Re: [Tsv-art] TSV ART review of draft-ietf-tls-sni-encryption

[Christian Huitema <[e-mail]>]

Thanks for the feedback, Bernard. We already fixed in the editor copy
some of the issues that you described based on other feedback received
during last call, but not all. Your comments about section 3.7.1 and
section 5 are interesting. I like your suggestion of moving some of the
text from section 5 to the introduction and believe that it will improve
the document.


Regarding section 3.7.1, you suggest adding a comment on related work,
and perhaps moving some of the discussion of ALPN there. I understand
the rationale, but I am a bit worried about going too far there -- the
current text reflects discussions and feedback on the TLS list. Also, if
we did add a section on related work we would probably need to reference
the deployment of encrypted DNS services, and I am a bit worried about
doing that late in the production cycle. How about a compromise, such as
pointing the issue in the introduction with a forward reference to
section 3.7.1?


-- Christian Huitema



On 9/9/2019 12:48 PM, Bernard Aboba wrote:
> Document: draft-ietf-tls-sni-encryption
> Reviewer: Bernard Aboba
> Review result: Ready with Nits
>
> This document has been reviewed as part of the transport area review
> team's
> ongoing effort to review key IETF documents. These comments were written
> primarily for the transport area directors, but are copied to the
> document's
> authors and WG to allow them to address any issues raised and also to
> the IETF
> discussion list for information.
>
> When done at the time of IETF Last Call, the authors should consider this
> review as part of the last-call comments they receive. Please always CC
> [e-mail] if you reply to or forward this review.
>
> I have not identified any transport related issues.
>
> NITS
>
> Expansion of acronyms on first use:
>
> Abstract: TLS
> Section 1: DNS
> Section 2.1: ISP, QoS, MITM
>
> Section 2
>
> s/mutiple/multiple/
>
> Section 2.1
>
> s/fradulent/fraudulent/
>
> Section 3.6
>
>    The downside is the the
>    client will not verify the identity of the fronting service with
>    risks discussed in , but solutions will have to mitigate this risks.
>
> [BA] Several problems with this sentence:
>
> s/the the/the/
> s/this risks/the risk/
> s/discussed in ,/discussed in [REF-TBD],/
>
> Section 3.7.1
>
> This section seems somewhat out of place in a section on Security
> and Privacy Requirements for SNI Encryption, given that it relates
> to hiding of the ALPN, and the text admits a weak case for linking
> the two problems:
>
>    Using the same technique for hiding the ALPN and encrypting the SNI
>    may result in excess complexity.  It might be preferable to encrypt
>    these independently.
>
> You might consider moving this section to Section 4.3.1, under Section 4.3
> Related Work.
>
> Section 5
>
> The first paragraph of this section strikes me as being potentially better
> suited to inclusion in Section 1 Introduction.
>
>    Replacing clear text SNI transmission by an encrypted variant will
>    improve the privacy and reliability of TLS connections, but the
>    design of proper SNI encryption solutions is difficult.  This
>    document does not present the design of a solution, but provides
>    guidelines for evaluating proposed solutions.
>