3 ms·
"We don’t need crypto unicorns so much as we need diligent engineering to deploy the crypto we already have." If we took this seriously, we would all still be
by peterwwillis 12y ago
"We don’t need crypto unicorns so much as we need diligent engineering to deploy the crypto we already have."
If we took this seriously, we would all still be using modified versions of DES, which, in a twisted way, we are. This isn't a good argument for ignoring new implementations of different cryptosystems.
"If a new crypto tool is first announced in a press release or popular science magazine, don’t use it."
This is far more accurate, but i'd extend this even further and define it as: "Don't use any new crypto." (Unless it comes from someone with high social influence in the sphere of cryptography research, of course)
- kedean 12y agoI don't think DES is an accurate comparison here. The world abandoned DES because a) everyone suspected it had a backdoor, and b) it was demonstrably broken by the EFF in 1999. We rely on a primitive until someone who's job it is finds holes in it. Engineers don't need to be inventing new encyrption.
- tptacek 12y agoDES was abandoned because its key size was too small. It also has a block size that's too small for safety, which is why we don't use Triple DES or Blowfish as bulk ciphers either. Ever since Biham and Shamir published the differential cryptanalysis, we've known that NSA didn't backdoor DES, but rather helped strengthen it against that attack. Concerns over a backdoor had nothing to do with it.