[E-voting] What is the problem ?

Aengus Lawlor aengusl at eircom.net
Mon Apr 26 10:32:46 IST 2004


Dr J Pelan <J.Pelan at gatsby.ucl.ac.uk> wrote:

> The case would be much more compelling if there was someone willing to
> knock-up a few prototypes. Anyone with time on their hands out there ?

Speaking of prototypes, does anyone have any guesses about how long it
would take to manually sort the ballot papers in an existing
constituency? Someone calculated how many unique cimbinations of votes
turned up in the 3 pilot constituencies, and (if I remember correctly)
it was only a couple of hundred of the potential millions of
possibilities. If you sorted the ballots, counted each bundle, and
entered those numbers (just a few hundred) into the count software, it
could produce the result of the count in minutes.

If fractional transfers aren't introduced, and you need to use unsorted
ballots because of the "random surplus transfer" factor, then you could
publish a random order for the ballots before the sorting, so that there
wouldn't be any question of the sort being tilted.

I imagine that manually sorting the ballots from a full constituency
into seperate bundles would take less time than our current manual count
procedure (which sort of does this now, but on a piecemeal basis).

Aengus




More information about the E-voting mailing list