Prof. Rivest of MIT’s CSAIL program: he may not be a rock star, like Google’s young Turk, Orkut Buyukkokten. But he has Brazilian fans as well.
Voto Seguro, a Brazilian e-voting security watchdog, has been evaluating its leader’s recent participation in a hearing before a committee of the federal Congress, as well as organizing and posting the materials it presented to legislators. See
- Brazil: E-Voting Issues Get An Airing
- Brazil’s E-Vote: Fernandes on Founding Faith
- COPPE-Out? The COPPE-UFRJ E-Voting Report
- Brazil: Prof. Rezende Links Jack Abramoff To Elections Fraud
The creation of a permanent subcommittee on electronic voting in the Tupi equivalent of our Judiciary committee is being celebrated as a victory for the bean-counter patrol.
And I, for one, congratulate Mr. Brunazo and his colleagues. These Brazilians, more than just about anyone else in the world, have studied the problem from a global perspective, and have a lot to offer us in that respect.
If only that damned language barrier did not get in the way. I will keep trying to do a bit of volunteer translating from time to time.
Lately, I have been listening to the audio recording of the hearing, posted to the group’s open forum, and see that a transcript will soon be available as well, for example.
In the meantime, some bullet-point takeaways from the presentation, some of which represent new, to me, factoids, to be fleshed out.
“Só no Brasil uma única autarquia acumula os três Poderes (Executivo, Legislativo e Judiciário) sobre o processo eleitoral”
Brazil is the only nation in the world with an autonomous election authority that concentrates executive, legislative and judicial powers under one roof.
“Os problemas nos logs ocorreu em todo o Brasil. No Amapá, por exemplo, os logs indicam uma diferença de 21 mil votos (6,8% do total) em relação ao resultado oficial”
The problems with the log files occurred all over Brazil. In Amapá, for example, the log files show a discrepancy of 21,000 votes (6.8% of the total votes cast) in relation to the official result.
The Voto Seguro folks have also gotten their hands on a parallel case from the 2002 elections, presented by a PT federal deputy.
Mr. Brunazo has reported that the current ruling party was nervous about having the legitimacy of its gains at the polls questioned by noise machinery, and therefore sealed a bargain with the TSE to avoid publicity over shortcomings in the software development process.
Which, if true, could well be as dunderheaded and short-sighted as Mr. Brunazo says it was (COPPE-Out? The COPPE-UFRJ E-Voting Report).
I look forward to getting more facts on that whole transaction.
“Ronald Rivest, o inventor da técnica de assinatura digital, afirma que assinatura digital não é suficiente para garantir a segurança do voto eletrônico. Recomenda a materialização do voto (voto impresso) em paralelo ao voto digital”
Ronald Rivest, the inventor of the digital signature technique, says that digital signatures are not sufficient to guarantee the security of electronic voting. He recommends a material record of the vote (the “printed vote”) be used in parallel with the digital vote.
Rivest invented the MD5 checksum procedure, which if you have hacked around some as I have you will recognize as a method for checking binary files for corruption.
But checksums cannot distinguish intentional tampering from file corruption from, say, garbled data transmission, as Prof. Fernandes of Brazil’s ITA has noted in the Alagoas case.
And they can be spoofed.
“Os maiores especialistas em segurança de dados afirmam que é mais fácil transmitir com segurança 1 milhão de dolares por meio eletrônico do que computar com segurança 1 voto digital secreto.”
Authorities on data security say that it is easier to securely transfer $1 million electronically than it is to securely compute a single secret ballot cast electronically.
Sexy talking point, that, but maybe a little too wonky.
“O Relatório Brennan, da New York University, listou 120 formas diferentes de se fraudar sistemas eletrônicos eleitorais”
The Brennan Report, from New York University, lists 120 different ways of defrauding electronic elections systems.
The report from the Brennan Center, published in October 2006, is titled The Machinery of Democracy: Voting System Security, Accessibility, Usability, and Cost.
The Center has also launched a campaign on federal election reform, and assisted the Working Families party in filing a lawsuit against paperless touchscreen voting here in New York. See Brennan Center and Supporters Call on New York City Board of Elections to Reject “Full-Face” Touch-Screen Machines.
And we did.
Working Families, by the way, is the only political party I currently support. I am not a member, but I donate when I can, and am following their work with great interest. They are interested in serious things. More power to them.
So long live the Brazuco-Gothamite good governance collaboration.
Disclosure: I am biased in favor of Brazilians. I even married one.
The rotten ones are incredibly rotten, but in general, that zebu and farofa diet and “crossroads of the world” cultural stew produces some extremely hardy and compelling human beings. Saravá!