[EM] Cubicle-society. Withdrawing Intermediate FBC-2.

[Paul Kislanko]

Mike Ossipoff wrote in part:

 

But here's something that many people seem to miss: Even when it isn't shown
that a count is fraudulent, the

count is still non-legitimate if it isn't vefifiable. A political system's
use of a non-verifiable machine-count

is ridiculous. 

 

Maybe it could be somehow possible for a machine-count to be verifiable.
But, untiil then, a handcount is

necessary. That's another disadvantage of rank-balloting voting
systems--Most of them aren't feasibly

hand-countable.

(End partial quote)

 

There are actually two aspects to verification: 1) were all votes cast
COUNTED (this has been a big problem in my precinct in Mississippi) and 2)
was the counting done correctly.

 

There's technology available that could be applied that does not require
"hand counting" of ranked ballots. I use a very weak version to conduct
"polls" for a college football/basketball/baseball "top 25 fans poll." 

 

The "was the counting done correctly" ("2") bit just requires that all the
ranked ballots that were counted be published. Once we have every ranked
ballot published anybody who knows the counting method and has a
programmatic implementation of it can use theirs to validate the "official"
counting program's results since they have the same input. So "validation"
only depends upon how available the ranked ballots are. 

 

A more subtle test of "legitimacy" is the "Did they count my vote or not?"
question. (1)

 

Here's what I would do to satisfy both objectives.

 

For each voting-place (precinct) define a "public key" that is unique to
that voting place. For instance 2^state-code times 3^county-code times
5^precinct number. (Credit to Godel for the "for-instance" - uniqueness is
guaranteed by the fundamental theorem of algebra.)

 

The machine that records the voter's ballot can generate a "private key"
unique to every ballot cast at that location (it doesn't matter how that is
generated as long as every ballot cast at that location has a unique
identifier) and prints that on the voter's "receipt" that confirms the
ballot was recorded.

 

The published record includes every ranked ballot with so that the accuracy
of the counting-by-whatever-method can be independently verified while
protecting the anonymity of the voter.

 

But any individual voter can find out which of their ballots is theirs by
providing their "private key" and the components used to create the "public
key" (i.e. I voted in this precinct in this county in this state) and if
their ballot is either not provided with those keys or does not match the
printed record of their ranked ballots then ANY voter would be able to
declare the election "fraudulent."

 

I guess there's a third angle to verifiability, namely that a vote was
counted under my voter ID should match my ballot.  This is actually the most
important to me as a voter. If a vote were "counted" under my voter ID I
need a way to see that what was counted was the same as the record I got
when I voted. I should be able to compare the "counted" ballot to the one I
submitted.

 

And that gets to why I think  "hand counting" is no longer useful as
"verification" - what is there to "hand count" when there are no paper
ballots except those printed by the machines that we're auditing? 

 

 

 

 

 

 

 

 

 

 

.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20120427/98228b95/attachment-0004.htm>