Re: [Cfrg] review of draft-irtf-cfrg-hpke-05

[Ilari Liusvaara <[e-mail]>]

On Mon, Aug 17, 2020 at 01:59:32AM +0000, Manger, James wrote:
> HTTP/2 Version Identification https://tools.ietf.org/html/draft-ietf-httpbis-http2-17#section-3.1
> took a different approach to this issue. It defined the final string
> "h2"; said only implementations of the final, published RFC can use
> that; earlier implementations MUST NOT use that; they can use
> "h2-<draft-number>", eg "h2-17"; a note to the rfc-editor asked for
> this text to be remove in the final RFC.

And on what other things did, Websockets (RFC 6455) had version number
that was the draft number of last incompatible change. And it happened
that the final version number in the RFC (13), was not the same as the
final draft number (17). The preceeding version to that was 8.

ACME (RFC 8555) would have used sequential version numbers for
incompatible method versions, starting from -01, but turns out that no
such changes happened for any published method (one method got version
-02, but was latter scrapped entierely).


-Ilari