[E-voting] Seanad election count
cansbro at eircom.net
Tue Jul 31 10:11:19 IST 2007
I've read the Overview, which I highly recommend. It is a quick read.
The vulnerabilities found, in even the short period of time allowed for
testing, is truly shocking. It included security vulnerabilities in
hardware, software, firmware, removable memory (electronic ballot
boxes), and bootloaders.
The individual detailed reports have a public version and a private
version. The private version will go to election officials and,
presumably, system vendors.
Vendor obstruction was also a problem which prevented the tests from
being as comprehensive as is needed.
David Malone wrote:
> On Mon, Jul 23, 2007 at 01:58:34PM +0000, icte-jlambe at johnlambe.com wrote:
>> If anyone wants to attend the Seanad election count for the NUI panel,
>> as an agent (of Martin Hogan, Green Party endorsed candidate), please
>> phone or text me at the number below.
> I went along to the one in TCD - it was interesting to watch for
> the mechanics of the postal vote. I've included below some comments
> I made below in a blog post.
> Unrelated, but also interesting looking are the links to the first
> reports on the study of three evoting systems used in California:
> I went in to watch the counting of Seanad votes for TCD on Monday
> to see how it is done. Basically, it is very like the general
> election, with the added complication of being a postal ballot. I
> was impressed with the number of people abroad who had bothered to
> return their ballots.
> The biggest problem seemed to be people who had not had their
> declaration-of-identity form correctly witnessed. The returning
> officer accepted ballot envelopes as long as they accompanied by
> signed and witnessed declaration-of-identity forms (even if not
> dated, not addressed or they had been put within the ballot envelope,
> which they are not supposed to be). Even after this a substantial
> number of votes had to be rejected.
> The number of votes per seat ended up at about 5600. This isn't too
> bad when you consider that in Bertie's constituency in the general
> election, the number of votes per seat was only 15900.
> After all this filtering, there were a relatively small number of
> spoiled votes - I guess a university education must help a bit when
> it comes to filling in ballot papers. I was surprised to see that
> votes marked with a single "X" or tick were rejected as spoiled.
> In a general election, the chances are that these would be accepted.
> I brought this to the attention of the returning officer, and she
> said that the 1937 act said that you had to mark the paper with "1"
> or "one".
> It seems that while the 1992 Electoral act allows a returning office
> to interpret marks like "X", the old 1937 act does not, and while
> the 1992 act amends some of the 1937 act, it does not change the
> rules for counting or interpreting votes.
> E-voting mailing list
> E-voting at lists.stdlib.net
More information about the E-voting