Brazil Announces Secure Email To Counter US Spying

Hugh Pickens DOT Com writes "Phys Org reports that Brazilian President Dilma Rousseff has announced her government is creating a secure email system to try and shield official communications from spying by the United States and other countries. 'We need more security on our messages to prevent possible espionage,' Rousseff said on Twitter, ordering the Federal Data Processing Service, or SERPRO, to implement a safe email system throughout the federal government. The move came after Rousseff publicly condemned spying against Brazilian government agencies attributed to the United States and Canada. 'This is the first step toward extending the privacy and inviolability of official posts,' Rousseff said. After bringing her complaints against U.S. intelligence agencies to the United Nations General Assembly last month and canceling a state visit to Washington, Rousseff announced that the country will host an international conference on Internet governance in April."

Comments

Re:Who wants email hosted by Federal Government?

[Marxist Hacker 42]

Especially not a federal government that uses Twitter to plan "secure e-mail"

Re: brace yourselves

[Anonymous Coward]

This is not "a foreign nations efforts against us".

This is "a foreign nations efforts to counter our efforts against them"

See the difference?

Re:brace yourselves

[CanHasDIY]

here come the liberal whiners to support Brazil and oppose the USA's effort to protect its own rational self interests.

here come the conservative whiners to support the USA and oppose Brazil's effort to protect its own rational self interests.

Hey, you know what? Fuck both of you for being part of the problem.

Re:Good luck with that.

[swillden]

with a secret signature that only the library knows for which crypto algorithm was used

Heh. Typical amateur security protocol design... can't even make it to the end of the second sentence of the description without handwaving some security through obscurity.

More importantly, your proposal addresses the part of the problem that isn't a problem -- the ciphers -- and ignores all of the rest, which is where the cracks show up: key management, protocol design, implementation quality and personnel. Much better to pick a small number of well-respected ciphers and then focus on all of the rest. You're still likely to fail against an adversary like the NSA, assuming they really care to put the effort in to read your mail, but you can make them work for it, and you can limit the amount of data they can get.