[E-voting] UK govt circular mentions open-source e-voting

Michael McMahon michael at hexmedia.com
Thu Jul 21 23:48:44 IST 2005


Aengus Lawlor wrote:

>No, Michael. The voters DON'T have to trust them. The "party workers" aren't
>counting the votes!!!!! 
>
That's the second time you have said that! You can take it I know that 
"party workers"
don't count the votes.

>They are observing the counters (usually civil
>servants), and it is the presence of observers with mutually opposed
>interests in the outcome that the voters can trust, and, if they don't trust
>it, they can apply to attend themselves.
>
>  
>
Whether they are trusting "it" (the process) or "them" (the observers) 
is making an absurd
distinction IMO. For one thing, you're assuming that the observers 
behave as expected, ie. that they look
out for their own candidates interests. In any case, exactly the same 
"mutually opposed interests"
apply with electronic systems. Each party or candidate would be free to 
commission their own
verification software. Or, they might not bother. The choice is theirs.

I also seriously question whether the average voter would have a clue 
what is going on
if he/she turns up at a present day count (assuming he gets a ticket).  
One individual working
alone couldn't possibly come to a judgement about whether a count was 
done right.

>>In a crypto e-voting system, the place of the observers is taken
>>partly by the
>>programmers who write the verification software, and partly by the
>>voters who choose to download that software and use it to verify an
>>election.
>>
>>That's why I was making the comparison.
>>    
>>
>
>Right, so you want to replace a system that you've no direct experience
>with, but that any member of the public who can be bothered can participate
>in, and immediate understand, with a system that any member of the public
>who can be bothered can particpate in, if they have a computer, even if they
>can't understand it, and end up just taking someone else word for it anyway.
>
>  
>
I certainly concede that any system has to be understandable to everyone 
concerned, though not
everyone needs to understand it to the same level (that's the same with 
the current system).
The challenge for the vendors of these systems is to facilitate that, 
and frankly they have done a pretty
lousy job so far.

As for needing a computer to participate, that is not strictly true 
(though maybe I implied it).
Verification of votes should be possible by telephone. VoteHere's 
receipt can be verified in that way.
In all probability, few people would actually test the verification 
data, and that's the bit which requires
a computer. The parties would probably do that, as well as other 
interested bodies, and individuals
with the technical know-how.

>This is exactly what the term "a solution in search of a problem" was coined
>to describe. Our "observation" system isn't broken, and it is precisely the
>area that technology is least capable of adequately replacing.
>  
>
I have never said the old (Irish election) system is broken.
What I actually said was that *eventually* IMHO, crypto based
e-voting systems will be used, because I think there are benefits to 
e-voting, and crypto based systems
are the most practical ways to do it.. For what it's worth
I don't see it happening in the short term in Ireland because none of 
the crypto systems handle
multi-preference elections very well. On the other hand, they work well 
with first past
the post systems, so they may well end up being used in those countries 
first.
It's interesting that the reaction in the US media has been quite 
positive. I'd bet
that once the inertia and skepticism of the various parties is overcome, 
it will happen
in the US.

- Michael




More information about the E-voting mailing list