Skip to main content

Last Call Review of draft-ietf-quic-version-negotiation-10
review-ietf-quic-version-negotiation-10-dnsdir-lc-spacek-2022-10-11-00

Request Review of draft-ietf-quic-version-negotiation
Requested revision No specific revision (document currently at 14)
Type Last Call Review
Team DNS Directorate (dnsdir)
Deadline 2022-10-11
Requested 2022-09-27
Authors David Schinazi , Eric Rescorla
Draft last updated 2022-10-11
Completed reviews Opsdir Last Call review of -10 by Qin Wu (diff)
Genart Last Call review of -10 by Robert Sparks (diff)
Secdir Last Call review of -10 by Joey Salazar (diff)
Artart Last Call review of -10 by Tim Bray (diff)
Dnsdir Last Call review of -10 by Petr Špaček (diff)
Assignment Reviewer Petr Špaček
State Completed
Review review-ietf-quic-version-negotiation-10-dnsdir-lc-spacek-2022-10-11
Posted at https://mailarchive.ietf.org/arch/msg/dnsdir/swTm6QGRLQqnsVC87asXiedii9s
Reviewed revision 10 (document currently at 14)
Result Ready with Nits
Completed 2022-10-11
review-ietf-quic-version-negotiation-10-dnsdir-lc-spacek-2022-10-11-00
The document contains no direct or indirect reference to the DNS.

I'm unfamiliar with QUIC protocol details and thus not qualified to make
detailed comments on the protocol. On the surface, it looks good.

The document is clearly motivated, and the proposed mechanism's description
contains helpful examples throughout the whole document.

A list of nits follows, very likely as a matter of personal taste - feel free
to ignore:

From my perspective, the document overuses forward references to itself, which
makes it harder to follow.

a) Section 8 Special Handling for QUIC Version 1
It would be nice if this section were mentioned in the beginning. For a while,
I thought the proposed protocol could not work because I did not notice special
handling in the Table of contents. I would put it forwarding, possibly as a
subsection of the Version Information section.

b) The first mention of "transport parameter" could use a reference to RFC 9000
sec. 7.4 to make it easier for the reader.

c) Also, section 5.  Server Deployments of QUIC forward could appear earlier, 
possibly as a subsection of the Version Information section.

For me personally more logical text flow would be:

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   3
   3.  Version Information . . . . . . . . . . . . . . . . . . . . .   8
      8.  Special Handling for QUIC Version 1 . . . . . . . . . . . . .  15
      5.  Server Deployments of QUIC  . . . . . . . . . . . . . . . . .  11
   2.  Version Negotiation Mechanism . . . . . . . . . . . . . . . .   4
   4.  Version Downgrade Prevention  . . . . . . . . . . . . . . . .  10

With enough jumping back and forth, the document makes sense, so again, feel
free to ignore me.