On Fri, Oct 1, 2021 at 8:04 PM Sam Whited <[email protected]> wrote:

> I have to respectfully disagree with this.
>
> Anecdotally, RFCs are hard to discover. Having them linked from a
> logical place in other RFCs is one way that discovery happens, and if
> you're looking for how to do channel bindings with TLS the first place
> you're going to look is the TLS RFC (and its list of updates).
>
> Secondly, this is an update, not a retconn. It in no way implies that
> TLS 1.3 always said this, or that the TLS 1.3 authors were involved in
> the channel bindings spec. TLS 1.3 does an analysis of its own keying
> material exporters and we rely on this and present a standard name for
> one scenario where it may be used, this does not involve new technology
> or even a novel use of EKM.
>

Well, perhaps it would be useful to step back for a moment from the
implications of "updates". We are currently preparing an 8446-bis.
What would you have that document say about this topic?

-Ekr


> —Sam
>
>
> On Fri, Oct 1, 2021, at 18:49, Eric Rescorla wrote:
> > I don't believe that this document should update 8446. As noted in S
> > 1, we didn't define these bindings because we didn't have complete
> > analysis. This document doesn't seem to either contain or reference
> > such analysis and until we have that, I think RFC 8446 shouldn't be
> > retconned into endorsing this construction.
> >
> > -Ekr
>
> _______________________________________________
> TLS mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/tls
>
_______________________________________________
TLS mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/tls

Reply via email to