Re: [openpgp] Fingerprints and their collisions resistance

[Andrey Jivsov <[e-mail]>]

On 01/03/2013 01:03 AM, ianG wrote:
...
> Now that SHA-3 is settled, it seems reasonable to clean out all of the
> SHA-1s.
>
...
>
> On another related point - have the MD numbers been allocated for SHA3
> in its various guises?

In the process of writing such a draft I noticed that the only place in 
OpenPGP where SHA1 is used in collision resistance sensitive way without 
the possibility to change it is fingerprints. For this reason OpenPGP 
fingerprints stand out because these are the data structures that 
technically make (or soon will make) RFC 4880 non-compliant with 
recognized standards. I would separate the issue of fingerprints 
depending on known SHA1 weaknesses from any other task that can be 
categorized as "OpenPGP V5".

Speaking of the Keccak in OpenPGP draft, I thought that it would be 
important to gather the feeling about the path of fixing the 
fingerprints. These issues are more dependent as seems. For example, if 
you have to use SHA-3-384 for fingerprints, it affects the decisions 
about SHOULDs for hashes elsewhere.

I have this Keccak in OpenPGP darft written, waiting to for the NIST to 
publish SHA-3 and the OIDs assigned.