Minutes of the May 7th, 2024 Meeting

[johngray-dev]

Present: Carl Wallace, John Gray, Mike Ounsworth, Corey Bonnell, Daniel Van Geest, Felipe Ventura, Markku-Juhani O. Saarinen, Petr Muzikant, Rajeev Ranjan

Welcomed Petr Muzikant and Rajeev Ranjan

Round Table:

Felipe

• His Composite Signatures implementation got accepted in the OQS provider! Congratulations Felipe!

Markku

• Interested in PKI integration
• SPHINCS+ hardware implementation covered by Intel.
• Markku mentioned some announcement was going to come from the NSA in regards to pre-hashing. This is relation to what NIST announced at the 5th PQC conference a few weeks ago.
• For hardware manufacturers, CNSA guidance is important. CNSA isn't very interested in Hybrid.
• There is no change in the standardization timelines (still summer).
• The Crypto conference accepted Markku's paper on the hardware SPHINCS+ implementation. Congratulations Markku!
• Dilithium is already supported from official test system - https://github.com/usnistgov/ACVP-Server/blob/master/gen-val/src/crypto/src/NIST.CVP.ACVTS.Libraries.Crypto/Dilithium/Dilithium.cs
• Action John: Submit an issue, and maybe someone will generate artifacts using this code base for the next hackathon as it is the standard for FIPS 140 testing. See Add NIST artifacts from the FIPS 140-3 test system #114

Carl Wallace

• Verifier for composite working, generator being worked on.
• We should clean up the old OIDS and algorithms forward. Should align with some set of OIDs and we don't need to support the old stuff.
• Cleaner set of OIDs for a .zip a client. DER verses PEM - (For V4 format) -
• Action: John to submit an issue to take care of these in a V4 format in the future (likely after first algorithms standardized). See Desired properties of V4 format artifact file #112

Petr Muzikant

• Agreed with cleaning the OID table. Would like to see the OID table get used more widely. Suggested making it more accessible by creating a portable JSON format to make it machine readable.
• Action: Petr to take a look at how to make the OID table more exportable (no timeline committed). See How to make the OID table more exportable #113

Ranjeet, Ranjan

• Working with Alexander on the interoperable CMP test suite

Corey Bonnell

• Don't have much to report, hope to get back to PQ.

Daniel Van Geest

• Congratulations for composite signature adoption
• composite signature artifacts on the way. A subset of the OIDs for now.

Mike Ounsworth:

• Composite KEM will likely get standardized soon after kyber, we should make an effort to include some artifacts for composite KEM.
• Action all: Work on Composite KEM if possible.

PUBLIC KEY, PRIVATE KEY FORMAT Disucssion:

• Daniel Van Geest Led discussion on public key encodings and OCTET STRINGS
  • No extra layer of ASN.1 encoding (From IETF 115 hackathon - see : https://github.com/IETF-Hackathon/ietf115-project-presentations/blob/main/IETF_115_Hackathon-PQKeysAndSignatures.pdf
  • See https://datatracker.ietf.org/doc/html/rfc5480#section-2.2 RFC 5480: Elliptic Curve Cryptography Subject Public Key
  • Daniel reaching out to the authors to help clarify so extra 4 bytes are not encoded. It is the OCTET STRING value that is encoded.
  • XMSS, LMS certificate encoding - has an extra wrapping in their certs (already published). Maybe there is something that can be done about this?
• Private Key Format - Change to using OneAsymmetricKey and use the optional public key field to carry the public key. This seems to be the preferred option. There was some discussion of using a private seed (as suggested by Diedre Connolly and Sophie) but that doesn't seem to be in the NIST documents. So carrying the public key in the private key as per OneAsymmetricKey would make the public key easily available (with the downside of larger keys and possible interop issues since it is optional).

Next meeting is Tuesday June 4th