond, but at the moment, I'm obligated to go flip the meat-like substance that I'm frying for breakfast.

                            I't started to rain here and it's predicted to last all day.  Since that will cut into gardening activity, I should have some time later today.  Stay tuned, please.  I find this dialog very enlightening.  

                            And Thanks!

                            Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                            by sxwarren on Sat Sep 23, 2006 at 06:38:42 AM PDT

                            [ Parent ]

                          •  likewise, and thanks! n/t (0+ / 0-)
    •  yes but (2+ / 0-)
      Recommended by:
      sxwarren, kraant

      I think telling people to vote absentee can be counterproductive, because it may give them a sense of security when there is none.

      until all the votes are counted, a few absentee ballots here and there won't do jack.

      Stronger than all the armies is an idea whose time has come. - Victor Hugo

      by racerx on Wed Sep 20, 2006 at 02:15:32 PM PDT

      [ Parent ]

  •  Over here in Ireland they have spent 60 millions (4+ / 0-)
    Recommended by:
    racerx, sxwarren, corvo, kraant

    on E-voting machines and after a number of blatant errors the government has scrapped them. A few red faces but the point is that they don't work. I read somewhere yesterday that a mini-bar key can open a voting machine. WTF? are on the March, boy George!!

    by Asinus Asinum Fricat on Wed Sep 20, 2006 at 01:58:28 PM PDT

  •  Sarasota County forced to have voters (2+ / 0-)
    Recommended by:
    sxwarren, kraant

    decide Nov 7th to keep the touch screens or scrap them for optical scanners, which produce a paper ballot and subject to random hand-counted independent audits.  A grassroots organizations called Sarasota Alliance for Safe Elections (SAFE) was formed early this year to gather petitions from citizens for a proposed amendment to be put before the voters in November.  Over 14,000 petitions were signed by both parties, but the Sarasota County Commission refused to put the proposed amendment  on the ballot, stating that parts of the amendment were unconstitutional.  SAFE had to file a lawsuit and the judge ruled in SAFE's favor.  The people of Sarasota County will decide Nov. 7th.  Check out the website at  SAFE is now in the process of raising funds for the attorney fees and to educate the public on this issue.  There is a lot of confusion and the opposing side is pushing the "high" cost of replacing the touch screens even though a study has been done to prove the optical scanners will be cheaper in the long run and costs to change to optical scanners will be recouped in 2 years. This is a hot issue in Florida and if SAFE is successful in Sarasota, it could lead the way for a statewide change to eliminate touch screens.  

  •  Paper ballots and purple fingers! (3+ / 0-)

    I'm jealous of the Iraqi electoral system.  Thanks, Rethugs.

  •  Fraud can occur by any voting method (2+ / 0-)
    Recommended by:
    kraant, sfRenter

    Can fraud occur by e-voting? yes. One could manipulate the program to miscount the votes. On the other hand, no matter how you vote, fraud can always occur.

    In Snohomish County, Washington, 20% of votes casted by absentee ballot was invalidated. Snohomish is a pretty big county, containing the northern suburbs of Seattle. In fact, the controversy over the Gregoire-Rossi governorship race in 2004, centered on how to count absentee ballots.

    The Bush-Gore controversy in Florida, centered on punch card ballots. When should a vote have counted? Remember hanging chads?

    Some people argue that we should refuse to cast e-ballots. They reason, the less e-ballots casted, the less fraud possible. However, one could program a computer to count a billion votes even if no one even casted a single vote on the computer. Thus, fraud can occur even if no one decides to cast an e-ballot.

    But before we try to abolish e-voting consider this. Casting by e-ballot is actually quite simple. The error rate is much lower than paper ballot voting. In Washington State, 5-10% of voters made errors voting by absentee because they mistakenly voted in both the Democratic and Republican primaries. E-voting simply won't allow that type of mistake to occur. Moreover, errors on optical scanners occur when you don't color in a circle quite right, or you accidently make a mark outside of a circle. Also, electronic voting is very accessible to disabled people.

    •  But it's easier with e-voting (4+ / 0-)
      Recommended by:
      sxwarren, corvo, kraant, BlueTide

      To make a significant difference in a state-wide election where there are paper ballots that are hand-counted in each precinct, you would have to corrupt thousands of local poll workers, and there is likely to be rather clear evidence of what you've done.  To do it with e-voting, you only have to corrupt one (or at most a handful) of programmers.  

      If you try the former, it's very expensive, somebody is likely to talk, and the fact of cheating is likely to be obvious to even rather casual observers.  If you try the latter, it's going to be much cheaper and much easier to conceal.

      Is it likely to be possible to corrupt a few precinct workers in one or a handful of precincts?  I don't doubt it, but that's not going to change the outcome of any except the closest elections -- and you don't know exactly how close any election will be until it's over.  On the other hand, with e-voting, it would be easy to shift 5 or 10 percent of the vote on a statewide basis.

      •  I was a poll judge (2+ / 0-)
        Recommended by:
        kraant, sfRenter

        In our machines, the voting machine created a printout that the voter could check with the computer screen. If the printout and screen did not match, the voter could ask the poll judges to shut the computer down and issue them a paper ballot. Thus, the type of fraud that I explained above, is quite difficult, because its too easy to notice.
        However, I did discover that it is easy for a poll worker to commit fraud. The poll workers took turns taking lunch. When I was alone, I could have easily voted a few dozen times. However, I probably would have been caught because the machine count would not have corresponded to the check-in tabulation. But if I were determined enough, may be I could pull it off.

        On the other hand, if I were counting paper ballots I could count: 1, 2, 3, 4, 11, 12, 13, 22.... You get the picture.

        •  Umm, what appears onscreen and on the printout (0+ / 0-)

          don't mean squat.  Ordinary, everyday computer viruses screw with the human-readable output as a matter of SOP.  

          A thousand people vote on a given machine - 489 for candidate "A" and 511 for candidate "B".  Each voter verifies that his/her paper receipt and the screen match the way that they voted.  The software in the machine switches 15 votes, so that the election comes out with candidate "A" winning by 504-496.  Unless there is a seperate hand recount of the votes from that specific machine using the receipts that each voter has personally verified, there is no way to tell that there was any fraud or that the results are incorrect.

          Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

          by sxwarren on Wed Sep 20, 2006 at 03:27:35 PM PDT

          [ Parent ]

          •  So you spot check (0+ / 0-)

            You should not find any discrepancies.  You need to spot check a significant number of random machines.

            •  Spot checking and statistical sampling are great (0+ / 0-)

              if you're doing quality control on automobile parts.  Not good enough in the system we use to give power to a person who may or may not want to launch nukes against Iran.

              Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

              by sxwarren on Wed Sep 20, 2006 at 08:50:28 PM PDT

              [ Parent ]

              •  E-voting is potentially much more reliable (0+ / 0-)

                The current system involves a lot of built-in errors.  Manual processing leads to variations.  Every recount they do comes up with a different answer.  If manual counting were the most realiable, that is how the banks would be handling their accounts.

                Spot checking is to protect from systematic fraud.  The devices would count accurately and reflect the intention of the voter.  People do lots of strange things to paper ballots, check boxes and write in names, vote for more than one etc.  A computer system can get a clarification of voter intent while the voter is standing there, not by later holding a punched card up to a light and looking for dimples.

                •  I began working with computers in 1968, (0+ / 0-)

                  wiring program panels for data-procssing equipment and operating a keypunch.  And I can say with great certainty that neither God nor von Neumann ever intended for humans to punch data cards by hand.  And I am religiously opposed to punch-card voting systems for that reason.

                  As regards your comparison to banks, your implication is that banks were hopelessly inaccurate and unreliable before the advent of computers.  Actually, the mere humans who handled the ledger books, BC, were very good at what they did and yet always had their work continually audited by others.  Taken in the aggregate, the calculations that are performed by banks these days are far more complex than simply counting votes and the adoption of computer systems by banks was not driven by a demand for greater accuracy but for greater speed and lower labor costs.  And yet, in all banks, the cash in the till is still counted by hand - to verify the accuracy of counting machines.

                  There is a widespread assumption on the part of folks who have never seen the "innards" of memory registers, CRC checks and software data buffers that "computers" are somehow magically more accurate at simple counting than humans.  This is simply not the case.  Remember this the next time your Internet Explorer locks up or your Windows XP "blue screens" - the root cause is that the result of some simple binary digit calculation was incorrect.

                  Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                  by sxwarren on Thu Sep 21, 2006 at 04:00:53 PM PDT

                  [ Parent ]

                  •  Well actually they did (0+ / 0-)

                    Hollerith introduced the punched card for the 1890 census.  It was punched with a manual punch like a train conductors.  So actually the idea of machine punched cards is a modern innovation!

                    They may have been able to balance the books by hand but a very large number of people were involved and they weren't amateurs who would only do it for a day every couple of years.  And yes, they double count the cash by hand because machines that count pieces of paper have error rates -- which is why I don't like voting plans that print out a piece of paper that is later scanned.

                    My involvement with computers began slightly after yours, 1970 but lower technologically as I used paper tape.  Cards were so nice of an improvement when I got to use them!  And, as you well know, XP does not crash because it fails to be able to add properly.  It is hard to imagine that a computer could run at all if it's ability to add were not flawless.  Computers are much better at adding than we are.  We do this funny thing with memorized sounds and addition tables etc.  Binary is so much cleaner.

                    •  Please don't twist my words. (0+ / 0-)

                      My statement regarding application problems and XP blue screens, that

                      ...the root cause is that the result of some simple binary digit calculation was incorrect.

                      is not even close to being equivalent to saying that XP crashes because it fails to add properly.

                      While it may be hard for you to "imagine that a computer could run at all if it's ability to add were not flawless", in point of fact, software (and even, occasionaly, hardware) screws up calculations all the time.  Both hardware and software are designed by humans.  What is it that leads you to believe that they would be inherently "flawless"?

                      "Computers", whether you're referring to hardware or software or "computer systems" which implies the integration of the two, do not add "better" than humans, merely faster.

                      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                      by sxwarren on Fri Sep 22, 2006 at 04:23:35 AM PDT

                      [ Parent ]

                      •  Don't be silly (1+ / 0-)
                        Recommended by:

                        Add up a couple hundred 6 digit numbers by hand.  How many times will have to add them to make sure it's right?

                        Once for a computer.  In 35 years I have never detected a mathematical mistake in any program I was testing.

                        And the blue screen generally is the result of a logic error, ususally in recovering from hardware devices other than the cpu.  If the CPU fails, to work properly, you're sort of done.

                        And yes, we have incredibly complex operating systems.  If you started in the wire days you probably remember when 32k was a lot.  But a voting machine is a relatively simple devices.

                        And whether you count ballots by saying 'one' 'two' 'three', or if you scan them, or if you do them electronically, you are going to report the result and they will, gasp, use a computer to communicate and tally the results!

                        •  A "logic error" that results in what? (1+ / 0-)
                          Recommended by:

                          An incorrect numerical value.

                          Jeez!  How many precincts are going to have to add up several six-digit numbers by hand?  If there are indeed single precincts with nearly a million voters, all that tells me is that we need smaller, more numerous precincts.  And, for God's sake, if your only task is to total several six digit numbers, use a damn calculator or an adding machine with a tape.  The I/O is a helluva lot more transparent and the energy usage a great deal less than using a computer system.

                          The problem, of course, is that e-voting computer systems are NOT simple machines.  They are full-blown computer systems.  They still use an operating system and software that handles the recording and tabulation of votes, and both are still written in code that does "stuff" that is incomprehensible to 98% of voters.  Therefore, the processing of the data is not transparent.

                          Gasp!  Precincts might actually use computers to transmit their hand-counted totals?  And no human is going to check to make sure that the transmitted tally was received correctly?  

                          Now you're being silly.  What we're talking about here is the use of computers at the point of voting to record and count the votes, the results produced by which become the official tally, with hand-counting relegated to after-thought or invoked only if a problem is suspected.

                          Using a computer system to record and count a couple thousand votes is like using a chainsaw to deadhead the delphiniums.

                          Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                          by sxwarren on Fri Sep 22, 2006 at 02:54:11 PM PDT

                          [ Parent ]

                          •  I must say sxwarren... (0+ / 0-)

                            the final two or three lines of your posts are really priceless.  I like your style.

                            so sorry for the epic...

                          •  Not priceless at all. (1+ / 0-)
                            Recommended by:

                            Like most commercial, out-of-warranty tech support these days, your credit card is automatically being charged tens of dollars per minute for this.  ;-)

                            Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

                            by sxwarren on Sat Sep 23, 2006 at 06:41:28 AM PDT

                            [ Parent ]

    •  Again, missing the point. (0+ / 0-)

      E-voting is inherently NOT transparent.  For that matter, neither is absentee voting, which I also do not advocate unless there is simply no other way for someone to cast a ballot at all.

      Yes, it IS possible to cheat in a system that is exclusively paper ballots, hand-counted.  However, the whole point to computers is that they allow us to, with very little effort, manipulate data on a scale and with a speed that is physically impossible with hard copy.  

      With computer voting systems, it is literally possible for a handful of people to alter the count simultaneously in a hundred closely contested precincts by an amount calcualted to be just enough to change the outcome without going outside the predicted margin of error.  And this can be done without being observed and without leaving any evidence behind.  How many hundreds of people acting in a super-secret, hyper-coordinated conspiracy would be required to ppull that off if everything were hard copy?  The logistics would be mind-boggling, as would the odds that someone would fuck-up and leave some evidence behind, as would the odds that someone would get caught and crack under questioning.

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 03:19:00 PM PDT

      [ Parent ]

  •  Well, uh, er... (3+ / 0-)
    Recommended by:
    sxwarren, corvo, kraant

    So, what precisely is "broken" with a voting system that uses paper ballots, hand-marked and hand-counted in full public view?

    Sometimes the OTHER PARTY WINS THAT WAY!

    Read James Loewen's "Sundown Towns"!

    by ChicagoDem on Wed Sep 20, 2006 at 02:34:11 PM PDT

    •  Yeah! It's like with those damn 'blogs'! (0+ / 0-)

      You can write or say anything you want to!  We need to put a stop to THAT, fer sure!

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 03:29:06 PM PDT

      [ Parent ]

  •  The Reason I Mentioned the Brazillian Machines (0+ / 0-)

    Pretty much everyone agrees that the system in Brazil is solid and trustworthy. There is a tiny bit of noise about the lack of transparency (anyone wanna bet it comes from the US?)

    Thus: maybe transparency is not the principal guarantor of a fair election.

    That being said, I am actually for transparency. I just don't think that it is any type of guarantee, and we shouldn't delude ourselves into thinking that once we get it, that our problems are over. Where do you think the term "ballot-box stuffing" came from?

    "[T]hat I have no remedy for all the sorrows of the world is no reason for my accepting yours. It simply supports the strong probability that yours is a fake."

    by Heronymous Cowherd on Wed Sep 20, 2006 at 05:52:28 PM PDT

    •  In any system in which data is recorded and (0+ / 0-)

      stored digitally, a handful of people - even one person acting alone - can instantly alter all the data without being observed and without leaving a trace.  That's the danger in the lack of transparency.  

      There is no doubt that ballot boxes have been stuffed in the past.  That technique, employed by the Chicago Mob in 1960, probably helped JFK get elected.  But it requires a hell of a lot of potentially observable screwing around by a hell of a lot of people to get the job done.

      Private life is all about managing pain. In business and government, this means externalizing and deferring costs whenever possible.

      by sxwarren on Wed Sep 20, 2006 at 08:57:44 PM PDT

      [ Parent ]

      •  Chicago in 1960 was fairly unique (1+ / 0-)
        Recommended by:

        As I recall (and I was living in downstate Illinois at the time), precinct committeemen selected their party's poll workers, and precinct committeemen were elected by the voters in their precinct.  Since there were some precincts in Chicago where there effectively were no Republicans (or no more than a handful), the Daley machine could tell a few of its members to register as Republicans, and who to elect as precinct committeeman in that precinct.  And you'd then have the fox guarding the henhouse.

        But those kinds of machinations could only make a difference in an EXTREMELY close election.  And if an election is significantly closer than a 1% margin, it's effectively a tie.  LOTS of things can change the outcome of that close an election -- bad weather in one particular part of a state, traffic jams in an area that's particularly strong for one party, cheating, or lots of other things, including simple random chance in who has to stay late at work and misses voting.  

        My concern with e-voting is that it's relatively easy to change the result of an election in which there is actually a clear winner, and to have nobody be the wiser -- and there are a lot more elections where the winner has a margin of 1%-10% than there are where the margin is less than 1%.

        (And as an historical aside, although I think Nixon probably would have narrowly carried Illinois in 1960 if only the votes cast by actual living voters had been counted, that still wouldn't have made him President.  He would also have had to carry Texas or Missouri which, while close and where allegations of improprieties were made, weren't nearly as close as Illinois.  I'm far from convinced that what probably happened in Chicago affected the outcome nationally.)

        And by the way, ballot boxes weren't stuffed in Chicago in 1960, since my understanding is that Chicago already used voting machines (the old-fashioned, lever-operated mechanical ones) at that point.  Which made it all the more strange when the results from the last couple of precincts in Chicago didn't come in until well after dawn on the next day, after the results from last few hand-counted precincts from rural downstate Illinois came in.

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site