News for hackers. Stuff that matters.
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by sp332 on Tuesday October 02, 2012 @05:06PM (#4604303)
Here's a competing author's take on the final round of this competition: http://news.ycombinator.com/item?id=4564190
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by tptacek on Tuesday October 02, 2012 @05:26PM (#4604424)
I don't think I understand Schneier's point here. This is only the second time (AFAIK) that NIST has held a contest like this to standardize a crypto primitive. In the first contest, there was a clear and urgent need to retire DES. In the case of SHA-2, it was clear going in that there wasn't a "speeds and feeds" problem with SHA so much as there was a gathering unease about the basic design of the family of hashes from which SHA-2 is derived: MD4 and MD5 are broken, and SHA-1 isn't looking long for this world. Meanwhile, hash functions are not as well studied as block ciphers and number theoretic crypto, and the pace of development in hash research is increasing.

There is value simply in diversifying the gene pool of hash functions.

Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by apawloski on Tuesday October 02, 2012 @06:30PM (#4604728)
I think it was just a case of Schneier being a bit of his usual contrarian self. But is he correct in his assertion that the security of SHA-3 is not going to be much greater than that of SHA-2? (I don't mean that as a leading question.)
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by tptacek on Tuesday October 02, 2012 @06:32PM (#4604742)
What does "greater" mean? I buy into the idea that the attack that breaks SHA-2 is unlikely to break SHA-3, and vice versa, and that's apparently a key design goal for the standard.
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by B-Con on Tuesday October 02, 2012 @06:58PM (#4604852)
"greater" here means that we would think that SHA-2 is easier to break than SHA-3. Technically speaking, it's a guess, but I think that the general consensus is that it is more likely that SHA-2 will be broken before SHA-3.
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by kbd on Tuesday October 02, 2012 @06:38PM (#4604769)
Here are his comments about this announcement:

http://www.schneier.com/blog/archives/2012/10/keccak_is_sha-...

Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by tptacek on Tuesday October 02, 2012 @06:43PM (#4604790)
Note that Schneier hedges in a very similar matter about AES, suggesting in _Practical Cryptography_ that XSL attacks made him dubious of AES's long-term security --- or at least, that if you're "paranoid about your data", that you should use Serpent instead of AES/Rijndael.
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by pbsd on Tuesday October 02, 2012 @08:11PM (#4605106)
Does Schneier really recommend Serpent to avoid XSL-type attacks? As far as I remember, XSL also applied to Serpent.
Re:NIST Selects Winner of Secure Hash Algorithm (SHA-3) Comp... by tptacek on Tuesday October 02, 2012 @10:16PM (#4605491)
Not directly, but his specific concern with AES is XSL, and he suggests Serpent for more security-conscious users.