Adrian's rebuttal

I hope Adrian doesn’t mind, but it’s a great rebuttal. Keep reading…

> There have been many constructive contributions to the debate on
> electronic voting. Regrettably however there have also been several
> ill-informed contributions. Sometimes this is a result of people simply
> not having done their homework. In other cases it is mischief making and
> attention seeking. In the interests of clarity I would like to make the
> following points:

“Attention seeking” is a new slur for the list. Perhaps Mr Cullen
wishes really hard that the media would stop paying attention to us?

> Contrary to some claims, the up front costs for the introduction of
> electronic voting are ¤43 million including VAT. While cost savings
> have not been our primary concern, it is expected that substantial
> savings will be made in electoral administration, particularly
> surrounding the count procedures. The vast bulk of the expenditure on
> this project is a once-off capital expenditure to purchase the voting
> machines, which have a life-span of some 20 years. .

Yes, and the other ¤39 million are incurred over the 20-year lifespan.
No contradiction here.

> The integrity of the new electronic system has been vigorously tested by
> six independent, internationally accredited test institutes:

Here we go again!

> * Physikalisch-Technische Bundesanstalt (PTB), the German institute
> which tested the voting machine software, confirming that it
> performs all the tasks required and that it has sufficient
> internal checks to identify any attempted interference.

The PTB reports clearly say that the internal checks detect
“malfunctions”, not “interference” which is (stated equally clearly
to be) dependent solely on the effectiveness of the seals. (Note:
there is no study of the effectiveness of the seals.)

> * Nathean Technologies, an Irish software firm, undertook an
> architectural code review of the election software and concluded
> that the code does not contain elements which can corrupt the
> correct running of the software.

They did? I mustn’t have read it properly, because I can’t find any such
declaration. I did find this:

} Recommendations
} a) Convert the database from Access 97 to Access 2002. The database
} technology utilised by the IES application ­ Opus DirectAccess ­
} readily supports the Microsoft Access 2000 format. This format
} presents several additional benefits ­ for example, the system would
} be less open to corruption when dealing with large database sizes.

Uh, _less_ open?

> * The Electoral Reform Society in the UK tested the PR STV (Single
> Transferable Vote) count rules against the 400+ STV elections in
> their database to ensure that the rules have been precisely
> applied.

And it only took them ten releases to finally get a version which
passed the ERS’s fixed set of test data. Will it work on different
test data or even real data? Er, no-one’s checked.

> * Zerflow Information Security Ltd undertook a rigorous security
> assessment of the system and confirmed that the measures
> introduced following the initial pilot elections have addressed
> any concerns they have with the system.

“Rigorous security assessment of the system”? Doesn’t he mean
“security assessment of the procedures to be applied in the use of
voting machines in the physical environment of polling stations”
which was “a subsidiary and less extensive exercise than the other
tests” (Pat “The Cope” Gallagher, Dáil Éireann, 2002-12-10)?

> * Kema Quality BV (accredited by Dutch Council for accreditation)
> examined and certified the physical voting machine components.

Yeah, “for compliance with international safety standards” (Martin
Cullen, Dáil Éireann, 2003-02-04).

> * TNO, the Dutch Electronic Products and Services company tested the
> voting machine and supporting equipment for compliance with
> international standards for environmental conditions (such as
> temperature, humidity, power supply voltage and interruptions,
> electromagnetic compatibility, insulation, energy consumption and
> transportation).

I’m sure we’re all reassured that humidity will not be a factor.

> In response to claims that Microsoft Access is not a suitable
> application for counting the votes, the Department has received expert
> advice on this issue from Nathean Technologies who reviewed the use of
> Microsoft Access in stand-alone security-hardened computers for the
> counting of votes and certified that it is fit for use in the elections.
> Furthermore, the election management system being used in Ireland has
> also been used in Holland and Germany with no problems and is also being
> adopted in France.

*choke* Is that the same Nathean Technologies who gave us the above
quote about database corruption? And who made two other recommendations
to “enhance” the security of the database but agreed to defer the
issues until after the June election, “given the proximity of the
election” and “given the Department is putting in place robust
installation and usage procedures to prevent any inappropriate
physical or networked access”?

> It is also worth noting that the votes stored in the ballot modules will
> be intact and kept by the returning officers for a period of six months.
> If so ordered by the Courts, all the votes cast in the election can be
> printed out and counted manually.

s/cast/recorded/

> The people elected should be the people with the most votes at the end
> of the count.

Is that “should” as in “ought to be”, or “should” as in “is expected to be”?

> With the Irish electoral system, seats can be decided by
> the smallest of differences. Indeed, at the last General Election,
> seats in 18 constituencies were decided by less than the number of
> spoiled votes. Similarly, there were 40 local electoral areas where
> councillors were elected by less than 50 votes. At the last Local and
> European Elections, 70,000 people who took the time to go and vote had
> their votes declared invalid.

An irrelevant tragedy, I’m sure we’ll all agree.

> This is a system which can end the problem of inadvertent spoiled votes
> and make sure that our elections are more accurate and therefore more
> democratic.

`You don’t HATE DEMOCRACY, do you?’ Maybe we could consider saying
“VVAT can make sure that our elections are more accurate and therefore
more democratic”.

> The system is also being introduced as it is an easier method for voters
> to cast their votes, it counts votes more accurately and more speedily
> and it will improve electoral administration.

Yes, it reduces all those complicated statutory rules to just “use
these machines and don’t ask awkward questions”.

> Our democracy is strengthened by vigorous debate. I would suggest that
> it is not too much to ask that this debate be informed and wild claims
> be subject to the scrutiny they deserve.

Indeed. This is worth quoting, especially in interviews with
spokesmen who go on about vote receipts and 70,000 tragically spoiled
ballots.

–Adrian.
venting prevents explosion