Hi Torben,

In discussion with our security folks, we felt that SHA-256 was more
secure. Of course there's a trade-off to be made about hash size as
well as choice of hashing algorithm.

The improved security of SHA-256 doesn't (only) come from its
increased output size. A truncated SHA-256 benefits from the enhanced
compression function used in the SHA-2 family of hash functions.

thanks,
Jochen

On Wed, Feb 3, 2010 at 6:46 AM, Torben Weis <[email protected]> wrote:
> Hi,
> while implementing federation for the C++ server, I realized that wave uses
> SHA256 for the rolling hash and then keeps only the first 20 bit.
> ----------------------------------------------------------------------
>
> /**
>
>  * Utility class for creating Hashed Versions with Crypto dependencies, this
> is intented
>
>  * for "full" i.e. not lightweight implementations.
>
>  */
>
> public class HashedVersionFactoryImpl extends HashedVersionZeroFactoryImpl {
>
>   /** The first N bits of a SHA-256 hash are stored in the hash must be <=
> 256 */
>
>   private static final int HASH_SIZE_BITS = 160;
>
> ----------------------------------------------------------------------------
>
> Why? In this case you could have used SHA1 in the first place. Is there any
> attack vector against SHA1 that cannot be used against
> SHA256_crippled_to_160 ?
> Or is this an attempt to save 96 bits? In this case better drop XMPP and
> donate some of the saved bits to SHA256 :-)
> Torben
>
> --
> You received this message because you are subscribed to the Google Groups
> "Wave Protocol" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected].
> For more options, visit this group at
> http://groups.google.com/group/wave-protocol?hl=en.
>

-- 
You received this message because you are subscribed to the Google Groups "Wave 
Protocol" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/wave-protocol?hl=en.

Reply via email to