Unfortunately, hand counting and transposition of hand written numbers
is probably as prone to error as the deliberate mischief.
I will never forget coming out of school with my business degree and a
lot of background in operations research. My first job at IBM was
production planning and I had all sort of visions of how I was going to
wow the company with all of the slick optimizations I was going to
develop for the parts planning and production processes. Then the
too-smart-for-his-britches kid ran into the reality which was that our
much balleyhooed, computer planned and controlled parts planning process
was totally crippled by manual data entry errors. There was probably
not a single one of the 60-80,000 parts in the inventory that was within
5% of having a correct quantity on hand. No way could you risk running
with a lean inventory as you'd likely run out of most everything in
So, I went on a religious crusade to get the numbers right and that did
more than any sophisticated process to make things run smoothly.
The only way you can trust a manual count is to have multiple people
doing it in parallel and they all have to agree. And if you feel
someone or something is diddling with the numbers at any level you need
to have more than one count.
Winsor Crosby wrote:
> I am no expert, but that might work, if you could guarantee no
> tampering with the optical reader and resisting the pressure for
> fast, TV reportable results with a central hookup of optical readers
> to a central compiler. Maybe a certified hand count with an optical
> reader check? There would be some mischief possible in variances
> between layout of the touch screen and the paper ballot that could
> cause enough confusion for people not to check whether the ballot
> represents the choices made on the screen.
> You are probably able to think of more ways to break it than I can. I
> am a babe in the woods in places like eBay. :~)
> I was surprised when Andrew said that Australia used a paper ballot.
> Evidently Canada does as well. Since voting is a local process I
> wonder why the need for expensive hardware and software except as a
> means to provide instantaneous results to provide drama on
> television. Certainly the size of the population would have nothing
> to do with it since final results would be a few sums of sums.
> Long Beach, CA
> On Aug 26, 2007, at 11:35 AM, Chuck Norcutt wrote:
>> I've always thought the best system would be a hybrid. Computer
>> controlled touch screen for easy presentation, selection and
>> (if necessary) of your choices. But the computer would be stand-alone
>> and not connected in any way to any network. It would print out a
>> readable paper form as the ballot. The paper form would be counted
>> by a
>> separate optical reader. No manual labor, paper ballots for safety
> List usage info: http://www.zuikoholic.com
> List nannies: olympusadmin@xxxxxxxxxx
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx