[E-voting] Re: terrorism and electronic elections

Michael McMahon michael at hexmedia.com
Tue Jul 12 10:02:10 IST 2005

>> VVAT /plus/ mandatory hand count of all ballots would be close to 
>> acceptable, if I felt sure that the hand count wouldn't eventually be 
>> dispensed with.  (If you've got a mandatory hand count, why bother 
>> with the electronics at all?  Just a huge unnecessary ongoing 
>> expenditure that has no true intrinsic benefits and risks losing 
>> everything.)


This assumes that hand-counting *all* the ballots is actually
necessary. In the US they only hand check a tiny percentage of
the old counting machines (eg. 1% of precincts in California),
and nobody has complained about that afaik.

I think there is a perception out there that
checking random samples is somehow similar to
exit or opinion polls, and is therefore based on some kind
of flaky pseudo-science. But that's not the case.
Sampling of the actual votes, rather than of voters
statements about how they did or will vote, is governed
by normal rules of mathematical probability.

I do share your concern that the checking might eventually
be dispensed with, but that's why the laws regulating e-voting
need to be primarily concerned with the details of auditing
and checking, and not with how the system itself is constructed.
At least if auditing is mandated by law, then there is a reasonable
chance that losing candidates will create a fuss, if the procedures
are not completely followed.

Personally, I believe it would be quite easy technically to add
a printer to the voting machines, and while I do have some
reservations about VVAT, I think it could be a viable solutio.
IMO, the best way to find out, would be to run a pilot in
one or two constituencies. This would help to determine:

1) how the public deal with the extra complication of
    the printed ballots

2) how much it adds to the running costs per election.


More information about the E-voting mailing list