[E-voting] The VVAT Debate - Technical or Not ?
colm at stdlib.net
Mon Mar 8 20:49:28 GMT 2004
On Mon, Mar 08, 2004 at 07:36:50PM +0000, Michael McMahon wrote:
> I'm also sceptical about sampling in general as a technique for finding
> errors and instances of fraud.
I'm sceptical of it's useful for finding instances of error, and
personally I think a 100% verification is workable. It would not be
urgent, and take place over a month, the likelyhood of an error (with
a VVAT present) leading to a change of result would be small - and
it would allow us to catch the problems.
I see a VVAT as being a disincentive to try electoral fraud in the first
place. The repursions are already very serious (jailtime, commercial
liability and so on) coupled with a chance of being caught lowers the
risk of occurance significantly.
I would apply the same argument to software quality; once there is
verifiability, the vendor has to become very serious about software
quality, or they will have a big problem on their hands. The big
problem is that there is little chance of problems/fraud being
detected and that is counter-productive.
I think the percentage of spotchecks that need to occur is a side-issue
and a procedural matter :)
Colm MacCárthaigh Public Key: colm+pgp at stdlib.net
More information about the E-voting