[TLS] Re: WG Last Call: draft-ietf-tls-mlkem-05 (Ends 2025-11-26)

[Eric Rescorla <[e-mail]>]

On Sun, Nov 30, 2025 at 5:08 AM Muhammad Usama Sardar <
[e-mail]> wrote:

> Super naive question: how critical is this "application profile standard"
> discussion for MLKEM draft? IMHO, if it is orthogonal enough, maybe we can
> move it over to a separate thread?
>
As best I can tell, not at all.

-Ekr


> On 30.11.25 09:57, John Mattsson wrote:
>
> If you interpret the word “standard” as defined in United Nations
> A-HRC-53-42,
>
> "The term “standard” refers to an agreed norm defining a way of doing
> something in a repeatable manner."
>
> Is this how most people in the IETF interpret this term? I don't think so
> and the argument of mixing and matching definitions from outside applies
> here as well. IMHO, we should attempt for complete definitions of our own,
> rather than letting people import their desired definitions or
> interpretations from outside.
>
> Besides, in my naive understanding (sincere apologies again if I am
> missing/misunderstanding something), D. J. Bernstein's concern seems to be
> on the word "profile", rather than "standard".
>
> In general, I think we agree that there is an ambiguity in "application
> profile standard", and things are left over to interpretation, which is
> leading to misunderstandings.
>
> IMHO, a collaborative way to solve this is to perhaps write a
> clarification document addressing D. J. Bernstein's concerns. I would
> assume that would need some lengthy debates. Perhaps that is too late to
> make RFC8446bis wait for it to resolve? Hence, a proposal for a small new
> draft.
>
> -Usama
>
> PS: While I am mostly in agreement with Ekr and John (except for the above
> UN interpretation), having seen the notice of moderation of D. J.
> Bernstein's posts and to be fair with him, personally I will not post on
> this topic any further until one of his posts land in the mailing list to
> give him equal opportunity to clarify/defend his perspective/position.
>
> PPS: I am not super invested in this topic in the long term. This is
> probably not something I can formally prove to be correct or wrong. It is
> most likely just a matter of definition: one can define it one way or the
> other, leading to very different results. Nevertheless, I do believe it's
> important to resolve for us to peacefully work towards the shared goal of
> securing TLS rather than having to deal with appeals. So if that helps, I
> am volunteering to initiate a draft to try to resolve any misunderstandings
> and at some point in time, shift the draft over to one of you to move it
> forward.
>