On Jun 1, 2023, at 12:59 PM, Christopher Wood 
<[email protected]<mailto:[email protected]>> wrote:


On Jun 1, 2023, at 4:52 AM, Ilari Liusvaara 
<[email protected]<mailto:[email protected]>> wrote:


With the HPKE registry as of 2023-05-18, I count 14 already. 12 of
those have equivalents composed of existing components.

And that is vulnerable to combinatorial blowup. E.g., adding CP-*
and AEGIS increases that to 33.

The two present ones with no equivalents are:

- HPKEv1-Base-X25519Kyber768Draft00-HKDFSHA256-AES256GCM
- HPKEv1-Base-X25519Kyber768Draft00-HKDFSHA256-ChaCha20Poly1305

(Both are post-quantum.)

If the point here is that the number of entries will grow combinatorially, then 
yes, I agree, but that assumes we’ll be adding lots of things to the registry. 
But as I suggested, I don’t think we should be doing that. I think we should 
just pick some sensible, widely supported defaults. Applications and 
implementations don’t seem to need the flexibility afforded by the alternate 
proposals (as I understand them).

+1

I think we make it easier for COSE (and HPKE) users to do better by providing a 
few good clear prominent choices. We can provide good expert choices.

COSE allows for proprietary alg IDs, so no one is prevented from doing 
different if they feel the need.

Carsten’s proposal seems workable too.

LL

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

Reply via email to