Hi, Thanks for all the hard work on the draft, and for the update. I wasn’t able to attend IETF126, but would like to add my vote towards “yes, I prefer aligning with LAMPS Composite document”. I believe that is the solution that is most palatable to adopters. After this change, I would love to see this draft go to WG last call!
I also am happy with the composite algorithm choices. Our algorithm of preference is ML-DSA-65-ES256. Thanks, Cathie On Thu, Aug 27, 2026 at 8:08 AM Lucas Prabel <lucas.prabel= [email protected]> wrote: > Thanks Lijun, > > > > The Editor’s Copy of the draft has been updated to take that into account: > ietf-wg-jose/draft-ietf-jose-pq-composite-sigs > <https://github.com/ietf-wg-jose/draft-ietf-jose-pq-composite-sigs/tree/main> > . > > > > The previous COSE test vectors were formatted to stay close to the format > used in the pure ML-DSA COSE draft's examples, but I agree your suggestions > are a better fit here. > > > > Thanks again for your feedback! > > Best, > > > > Lucas > > > > *From:* Lijun Liao <[email protected]> > *Sent:* 24 August 2026 05:27 > *To:* Lucas Prabel <[email protected]>; JOSE WG <[email protected]>; > cose <[email protected]> > *Subject:* Re: draft-ietf-jose-pq-composite-sigs: ECDSA encoding > > > > For the examples in COSE (Appendix A.2), take Figure 7 as example: > > > > { > > "priv": "0000...0107", > > "mldsa_seed": "0000...0000", > > "ecdsa_d": "0000...0001", > > "key": “a501...0107", > > "key_diag": "{1: 7, 2: h’08e7...1e3e', 3: -54, -1: h'ba7...51f5', -2: > h'0000...0107'}", > > "sign1": “d284...6863", > > "sign1_diag": "18([h’a204...3835', {}, h'6865...6863'])", > > "raw_to_be_signed": “846a...6573", > > "raw_message_representative": “436f...35f1", > > "raw_signature": “6efa...6863", > > "raw_public_key": “ba71...51f5" > > } > > > > 1. For the top level fields priv, melds_seed, ecdsa_d, key, sign1, > raw_to_be_signed, raw_message_representative, raw_signature, > raw_public_key, the value is hex representation of the raw value, this is > very inefficient and there are no reasons to use it in COSE. Please use > bstr (bytes) instead str (text). For example, "priv": h'0000…0107’. > > > 2. All top level fields are identified by tstr (text), there is > allowed in COSE, but it takes too many spaces. Please considering use > integer instead. > > > > Best > > Lijun > > > > *From: *Lucas Prabel <[email protected]> > *Date: *Friday, 21. August 2026 at 16:48 > *To: *Lijun Liao <[email protected]>; JOSE WG < > [email protected]>; cose <[email protected]> > *Subject: *[COSE] Re: draft-ietf-jose-pq-composite-sigs: ECDSA encoding > > *CAUTION!* *External Email.* Do not click links or open attachments > unless you recognize the sender and know the content is safe. > > Hi Lijun, > > > > Many thanks for your review. I tried to address most of the editorial > points (line wrapping, P-256/P-384 naming consistency, the r/s wording in > 4.5.1, the A0 tag explanation) and you can find the changes in the Editor's > Copy on the Github repository: PQ/T Hybrid Composite Signatures for JOSE > and COSE > <https://ietf-wg-jose.github.io/draft-ietf-jose-pq-composite-sigs/draft-ietf-jose-pq-composite-sigs.html#name-ecdsa-encodings> > . > > > > For Appendix A.2, could you indicate what raised your concern in > particular? > > > > On your suggestion to a normative reference to LAMPS instead: I agree that > it would be a cleaner approach. We'll leave the current text as is for now > and plan that change for a future revision if there are no objections from > the group. > > > > Concerning the poll, it was taken during the JOSE session of the IETF 126 > meeting, but anyone attending (onsite or remotely) could vote. In any case, > we see it only as an indication, which is also why we'd welcome more input > from both the JOSE and COSE mailing lists before settling this. > > > > Thanks again, > > > > Lucas > > > > *From:* Lijun Liao <[email protected]> > *Sent:* 31 July 2026 18:19 > *To:* Lucas Prabel <[email protected]>; JOSE WG < > [email protected]>; cose <[email protected]> > *Subject:* [jose] Re: draft-ietf-jose-pq-composite-sigs: ECDSA encoding > > > > Hi, > > First, > > "A poll was organized, with the following result: > *I prefer aligning with the LAMPS Composite document?* > *Yes:* 21 / *No:* 12 / *No opinion:* 10" > > Was the COSE WG consulted as part of this poll? > > I also have the following editorial comments: > > 1. *Section 4.5.1, Ecdsa-Sig-Value (Signatures)* > > "r and s are the raw, fixed-length, big-endian values already used today > by [RFC7518] and [RFC9053] (32 bytes each for P-256, 48 bytes each for > P-384)." > > Although *r* and *s* are typically fixed-length values for a given curve, > they may also be encoded using fewer bytes if leading zeros are omitted. > > 2. *Section 4.5.1, Ecdsa-Sig-Value (Signatures)* > > Please also specify how many bytes are used to encode and decode the > length field. Otherwise, implementers need prior knowledge of ASN.1 > encoding rules. > > 3. *Section 4.5.1, Ecdsa-Sig-Value (Signatures)* > > Please wrap the long lines in the last code block. Also, after "parameters > field", add the explanatory text: *"A0: explicit tag 0"*. > > 4. Please use either *P-256 / P-384* or *secp256 / secp384* consistently > throughout the document, rather than mixing the two naming conventions. > 5. Please wrap the long lines in * Table 7: JOSE/COSE Composite Label > Values*. > 6. Please wrap the long lines in the examples in *Appendix A*. > 7. I am not sure that the example in *Appendix A.2* is conformant with > the COSE specification. > > Best regards, > > Lijun (from cose WG) > > > > [image: Image removed by sender. Banner] <http://www.nio.io/> > > This email and any files transmitted with it are confidential and intended > solely for the use of the individual or entity to whom they are addressed. > You may NOT use, disclose, copy or disseminate this information. If you > have received this email in error, please notify the sender and destroy all > copies of the original message and all attachments. Please note that any > views or opinions presented in this email are solely those of the author > and do not necessarily represent those of the company. Finally, the > recipient should check this email and any attachments for the presence of > viruses. The company accepts no liability for any damage caused by any > virus transmitted by this email. > > > > [image: Banner] <http://www.nio.io/> > > This email and any files transmitted with it are confidential and intended > solely for the use of the individual or entity to whom they are addressed. > You may NOT use, disclose, copy or disseminate this information. If you > have received this email in error, please notify the sender and destroy all > copies of the original message and all attachments. Please note that any > views or opinions presented in this email are solely those of the author > and do not necessarily represent those of the company. Finally, the > recipient should check this email and any attachments for the presence of > viruses. The company accepts no liability for any damage caused by any > virus transmitted by this email. > > > _______________________________________________ > jose mailing list -- [email protected] > To unsubscribe send an email to [email protected] >
_______________________________________________ COSE mailing list -- [email protected] To unsubscribe send an email to [email protected]
