Hi Lucas,

Thank you for your comments, I am attempting to address them here:
https://github.com/cose-wg/draft-ietf-cose-dilithium/pull/14

May I add your name (Lucas Prabel) to the acknowledgements section?

Regards,

OS

On Thu, Nov 21, 2024 at 4:57 AM Lucas Prabel <lucas.prabel=
[email protected]> wrote:

> Hello,
>
>
>
> I am pleased this document exists and think support for ML-DSA in JOSE and
> COSE will prove useful, especially now that ML-DSA is being specified by
> more and more working groups. After reviewing it, here are my comments:
>
>
>
> *Section 3*
>
> - typo: “paramaterized” --> “parameterized”
>
> - In Table 2, I would call the 2nd column “value” instead of “alg”.
>

Thanks I took both of these.


>
>
> *Section 4*
>
> - I agree with Neil Madden and think this section is a bit underspecified.
> In particular, I think it needs precise specification of the Key Parameters
> associated with AKP.
>

I agree, I have attempted to address this feedback as part of the same pull
request.


> - In the given COSE example (Figure 2), key parameters are called “pub”
> and “priv”, but called “public_key” and “private_key” in Section 8.14 and
> 8.15.
>

Thanks for pointing this out, I aligned them.


> - typo “the registration of the following algorithms in [IANA.cose]” → “the
> registration of the following key type in [IANA.cose]”
>

Fixed.


>
>
> *Section 5*
>
> - I think the first sentence can appear a bit confusing in its use of
> "private key" as the term refers to two different things in its 2
> occurrences (in the 1st one, "private keys" encompasses both the seed and
> its expanded representation, while in the 2nd, it only refers to the full
> expanded key). FIPS 204 seems to be a bit clearer by using "private key"
> exclusively to mean the full expanded private key.
>

I agree, hopefully it is better now.


>
>
> *Other*
>
> I would like this document to clarify its stance on the use of HashML-DSA
> from FIPS 204. Recent IETF discussions suggest moving away from including
> HashML-DSA as an option, focusing instead solely on ML-DSA without
> pre-hashing. For this document, while I think it was agreed that any
> pre-hash specifications would be detailed in a separate draft if needed, I
> think we should add a short sentence stating that this draft doesn't
> include the pre-hash option. Besides, I think we should add some words
> about how to handle the ML-DSA context string.
>

I agree, I hope what I have is sufficient to address this.


>
>
>
>
> Thanks for your work,
>
> Best,
>
>
>
> Lucas
>
>
>
> *From:* Michael Jones <[email protected]>
> *Sent:* mardi 19 novembre 2024 17:48
> *To:* [email protected]
> *Subject:* [COSE] WGLC for draft-ietf-cose-dilithium
>
>
>
> Hi all,
>
>
>
> This message starts the Working Group Last Call (WGLC) for
> https://www.ietf.org/archive/id/draft-ietf-cose-dilithium-04.html (ML-DSA
> for JOSE and COSE), as was discussed at IETF 121 in Dublin.  The WGLC will
> run for two weeks, ending on Tuesday, December 3, 2024.
>
>
>
> Please review and send any comments or feedback to the working group.
> Even if your feedback is “this is ready for publication”, please let us
> know.
>
>
>
>                                                                 Thank you,
>
>                                                 -- Mike and Ivaylo, COSE
> Chairs
>
>
> _______________________________________________
> COSE mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>


-- 


ORIE STEELE
Chief Technology Officer
www.transmute.industries

<https://transmute.industries>
_______________________________________________
COSE mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to