Benjamin Kaduk has entered the following ballot position for
draft-ietf-cose-x509-08: Yes

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
https://datatracker.ietf.org/doc/draft-ietf-cose-x509/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

(I posted a PR on github to fix a few editorial nits.)

There are some issues still in the open state at
https://github.com/cose-wg/X509/issues
Some of them have been fully or essentially addressed and could probably
be closed, but a couple seem to still be noteworthy:

https://github.com/cose-wg/X509/issues/30 and
https://github.com/cose-wg/X509/issues/31 cover related issues, relating
to the "trust relationship" between signer and host of URI (that we say
needs to be authenticated), and whether there are similar considerations
relating to other header parameters.

The answer seems to be that "yes, there are sometimes such
considerations", and it would be okay to document them if we have a
concise explanation.  That COSE mandates x5u appear in the protected
headers is a divergence from JWS, but it would feel out of place to
attempt to amend JWS in this document; the other header parameters can
appear either in the protected or unprotected buckets, which allows for
pretty much all use cases.  JWS does have some text relating to header
parameters that must be integrity protected "if the information that
they convey is to be utilized in a trust decision", which is vague
enough that it may not actually be helpful to replicate that
terminology.  (We did not seem to have immediate agreement on what it
meant when this topic was discussed in the WG.)

There is perhaps just one remaining point in
https://github.com/cose-wg/X509/issues/29 ; whether we should be more
explicit that 'x5t' refers to the end-entity cert.  I'd be okay with
doing so, but it doesn't feel particularly critical.



_______________________________________________
COSE mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/cose

Reply via email to