Hi Martin, would you mind working out a PR? I think being able to compare 407 to a concrete alternative would be helpful. Just so that we're on the same page, here's a quick summary of the issues that 407 is designed to solve. (These may or may not be problems in your view, and I don't claim this list is exhaustive.) - 233: No acceptance signal until after HRR, so the procedure for computing CH2 is underspecified. This can be avoided by advertising the same preferences in CHI/CHO, but the spec doesn't require this. - 373: To fix 3233, can we put an acceptance signal in HRR.random? Probably not, since HRR.random has a value specified in RFC8446. - 358: RFC8446 allows the value of an extension in CH2 to differ from CH1 only if the extension appears in HRR. - 333: "Split mode" is broken, since the client doesn't know who the cookie is for.
Best, Chris P.
_______________________________________________ TLS mailing list TLS@ietf.org https://www.ietf.org/mailman/listinfo/tls
