5 ms·

-----BEGIN PGP MESSAGE----- Version: GnuPG v1.4.7 (Darwin) Comment: Remember Lexington Green! hQENA+4fGjU8+s3CEAP45kAPM5FMQJpNHb+0ve98gqLhXwwmQ0e6V

• randallsquared 17y ago
  A temporary stopgap, at best, especially in a world where so few people use it. Encryption is maximally useful when people don't see -- or it doesn't matter that people see -- that you're using it. My guess is that people who really, really want to protect what they saying and that they're saying it use nonsense spam as a steganographic medium.
  • philwelch 17y ago
    Maybe they already are--what other purpose would there be for nonsense spam?
    • bd 17y ago
      Nonsense spam messes with adaptive spam filters. They learn bad rules, thus it becomes harder to identify real spam. ---- Edit: I don't know how effective it really is, but this seems to be the motivation for why spammers at least try this. Many more details here: http://en.wikipedia.org/wiki/Bayesian_poisoning http://en.wikipedia.org/wiki/Bayesian_poisoning
      • philwelch 17y ago
        Wouldn't you need significantly more noise spam than signal spam for that to work? Is that what we indeed observe?
      • randallsquared 17y ago
        But since that's a public good (for spammers) it's hard to understand why they'd do it instead of sending a profit-making campaign, any given time.
• mschy 17y ago
  If I ran an oppressive, evil regime, one of the first things I'd arrange would be traffic analysis of encrypted or otherwise suspicious email.
  • lt 17y ago
    If everyone was sending encrypted emails it would be unpratical, though.
    • pyre 17y ago
      You can still analyze that Bob emailed Cindy and that Cindy emailed Marge, even without knowing the content of the emails. IIRC, the FBI has some sort of software that does this with telephone communications to identify 'networks' that was originally developed to combat the mafia.
      • rdtsc 17y ago
        Yes, and you can break Bob and Cindy's fingers to get their passphrases. That is a very easy step. That is why steganography should be used. There shouldn't even be an ecrypted text for them to try to decrypt, at least as far as they know. Because once, there is, they can always get the password. The next best thing is is fake encryption or some kind of nested encryption, so that the outer layer decrypts to something plausable but ultimately benign. A password doesn't even fall under 'free speech', so even in a free country like US one can get slapped with 'obstruction of justice' if one doesn't provide a password. I'll leave it to your imagination to what happens in other, more oppressive countries...
    • catzaa 17y ago
      That may be true for most emails. But I guess that large webmail providers can band together and ensure that e-mails between their systems are encrypted (i.e. all messages sent by gmail to other gmail addresses or hotmail addresses are encrypted). Then users would just use HTTPS to check their mail and everything would be encrypted. This is not perfect, but it is at least a stopgap.
  • 17y ago