[cryptography] NIST Randomness Beacon

Adam Back adam at cypherspace.org
Tue Nov 12 04:10:13 EST 2013


(Top posted, so sue me, my text explains itself without the history).

Thats a big cc list.  I think you could create a beacon with bitcoin hash
chain by having miners reveal a preimage for 6 old, consecutive blocks where
the newest of the 6 old blocks is itself 6-blocks confirmed.  (ie reveal
preimage on blocks 7-12.  The xor of those preimages defines a rolling
beacon (new output every block, just with reference to blocks 7-12 relative
to the current block depth).

The security against insider foreknowledge is not fantastic, as its relating
to the trustworthiness of the 6 random miners (which have probabilty of
winning relating to hashpower, which doesnt always relate to
trustworthiness).

Adam

On Mon, Nov 11, 2013 at 05:42:54PM +0100, CodesInChaos wrote:
>On Sun, Nov 10, 2013 at 9:54 AM, Andy Isaacson <adi at hexapodia.org> wrote:
>> For example, suppose you use the low bits of the bitcoin blockchain
>> hash.  An attacker with 10% of the hash power could probabilistically
>> attack such a system by chosing blocks with a specific value in those
>> bits;
>
>This can be avoided by running a sequential computation based on that hash. 
>For example by hashing it 2^40 times.  Obvious downside is that verifying
>that the computation was performed correctly is just as expensive (but
>parallelizable).
>
>Perhaps there is a function that's sequential and slow in one
>direction and fast in the reverse direction.


More information about the cryptography mailing list