lol
the guys name is seriously 'Kitcat'????
The Open Rights Group (ORG) has condemned the May 2007 pilots of e-voting and electronic vote counting in the English and Scottish local elections, saying the technology involved is simply not suitable for use in statutory elections. "We came into this not as a blank sheet," ORG e-voting coordinator Jason Kitcat concedes. "But …
With everyone banging on about their rights, might it be time to include the associated responsibilities?
Force people to vote (preferably using a paper ballot) and include a "none of the above" box on the form. This would allow people to express their justified boredom with the same ol', same ol'. It may be impotent frustration initially, but when the political elite are forced to acknowledge the scale of the problem of voter apathy through such a scheme, the excuses will be whittled down.
It's possible that the motivation behind some of the initiatives is about trying to re-engage people (i.e. greater postal voting / e-voting), but the execution of these plans has been diabolical at best and brought shame to one of the oldest democracies in the world ("A farce worthy of a banana republic" according to the judge in the Birmingham postal vote fraud).
Of course, there's every chance that the mindless, latté-sipping drifters of our celebrity-obsessed consumer society will continue with their "am I bovvered " approach to this. What else can we do to (metaphorically) put a rocket up their ar$e and get them interested in something that really does affect them on a daily basis? Too many lives have been lost defending our freedom to allow these poorly schooled morons to perpetuate their special brand of laisse-faire cretinism.
Actually, this also ties in nicely with a previous post of mine ("Technology for technology's sake"). All this extra expense and it still doesn't work! Remember that story about the Yanks spending $millions developing a pen that would write in space? The Russians saved a few bob by taking a pencil instead.
"Our view is that you can't actually build an electronic voting system that meets these [Kitcat is presumably refering to a voters ability to check that their vote was counted] requirements," Kitcat told us.
Voters are given no guarantee in today's system that their vote is counted. They are given assurance yes, but no guarantee. Furthermore, today's systems do not even provide privacy - the way in which a voter cast her vote can be revealed!
e-voting systems aim to provide strong cryptographic guarantees. The none-technical reader should note that a crypto guarantee is far stronger than an assurance. One such e-voting system is the well known FOO protocol, developed in 1992, which has been verified to provide the following properties:
Privacy: the way in which a voter cast her vote is not revealed to anybody.
Fairness: no partial tally of results may be obtained until the official count.
Eligibility: only authorised voters may vote and at most once.
Individual verifiability: a voter can verify that her vote was really counted.
Kitcat's claim that you cannot build an electronic voting system that allows voter verifiability is simply unfounded - you can. However, FOO is not suitable for practical deployment*. I am not aware of all the literature in the field, but I suspect there are e-voting protocols in existence which satisfy at least the four properties introduced above. And maybe even:
Receipt-freeness: the voter is unable to prove that she voted in a particular way.
Universal verifiability: anybody can check that the published tally really is the sum of the votes.
Receipt-freeness prevents vote selling - if a voter buyer cannot be convinced that a voter did indeed vote in a particular manner, he will not pay the voter.
Footnote:
* FOO requires voters to participate in three sessions, thus making it impractical for use in local/general elections.
Why not use the electronic equipment to produce a machine punched card with your vote. You can then manually put this into the counting box. The cards can then be machine read with (presumably) a pretty high accuracy due to the uniformity of the holes punched.
Or am I barking up the wrong tree?
You are not so far off...
The idea I read was from Bruce Scnheier. You use an electronic system to prepare (note: prepare, not cast) your vote. When you are ready, you press the "OK" button, and a small printer whirrs out the voting paper, pre-printed, and stores voting info into a DB.
From there, the voter checks the paper, and either cancels, retries or maintains. This is also stored.
The voter then walks over to the ballot box and slides the print in.
At the end of the vote, the total number of votes are tallied, and compared to a printout. If everything matches up, then there is a chance that the computer is right, as long as no voters complained about the wrong slip being printed.
A cursory count of the ballots can also be done by hand and compared to the box.
If all is ok, then the computer is a backup and "recount" to the manual paper system.
If there is a discrepancy, then the paper votes are law, as they were checked by the voters by hand before voting.
In this way, the standard voting process is achieved, on paper, but using the computer as the recount. Also, if all is OK, then the computer can also upload the statistics part of the vote data to internet and inform the voter that they have voted and that their vote has been taken into account, or that a manual recount has taken place that is overriding the computer tally.
This would alleviate the needs of multiple manual recounts, but voting is a human affair and it is wrong to delete the human element from the checks and balances. A computerised system is there to assist, but in no case is it there to replace (and voters like putting their piece of paper into the box).
It reminds me again of a short story I read on the internet about a US presidential election 100% run on federal mandated electronic voting machines. At the end of voting day, the president elect was a 22 year old unpartisan candidate unknown to the general public who was until that day a programmer from the company that made the electronic voting software, with a totally unexpected unexpected landslide win....
"Come to the podium M. Acker. My fellow Americans, I present John H.Acker, the new president of the United States..."
Cheers,
Daniel
Yes there are cryptographic algorithms for voting that provide all sorts of lovely provable properties. However you have to factor in that the average person is incapable of understanding the principle of operation of such a system, let alone auditing one. A black box that claims to use cryptomagic is no more trustworthy to Joe Bloggs than a black box that claims to add up correctly.
Look at the state of online banking over https. A customer, who always checks that the url is at the correct domain and that a padlock symbol appears in the status bar, is above average. Yet this is a long way from truly auditing the system and still relies on a lot of blind faith. Did you check the host certificate's chain of trust? Has the browser been modified? Has the OS,graphics driver,keyboard driver,keyboard,browser compiler or CPU been tampered with? What about the compiler used to produce the software that did the design of the northbridge in the machine that's doing the auditing calculations?
Full e-voting will never be guaranteed secure. Electronic counting of some easy to recognise physical voting token sounds like a much better way to me.
@Sir Runcible Spoon,
e-voting does not necessarily mean voting on a computer. The people involved with making these decisions don't even know this. e-voting incorporates all technologies which allow voting by an electronic means. Therefore your punching machine and an electronic ballot counter fits the definition and is thus an e-voting system.
Daniel proposes a method _almost_ like the "Voter Verified Audit" one, but including the step:
The voter then walks over to the ballot box and slides the print in.
This has the problem of being enough like the traditional paper ballot to be vulnerable to the "shirt-tail vote". This starts with one voter palming the real ballot and sliding a blank sheet of paper or fake ballot into the box. It will be discarded, but no matter. The ballot thus obtained is marked with the party's choices, the first "voter" is paid, and the second (third, etc.) voter is given the pre-marked ballot to deposit, returning the new unmarked ballot each time to the waiting party officials.
Not that the scheme with a visible (but not touchable) printed ballot is immune to problems. Since the "ballots" are kept in a single roll, it is possible to discover with fair accuracy what votes a particular voter cast, followed by reprisals from the authorities.
Any quick fix is likely wrong. There is (was?) a freely available (and free-software) system from Australia, inventors of the secret ballot, some years ago, that used a keystroke audit trail for verification, but I do not know what came of it.
Considering the amount of FUD/spin/shit the general masses are willing to accept from people they precieve to be in authority its a wonder people still think its all a good idea.
Germany in the 1918/19 was the first country to allow universal suffrage and using PR (which on the surface may seem even more democratic). Fast forward 15 odd years and look where they were.
Universal Suffrage Democracy is not all its cracked up to be if u ask me, even the Greeks didn't let everyone vote.
What you all need to do is make me your dictator, I will ensure that people will be proud of Britain but also that there will be social equality, does that make me a National Socialist i wonder :)