Re: [keyassure] publishing the public key

[Peter Gutmann <[e-mail]>]

[Several of my posts didn't make it to the list, these are re-posts of earlier
 messages]

Henry Story <[e-mail]> writes:

>In draft-dane-04 [1] one can currently only publish either a signature or a
>certificate in the DNS. That is the only allowed formats of the resource
>record are as specified in section 2.2
>
>[...]
>
>Why not publish the only piece of the certificate that is important in public
>key cryptography: the public key.

Whatever format gets used, could we restrict things to just a single form, or
at most two forms?  At the moment we've got two types of certs, two hashes of
certs, and a raw public key (or, most likely, two of them).  In practice
there's going to be some de facto standard that everyone uses (e.g. SHA-1
hashes of certs for "fingerprints", which every browser seems to support even
though there isn't even a standard specifying them), and adding a whole pile
of formats is going to reduce interoperability or (more likely) lead to
implementors ignoring all but whatever ends up as the de facto standard.  I'd
use the cert fingerprint (which is already the de facto universal identifier)
as a MUST and leave the rest as options.  TLS already requires that you send
the full cert, TLS implementations already generate a cert fingerprint, so all
that's left to implement besides (shudder) the res_query() lookup is a
memcmp().

Peter.