Jason Kitcat wrote:
> Hi,
>
> Thanks for the detail again.
>
> Ed Gerck wrote:
> >The 10 cents figure I quote is thus Safevote's estimate for the
> >total cost of an Internet
> >public election using Safevote's technology, per voter, including
> >licencing and auditing
> >costs. This figure is, of course, hopelessly invalid and
> >non-sensical for one voter -- or
> >even 1,000 voters. But it makes sense for 1,000,000 voters.
>
> Of course economies of scale come into play - as with most things.
> But the fact that each US county is likely to be using different
> systems surely impacts this???
This assumption may not be true -- it is certainly not true today, for
example. Further, after some inevitable initial flops or quasi-flops counties
are under public scrutiny to follow what works under a reasonable price
and thus time will help weed out flawed or costly systems -- especially
with the current public awareness in the US. These issues are now under the
public microscope and they may not wear out so easily in two months (as
they did in the past, for example when Kennedy was elected).
Safevote is making public its specification and some local vendors are
looking into using this technology much in the same way that one uses
a plug-in -- the whole voting product is still much larger because you
still have to include voter registration, ballot formation, reporting, training,
archival services, tech support, installation, and so on.
Also note that in a conventional voting system, the process of registering
voters and producing voters lists and/or voter credentials often accounts for
more than fifty percent of the overall cost for administering elections.
(See Harry Neufeld, "The Range of Advanced Technologies Available for
Election Organizations," in Let's Talk About Elections, ed. Carl W. Dundas,
London: Commonwealth Secretariat, May 1997). Thus, an improvement towards
automation of the process by means of computers and networks may have a
considerable effect in public elections, provided that the requirements for
privacy, security and integrity are met -- way beyond mere voting and tabulation.
> >Other Internet voting proposals may quote different values, of
> >course, but Safevotes
> >technology can bring the cost per vote to the range of $0.02 to
> >$0.10 -- as a function of
> >auditing requirements and satisfying the assurance needs of public
> >elections. These are
> >estimated values but they are not pulled out of "thin air". They are
> >quite consistent with
> >the average cost of comparable transactions in banks and private
> >sector Internet voting.
>
> Fair enough. However do this figures include the capital expenditure
> of hooking up internet connections, buy routers and buy computers? I
> accpept these are one-off costs but I'm interested to see how they
> impact the cost/benefit analysis potential users will be doing.
Since the technology works with off-the-shelf PCs (as long as they obey
some minimum requirements), counties may turn their PCs into voting
machines temporarily (as we did in Contra Costa, BTW) and PCs may
be rented for a weeek or so. In the US, there is one public election almost
every day on average -- so, machines can be reused all year round.
Surely, initial capital expenditure may offset initial costs (say 20 cents per
voter in the first year) but companies can also provide the equipment buffer
needed by counties, without any impact on election integrity. We are talking
here about standard PCs that can be used in any state with possibly differently
certified software, for many different uses (voter registration, vote casting,
remote ballot box, tallying, reporting, etc.) in elections, not some special DRE
machine which can only be used to count votes and only in some states.
And even if initial costs are 2x or 3x the expected near-term cost of 10 cents
per voter, this is still a 10x savings over current costs! Not to mention an
infinite improvement over current accuracy and reliability ;-)
Cheers,
Ed Gerck
