[E-voting] Press Release Third draft with minor corrections, additional final paragraph and spellcheck..

Aengus Lawlor aengusl at eircom.net
Thu Apr 15 16:28:43 IST 2004


Casey, Dermot (GE Consumer Finance) <Dermot.Casey at ge.com> wrote:
>>>>
>>>> Due to the procedures put in place by the presiding officer, it is
>>>> impossible to tell if the discrepancies are the result of human
>>>> error, machine error or even tampering.
>>>
>>> The procedures put in place by the presiding officer, if followed
>>> properly, are precisely what is supposed to tell us that the
>>> discrepancy
>>> must be in the machine. The problem isn't that the procedures
>>> obscured this information, it is that the procedures weren't
>>> followed.
>
> Do we know the procedures weren't followed ?

Yes, we do. The "Post Election Report" states:
"The forms used in the polling stations are poorly worded and led to
substantial confusion when it came to carrying out the reconciliation of
each station.  In the end it was decided to perform the count without
reconciling the stations as it proved impossible to do so on the night.
In particular many presiding officers filled in the total number on
their register or the total number of tickets issued to them instead of
the number of voters marked off on their register (the number of tickets
issued).   Some staff also recorded each person that they sent to a
standby machine as a voter who left their station without voting."


>>>> time that "There is also no correlation in the polling station
>>>> between the
>>>> number of votes cast and the number of votes counted, even
>>> though this
>>>> is an elementary procedure in most if not all polling
>>> stations in the
>>>> former Communist countries of Eastern and Central Europe.
>>>
>>> This is a weak argument. As I pointed out yesterday, it's not
>>> possible to do PR-STV counts in the polling stations. And there
>>> certainly is a cross-check made between the number of votes cast
>>> and the number of votes counted. It's the discrepancy in this
>>> cross-check that is at issue.
>>>
>
> To me this is a very simple count issue
>
> 1. 1000 voters crossed off the registered voters roll
> 2. 1000 ballots papers in the boxes
> then there is no issue

Dermot, you can't count the number of ballots in a box until you open
the box. And the box isn't opened until it is brought to the count
centre. Therefore you can't count the number of ballots in the boxes in
the polling station. We have to do it this way for PR-STV elections, and
it would not be sensible to have different procedures for referendums,
whether they are "elementary procedures" in Eastern Europe or not.

The number of votes in a ballot box is checked against the number of
voters crossed off the register. This is done when the box is opened.
(An envelope containing information about the expected number of votes
is attached to the box).

The proposed electronic system _will_ make it possible to verify, at the
polling station, that the number of ballots recorded by the machine is
the same as the number of people who voted. It's not clear if this check
was made, or was supposed to have been made in this case.

> Its not about doing PR-STV counts. Its about making sure firstly that
> the correct number of votes are recorded (which is the problem in this
> case) and then making sure the contents of those votes are OK. The
> second part doesn't work without the first......

That's _exactly_ my point. You are making allegations about the second
part ("716 missing votes") when all you know is that the first part was
screwed up.

> Feel free to change

I'm not trying to make any personal criticisms of this Dermot. I'm just
concerned that this draft leaves ICTE open to criticism of
sensationalism. The information we have _is_ sensational, but the
suggestions included in this draft press release aren't backed up by the
information that we have.

Aengus





More information about the E-voting mailing list