[LINK] The Register on vote hacking

[George Michaelson]

I think the reason people talk about online voting is that they want the cost
of vote counting amortized *in time* across the time to run the poll.

For each vote, tally its impact on the voted matter. At close of poll, 
results are available in a fraction of the time at present. 

Of course, this is also anathema: imagine what it does to the process
if the 50%+1th and onwards voters know their vote is wasted? (this is
more true in first past the post and less true here)

Also, early sight of vote patterns definately skews the result as people
shift voting behaviour given knowledge of the vote in progress.

So people proposing this are really proposing quite radical changes in
the social dynamic of voting, and as you point out are not just 

 "making it easier for you" (tm telstra in all states)

(I'm sure the AEC have other motivations to discuss it btw, but I see
 their role as more neutral, exploring the space)

What do people think about the non-repudiation aspects of simple token
schemes here? Privacy etc can be well managed if the register-to-vote
process is decoupled from the 'prove you can vote' and 'prove you have
not already voted' and 'lodge a vote' phases, conceptually anyway.

Lots of times I've seen people like SAGE-AU or USENIX hand out random
text to be keysigned by pgp keys, to then lodge as proof of something,
Greg Rose (who is no fool) helped set one up for AUUG I believe?

And of course there is a TTP role for non-repudiation and timestamping as 
well (lets not get into the keygen arguments, just the certification of
(x) doing (y) made by (z) at time (t) kind of thing.)
-George

--
George Michaelson         |  DSTC Pty Ltd
Email: ggm@dstc.edu.au    |  University of Qld 4072
Phone: +61 7 3365 4310    |  Australia
  Fax: +61 7 3365 4311    |  http://www.dstc.edu.au