November 05, 2004 | Graham

Dubya stole this election too

Silly me. I assumed that there could be no way that Bush’s result in this election could be challenged, but there is. In the Florida election 2000 it was the voting machines with their hanging, dimpled and pregnant chads. This year it is the computerised voting machines without their hanging, dimpled and pregnant chads! Just read this piece by Greg Palast.
Blogger Danny Schecter was defending the role of bloggers in the US elections this morning on Peter Thompson’s ABC breakfast programme. Bloggers were being criticised for calling the election early for Kerry. This was done on the basis of exit polls. I have sympathy for them. I more or less did the same thing on the basis of reports of the exit polls on ABC radio, but only in a couple of telephone conversations. It appears that the exit polls were wrong. But were they?
The Blogosphere has a different view, or at least some parts of it. Following the trail left by Schecter and Palast leads to this post on the Daily Kos. According to this theory, the exit polls got it right, and it was the voting machines that got it wrong, thus the role of the blogger as all knowing, all prescient, and 24 hours ahead of everyone else’s deadline with the truth, is justified, vindicated and upheld.
The evidence is that the exit polls were apparently right in the non-contentious states, but wrong in at least two key battlefield states – Ohio and Florida. The blogger explanation as to why this might be so is that obviously the voting machines had to have been tampered with. In these machines in the US, once you commit your voting choice to electrons, there is no auditable paper trail of what you did. (I have no idea whether there is a computer log that serves the same purpose.)
I haven’t viewed the polling evidence because the link to it is dead on the Kos, but the proposition is arguable, which is a pity, because it is just the sort of thesis that Michael Moore could base another movie around, and I’m not sure I could stand that. There have been problems with US computerised voting machines, including scrappy code and compromimsed security. In addition, some of the manufacturers of these machines have Republican Party links. I’d want to know a lot more about the machines used before getting too excited one way or the other.
Pity that the US doesn’t take a look over here. I’m reliably told by Tom Worthington that the system invented at ANU to conduct the ACT elections doesn’t have any of the problems of the US systems, being open-source and producing its own internal audit trail.
Another Australian, Craig Burton, has voting software which he contracts out to clients that include the British Labour Party who have used him to conduct their internal ballots.
Will the bloggers prevail in this argument? I’m not sure. An alternative way of looking at the whole issue is that, having been wrong about the exit polls, bloggers are now casting about for a plausible alibi to explain their own failings away, making them just as venal, wrong-headed and self-interested as some of the media that they have made a name for criticising. They could just be compounding the original exit poll error.

Posted by Graham at 6:52 am | Comments (1) |

1 Comment

  1. There is a “race condition” in voting now such that some votes get counted and others do not, purely driven by the expectation of a same-day result. This is a new digital divide brough about by some states having automated voting and others not. It is exacerbated by absentee voting and provisional voting (first tried this election) being paper systems. 130K provisionals were not counted in Ohio and more than half of 1.3m absentee votes were not counted in 2000.
    The touchscreen machines will be probably be acceptable with the audit issues are sorted out, but I am more in favour of scanning. Exit polls, however, cannot be relied upon for very much confirmation of results as their margin of error is higher (fewer procedural controls over security, no audit, not supervised, not private, not anonymous etc).
    The big problem is that every vote should be counted. This means that there has to be a more reliable way for travellers, embassy staff, military and others out of the county or country to vote. Overseas in 2004, the military have resorted to downloading PDF ballots, printing them, scanning them, then emailing them to the DoD whereafter they are FAXed to local government.
    Internet voting works, it is secure and it is acceptable if it is made available as early voting as well as on-the-day voting. Don’t refer to the cancellation of Accenture’s DoD system after Aviel Rubins’ report as last word on Internet technology.

    Comment by Craig Burton — November 5, 2004 @ 12:28 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.