Pedro Rezende
Wrote paper: "Electronic Voting Systems: Is Brazil ahead of its time?"
afraid not to be able to enter the US
organization asked for a substitute
He suggested I represent him
we share the same objectives
we have a different approach
"I do not agree with you but I will fight for your right to say what you think"
goal talk
quick explanation urna
overview paper Rezende
my talk
short introduction
fraud in 1994
first election in 1996
demo
photos
urna
urna2
detail: printer
detail: side view
detail: printouts
parallel voting
simulation
goto www.tse.gov.br
choose Eleições > Urna Eletrônica > Simulação
goto bottom and click on "Vote aqui"
a valid sequence is "91101 CONFIRMA 91 CONFIRMA"
process
flash cards
zeressima
bulletin
floppy disk
the logistics are incredible
population 170 million
voters 110 million
450000 urnas
2 million poll workers
progress
1996: 33% (cities > 200.000)
1998: 67% (cities > 41.000)
2000: 100% = 109 million
first country with fully automated elections
general observations
lottery wouldn't work
physical ballot wouldn't work
receipts are problematic
MANY candidates
braille/headphones
voter registration
4 million re-registered
1 million cancelled
testing by TREs
paper Rezende
scandal in the Senate in 2001
call for printing
Senator Requião
Judge Jobim
change in law
printing on 3%
printing 100 % for 2004
2002 elections included prototype for printing
led to terrible delays
"recount could undermine credebility"!!!!
alternatives
parallel voting
software auditing
hashes and signatures on the code and executables
signatures on the ballots
changes for 2004 elections
printing to be eliminated
signatures on ballots to be implemented
laws were approved without any technical discussion whatsoever
suspicious fiddling with procedures
Jobim
politician
contributed to writing the constitution in 1988
judge highest federal court
admitted including to articles that were not voted
currently president of the highest court
urna used worldwide
Argentina
Paraguay
Mexico
Ivory Coast (?)
conclusion
higher complexity/cost should NOT be used as an excuse not to implement vv methods
it seems that enterprises dominating the e-voting market are exactly trying to uphold and advance this flawed reasoning
Brazil is ahead of its time in showing the importance of this debate
It will be very important to see how these topics develop in the USA
report BSC
observer during 2002 elections
with Prof Custódio
access as a political party
structure
security requirements for voting
classify voting technologies
different voting techniques give different results, esp. wrt. auditability
application to the urna
conclusions
urna is improvement
seems secure, but we cannot be sure
linking identity voter X vote theoretically possible
security and transparency based on software difficult/impossible
correctness software
official software
parties have no effective means
too much trust on the officials
physical evidence
printing
something else
paper audit trail
most voters don't verify
nobody believes it is necessary
motivating people for an audit recount difficult
cheating seems easier with it
TSE claims it is undoable
future trends
extension Chaum's scheme
auditability through hardware
simplify source code
state
events
finite state machine
statistical analysis of fraud
About
Title
Brazilian Urna
Subject
-
Company
-
Author
-
Version
26
Comments
-