Re: [dnsext] Adopting draft: draft-hoffman-dnssec-ecdsa-04.txt

[Paul Hoffman <[e-mail]>]

On 1/5/11 10:52 AM, Basil Dolmatov wrote:
> I did not make any GOST references, moreover I did not imply any GOST
> peculiarities in my comment, it is true for _any_ algorithm (the fact
> that GOST belongs to "space of open standards" too is worth mentioning,
> but irrelevant).

GOST limits which hash algorithm can be used in signatures to one 
choice, so that is indeed relevant.

> In any space of cryptographic standards the strength of algorithm is
> measured not by "bits of strength" or "bits of key or hash length" but
> by the number of operations which should be performed for given attack.
> If one bothers to compare the algorithm strength more accurately then
> the product of (operations)*(memory required) is used.

Both are used as measurements, and the one that is most commonly used in 
discussion is equivalent bit strength.

> Other figures have mostly marketing meaning rather than technical one.

That was not the IETF consensus when we adopted RFC 3766, which was 
heavily reviewed in the IETF's cryptographic community.
_______________________________________________
dnsext mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/dnsext