Re: Guess they didn't use an iSeries... Windows for a voting solution?Puleeeeze!

The thumb ink is suposed to prevent you from voting more than one time.

Jim Franz wrote:

the thumb ink proves you voted... but not who for..
jim

----- Original Message ----- From: "Booth Martin" <booth@xxxxxxxxxxxx>

To: "Midrange Systems Technical Discussion" <midrange-l@xxxxxxxxxxxx>
Sent: Tuesday, April 04, 2006 4:46 PM

Subject: Re: Guess they didn't use an iSeries... Windows for a voting solution?Puleeeeze!

I still am fascinated by the indelible ink on the thumb that other
countries use.

Raul A. Jager W. wrote:

The simpler the voting machine, the easier to audit it.  But, it still
requires a "guru".  The best is to print a piece of paper that the voter
can read, see that it has only the name of his candidate, and drop in an
urn.
Count manualy to validate the results.
___________________________________________________________________________________________
Peter Dow (ML) wrote:

But if verification is passed back through the voting machine, what's to
prevent it from changing it?  I vote for Bulbous Frumpwort, it gets sent
to the iSeries as Jamus Plugly, the iSeries sends back that I voted for
Jamus Plugly, and the voting machine shows me I voted for Bulbous
Frumpwort.  How did having an iSeries in the loop help?

*Peter Dow* /
Dow Software Services, Inc.
909 793-9050
pdow@xxxxxxxxxxxxxxx <mailto:pdow@xxxxxxxxxxxxxxx> /

Larry Bolhuis wrote:

Some time back some friends and I were talking about how the heck you
could possibly trust winDOHs anything to actually count votes in such a

way that Reliably is insured and Fraud is impossible. We decided (rather

quickly) that you just couldn't. With i5/OS however it seems to me that
the answer is so simple that it's elegant. Use any freekin front end
that you want to do the voting, web browser, 5250 screen, windows
program with ODBC, whatever, don't care. In any case when the vote hits
the database (which will *ABSOLUTELY be journaled) you scrape the
journal. The Journal contains everything the voter did and you use that
transaction to spit out the voters ticket, NOT the voting machine. In
other words what the voter gets out of the voting machine is not simply
a reflection of what they put into the machine, rather it's downstream
of the actual votes that were posted. If the voter then verifies the
ballot in their hand is correct, shazamm no fraud, and you vote is
already in the DB. If required a list of votes could also be printed at
the precinct as the tickets come out but hey, it's in the journal so
it's in the DB. Done deal.

Another example of the System i, and i5/OS being an integrated solution,

and kicking the **** out of M$.

- Larry

Joe Pluta wrote:

Cook County is embroiled in controversy yet again, although this time it's the machines not The Machine. Nearly a week after elections, votes are still being counted - votes that were supposed to be counted automatically by the Sequoia Voting Systems electronic balloting machines. And while Sequoia insists that their systems are more secure than Diebold's because they don't use Windows, as it turns out this is only the voting machine

itself.  The back end "WinEDS" system (which does the tallying and

reporting) uses Microsoft SQL Server as its database, and evidently it has a

problem counting a million or so votes.

DB2/400 posts a million transactions in seconds (or maybe minutes, but
certainly not days!).

I'm being a bit facetious here. From what I could see, voters vote on a

touch screen, then the touch screen results are printed out on special

forms, which are then fed into a reader. It's the reader that seems to be

the bottleneck.  But I'm just guessing.