On Fri, Oct 24, 2025 at 10:52 AM Paul Wouters <[email protected]> wrote:

> On Fri, 24 Oct 2025, Sean Turner wrote:
>
> > Subject: [TLS] Re: KeyShareEntry MUST be generated independently - was
> Re:
> >     Mohamed Boucadair's No Objection on draft-ietf-tls-hybrid-design-15:
> (with
> >      COMMENT) (fwd)
> >
> > Hi! It appears that the emerging consensus here is to make no changes as
> a result of this comment.  Mostly this is because it appears that the MUST
> will be ignored.
> >  If you disagree with this, please indicate why by 30 October 2025.
>
> I think reading it context does help to frame the issue properly for
> evaluation:
>
>
> https://datatracker.ietf.org/doc/html/draft-ietf-tls-hybrid-design-15#section-3.2
>
>
> A note I have though is as the start of that paragraph states:
>
>         [TLS13] requires that ``The key_exchange values for each
> KeyShareEntry
>         MUST be generated independently.'' In the context of this document,
>         since the same algorithm may appear in multiple named groups, this
>         document relaxes the above requirement to allow the same
> key_exchange
>         value for the same algorithm to be reused in multiple KeyShareEntry
>         records sent in within the same ClientHello.
>
> The question is if this "relaxing of a [TLS13] requirement" means this
> draft Updates: [TLS13]. I think it can be argued that it only updates
> the requirement for hybrids, and so one does not need to know this
> "relaxing requirement" in other cases, so an Updates: clause is not
> needed. But if new drafts specifying hybrids come out, it might not
> be obvious to people this is the case unless those drafts mention this
> ssue specifically via a normative reference to this draft. So that is
> an argument to argue in favour of an Updates: clause.
>

> The document currently does not have an Updates: clause for this.
>

We all agree that this is overriding the text in 8446, right?

I don't think "Updates" is really clear enough to know whether we need the
header or not (this is one reason I'm not a fan of the tags in the first
place). With that said, as I observed the other day, RFC 8446-bis is
actually ahead of this draft in the pipeline, so I think if we are going
to continue to allow this practice, we *should* modify 8446-bis accordingly
to permit it.

-Ekr
_______________________________________________
TLS mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to