I think there is no need to repeat this, but I fully agree with Ilari on
this matter.

As a final note, the contents of the registry do matter. I consider HPKE to
> be a fairly low-level primitive, whereas I (perhaps naively) view COSE as a
> higher-level application protocol. We have the expertise to be opinionated
> about what goes into that registry for applications to use, and we should
> exercise that opinion to provide folks with a very limited set of options,
> ideally target precisely one. The
> "HPKEv1-Base-DHKEM(P256,HKDFSHA256)-HKDFSHA256-AES128GCM" label seems like
> a perfectly fine candidate for that.


I can't agree with that.
COSE-HPKE might be a high-level application layer standard, but it's not
specific to a particular application. Whether or not we can limit the
cipher suites should be decided based on whether the specification is
premised on a limited use case or environment, and whether it's specific to
a certain application. The COSE-HPKE spec should be a generic standard that
doesn't assume a specific use case, and limiting the cipher suites is not
the right choice.

I wouldn't object to limiting the cipher suites in the ”Firmware Encryption
with SUIT Manifests" specification that utilizes COSE-HPKE. This is because
the use case is limited and it is possible to restrict ciphersuites based
on reasonable grounds.

I suggest requesting registrations only for the algorithm suites that
> people want to use in COSE-HPKE


OK. Let's register 72(6*3*4) cryptographic suites then. In the future, it's
difficult to determine which cryptographic algorithms may become insecure,
whether due to specification or implementation. If we are considering
removing some algorithms, I would like to request a reasonable
justification. Can such a thing be done in this specification, which does
not assume any specific use cases?

Best,
AJITOMI Daisuke

2023年6月1日(木) 18:52 Carsten Bormann <[email protected]>:

> On 2023-06-01, at 11:07, John Mattsson <john.mattsson=
> [email protected]> wrote:
> >
> > I don’t think COSE in the past had something that can be described as
> cipher suites.
>
> To me anything that has a “w/” or a “+” (mostly) in
> https://www.iana.org/assignments/cose/cose.xhtml#algorithms
> is a cipher suite.
> Sorry if that term is not rigorously defined...
>
> Grüße, Carsten
>
> _______________________________________________
> COSE mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/cose
>
_______________________________________________
COSE mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/cose

Reply via email to