back to article How scanners and PCs will choose London's mayor

Very few politicians are recognisable by their first names only, but next week, two such larger than life characters will face each other in the closest battle for the office of London Mayor since it was re-established in 2000. The polls have the Labour incumbent Ken Livingstone running neck and neck with the Tory contender, …

COMMENTS

This topic is closed for new posts.
  1. Ben Griffiths
    Thumb Down

    More People

    I can't understand this. The vote is cast on paper, the machine counts it, except when the machine can't figure it out, then a human counts it. A human then checks the machine by doing a manual count of some ballots.

    Why not just keep the humans. I think the advantage of having counting done in the open outweighs the errors the humans will make.

  2. philbo

    Did he really say this?

    "Bennet says that such an audit would be "meaningless" and bad for voter confidence. This is because the rules that govern the counting procedure do not allow for both a manual and electronic count."

    ...IOW it's better for people to suspect that the electronic count is flawed than to have it proven either way. It could only be bad for voter confidence if it comes up with a different result - and surely you'd want to know if that were the case?

    How ****ing stupid can you get?

  3. Chris Miller
    Thumb Up

    Welcome back Lucy

    We've missed you :)

  4. Spleen

    Nice of them to admit it

    The guy in charge has essentially openly admitted that the primary concern is "voter confidence" (i.e. no open revolts occuring) rather than ensuring that the voice of the people is accurately counted.

    Still, it doesn't matter because the incumbent is an honourable man who in no way would do anything immoral like sending his close friends tens of thousands of pounds of taxpayers' money which then disappears off the accounting books and threatening people with the race card when they ask what happened to it. So even if the system was vulnerable he'd never dream of abusing it.

    More reasons not to live in London.

  5. Paul Slater
    IT Angle

    Accountability

    "It might be possible, Mercuri contends, for a hidden piece of code to be activated, or for a machine to be subverted by scanning a particular bitmap image"

    Why is the software not open-source, in that case? This is the public, voting to public officials, after all.

  6. Paul

    Given that...

    ... even something as simple as a National Lottery terminal frequently struggles to recognise marks on a piece of paper, why on earth would you consider using the same technology in something as important as an election?

    Or is it that all the terminals are rigged to let a certain candidate to win?

  7. Mike Crawshaw
    Thumb Down

    No, no, no.

    "We could do a sample manual recount, but if it turned up a problem, we wouldn't be able to do anything about it, which would be the quickest way to collapse voter confidence in the result,"

    No, the QUICKEST way to collapse voter confidence in the result is to state that "we think it'll work, so we're not going to check it, and we won't do anything if we think there's a problem cos we don't want anyone to know if it doesn't work."

  8. Jim
    Paris Hilton

    To add another voice...

    "We could do a sample manual recount, but if it turned up a problem, we wouldn't be able to do anything about it, which would be the quickest way to collapse voter confidence in the result,"

    Eh? You find an error and you can't do anything about it? Now this may sound mad but how about... Wait for it... Declare the election as void and start again?

    Yeah, it's a pain in the arse but less so than several years of endless moaning that the result was fixed, blah, blah, blah...

    On a different note, is it only me that read ORG as OR-G (maybe something to do with OGC)?

  9. Duncan Hothersall
    Heart

    Some inaccuracies here

    Good article, but some inaccuracies and a significant error of omission I think.

    In the Scottish general election last year, the vote counting machines were not the final adjudicator for a single ballot. Any ballot which could not be confirmed was passed to a human to be checked. To say that 70,000 of the spoiled ballots were machine-checked only is simply incorrect. All of the spoiled ballots were passed to humans to check. If there were failings in the checking they were human, not machine errors.

    More importantly, while some machine failures caused counts to have to be stopped, the main problem in the Scottish elections was spoiled papers - actual spoiled papers, rather than incorrectly identified spoiled papers. This was put down to the flawed design of the combined ballot paper for the Parliamentary elections - a theory borne out by the fact that hardly any council papers were spoiled, despite it being the first time STV had been used.

    In other words, the use of counting machines was not the cause of the main problems with the Scottish elections. Bad ballot design, and public stupidity, were the main problems.

    Those who would remove automated counts from the system altogether should make clear that for the STV council elections, this would mean results taking 3-4 days to be computed from the larger wards, and would cost an enormous amount more in training and human resource costs.

  10. Lou Gosselin

    Solutions

    > "...a hidden piece of code to be activated, ..."

    > Why is the software not open-source, in that case?

    Two problems with this. Just because the manufacturer opens the source doesn't actually mean the machine is running this source.

    Secondly, the manufacturer claims the machines would not be reprogramable, so there may not be a way for a third party to verify the code on the machines.

    I still agree they should demand open source, at least people will catch genuine coding errors. Better than security by obscurity.

    I've long thought that all these issues could be solved the "NASA" way - tons of redundancy. Why not get 3 different vendor implementations to count the ballots?

    All the votes would be counted by machine A, then go to B for recount, then C for another. Then look for statistical discrepancies and investigate those. You'll quickly find out which systems have bias - deliberate or accidental.

    If the ballots are stamped with serial number before going into machines, then it would be possible to track exactly which ballots were disputed.

    Heck, it would be possible to have one scanning machine, that is incapable of anything but scanning. All these images go into a feed bound for each vendor's scanning software. This would bring down the cost of duplicating the scanning hardware between vendors.

    Why is this so hard to solve? Most of us have to solve much more difficult problems on a daily basis.

  11. This post has been deleted by its author

  12. Anonymous Coward
    Stop

    "Open Source" isn't a solution

    Open Source software might have some merit when you download the source yourself and compile it on your own machine, but what good would it do on any machine involved in "e-voting". You'd still have to take someone elses word for the fact that the machine was actually running the source that you downloaded somewhere else and verified to your own satisfaction.

    If you have to take their word for it any way, what difference does it make whether the source is open or closed?

  13. Wayland Sothcott
    Thumb Down

    May the best man win

    I suspect that will be Boris :-(

    It's very strange that they can't check it manually and the electronic count rules, no matter if it's wrong. As long as you don't know it's wrong then it's correct. Schodingers cat in a box?

    This has got to be the ideal system for allowing governmental election fixing.

    When you think of it as allowing for electon fixing then all the rules make far more sense. It's easier to pwn a machine than a human.

    The way round this is to somehow mark your ballet so it's human readable but not machine readable. I would trust the human ballet more but you risk spoiling the paper.

  14. Frederick Karno

    Why change ????

    All ballots imo should be hand counted it is a tradition if it aint bust dont fix it !!!!

    There is no need for technology to be involved it is complex and fraught with problems.

    Dumping a load of papers into a machine gives no sense of what might go wrong until way after results have been given out as happened in Scotland ,how that could ever happen is beyond belief,and the results allowed to stand.

    What bothers me even more are the number of postal ballots that will be cast these to me have to stop they are too easy to forge.Small numbers of voters change local election results dramatically.

  15. Pierre

    I spy, with my little eye....

    ... problems BEGGING to happen....

    How the heck an administration unable to prevent the stupidest data losses ever could think they're ready for that kind of thing. I'm very, very doubtfull about e-voting and non-controlled electronic counts in the first place (as everyone should be, because it eliminates the possibility of any independant control). But in this precise context, it's really funny (funny because I live far away, that is). Now, are the voting (or vote-counter) machines linked to our Heathrow T5 overlords?

  16. Pierre

    @ Lou Gosselin about solutions

    3 different vendors making concurrent machines might solve the problem, but only if you can be sure that the manufacturers don't share common interests (which you'll never know) or that they don't talk to each other (which you can't prevent). So, well, manually counting pieces of paper, with volunteers from each concerned party scrutinizing the process, can't be beaten.

  17. Steve

    spackle

    "We could do a sample manual recount, but if it turned up a problem, we wouldn't be able to do anything about it, which would be the quickest way to collapse voter confidence in the result,"

    The scary thing is that this double speak is so prevalent in our government that no-one seems to question it anymore. We don't need elections that people can be "confident" about, we need elections that people can be CERTAIN about.

    It seems people just like to be able to use what they think are technical terms like "voter confidence" so they feel like they are discussing something complicated.

    They then completely miss the fact that this guy has just said that "it's more important that you trust me than it is that I'm honest."

  18. Anonymous Coward
    Alert

    Why not weigh the ballots

    If the scanner controlled distribution to bins ( one per candidate) and the ballot papers were fed face up then the scruitineers could verify the contents of the bins and weighing the bins would give comfort that there had been no gross error or indeed the contents of each bin could be sent through a simple sheet counter.

  19. Anonymous Coward
    Go

    Manual is not that accurate

    Historicly ballots such as this were counted by humans. To do this for London would have taken about 3-4 days, rather than 10 hours. This is the main driver for implementing an automated count at an election.

    The assumption of the ORG and a lot of the posters here is that a team of humans would have counted the number of votes more accurately than a computer and would have been subject to less bias. If this is the case why have so many manual counts have to be recounted in previous elections?

    In the actual London election weren't some of the ballot boxes were manually verfied (that is the process of counting the number of ballot papers, not the votes on them) as the computer provided different numbers to the polling station staff. In all cases was not the number from the computer the same as the actual number of ballots, not the number from the staff.

    Surely we can move past this ludite attitude that the machines can't do a job as a human, when it comes to performing repetitive tasks mant times, like counting large numbers of votes they can!

This topic is closed for new posts.