Re: [dane] draft-ietf-dane-openpgpkey-00 comments

[Viktor Dukhovni <[e-mail]>]

On Tue, Apr 22, 2014 at 10:43:18AM +0200, Petr Spacek wrote:

> 3) Algorithm agility:
> It is clear to me that SHA2-224 hashing is there "just" for privacy and
> nothing else. Still, I think it would be beneficial to have algorithm
> agility built-in.

In this specification sha2-224 does not play a security role.  It
is used not for privacy but rather as a short-enough and yet strongly
collision resistant representation of potentially longer email
addresses that would not fit into a DNS label.  It is expected that
the number of email addresses with SMIMEA or OPENPGP keys in any one
domain will be substantially less than 2^{112} (~ 10^{34}).  A domain
with 10^9 users will have two users with the same lookup key
with probability roughly 2^{-62} or ~10^{-16}.

There is no need for "algorithm agility" here.  This is a lookup
key construct, not a tamper-resistant signature.  In fact multiple
algorithms would be entirely counter-productive in this context.

-- 
	Viktor.