Big News in Crypto

[rookcifer]

SHA-3 winner has been announced. Schneier's entry of Skein did not win. Instead Keccak won. This is the second time a Schneier entry has made the finals and lost (the other was with Twofish who got beat out by Rijndael to become AES).

For those who don't know, SHA-3 will become the new hashing algorithm standard soon.

[Stonecold1995]

Cool! When do you think applications like TrueCrypt will adopt this?

Sadly I don't see SSL-"secured" sites using this any time soon. So many sites still use horribly weak hash algorithms, even worse than SHA-1 most of the time (as far as I know, correct me if I'm wrong).

[rookcifer]

Cool! When do you think applications like TrueCrypt will adopt this?

NIST will have to finalize everything first, and then release specifications, etc. I mean the design is already known and open, but they will have to release docs and implementation info, etc. I would say it will take most software close to a year to start implementing it, though maybe a little sooner.

Sadly I don't see SSL-"secured" sites using this any time soon. So many sites still use horribly weak hash algorithms, even worse than SHA-1 most of the time (as far as I know, correct me if I'm wrong).

Well, MD5 is what you need to avoid. It is utterly broken. SHA-1 is still thought to be secure, but has chinks in its armor. SHA-2 is still very secure by all accounts.

[Lars Noodén]

Doesn't Ubuntu still sign the packages with MD5?

[Stonecold1995]

Doesn't Ubuntu still sign the packages with MD5?

Yes.

[zombifier25]

I thought they also use SHA-2.