On Thu, 9 Oct 2025 at 19:58, Simo Sorce <[email protected]> wrote:

> On Thu, 2025-10-09 at 19:20 +0530, tirumal reddy wrote:
> > Thanks Simo for the review. Please see inline
> > >
>
> > > 3) The use of AKP here really shows how AKP is not a great option
> > > (IMHO), as it forces a distinction that makes little or no sense, given
> > > there is no risk of confusion or misuse between direct agreement and
> > > key wrap uses with the same keypair. I can see how some protocols may
> > > want to use either/or with the same key depending on the other
> > > circumstances (section 10).
> > >
> >
> >
> > The WG needs to discuss and decide whether to use AKP or OKP for this
> purpose.
> >
> > The key trade-off seems to be between enforcing algorithm binding in the
> key structure to reduce misuse and keeping flexibility to avoid layering
> issues. If JWK starts enforcing operational policies (like “this key must
> only be used for this algorithm”), it may interfere with higher layers
> (such as application logic or key management) that should be making those
> decisions. One possible balanced approach would be to continue using AKP,
> but make the "alg" field optional when the key is used for key agreement.
>
> The problem is that once JWKs carry the algorithm the only option is
> not whether or not alg is used, but whether or not multiple algorithm
> should be considered equivalent and interchangeable for some mechanism,
> and I believe that will not be a good compromise, which is why I
> brought it up here. I do agree that the WG really need to think hard
> whether it is proper to try to enforce policy mechanisms at the storage
> format/information exchange level, or not (I think not).
>

I suggested the above option to help resolve the WG discussion on whether
to use AKP or OKP. The draft can be updated to list the respective
pros/cons and allow implementations to make a choice.


>
>
> > >
> > > 4) I object to the characterization that forcing the use of SEED only
> > > as a private key promotes interoperability, so I would drop that
> > > preamble from the phrase that defines what the specification mandates
> > > in this area (last paragraph of section 10).
> > >
> >
> >
> > No, it simplifies key transport and allows keying material to be
> encrypted, exported/imported between HSMs without ambiguity.
>
> I understand that this is the position of some people in here, but I do
> not subscribe to the idea that this is better for interoperability.
> Because it prevents the exchange of keys from systems that do not
> preserve the SEED, or can't extract it.
>
> (There is no consensus that NIST standards, as written today and until
> they are amended, allow for the extraction of a SEED from a
> cryptographic module, and therefore standardizing on SEED means there
> may be interoperability issues with modules that allow only the "NIST
> defined" secret key to be extracted and do not preserve the SEED.)
>
> I am not asking to remove the requirement to standardize on SEED only,
> just that it is not incorrectly labeled as an interoperability choice,
> because it is not.
>

The decision that it is for interoperability is already discussed and
finalized in Section 4 of draft-ietf-cose-dilithium
<https://datatracker.ietf.org/doc/draft-ietf-cose-dilithium/?utm_source=chatgpt.com>.
This
document aligns with that decision

-Tiru


>
> Simo.
>
> --
> Simo Sorce
> Distinguished Engineer
> RHEL Crypto Team
> Red Hat, Inc
>
>
_______________________________________________
jose mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to