Re: [Dcrup] [taugh.com-standards] Benjamin Kaduk's No Objection on draft-ietf-dcrup-dkim-crypto-13: (with COMMENT)

[Benjamin Kaduk <[e-mail]>]

On Sat, Jun 16, 2018 at 10:56:39AM -0400, John R. Levine wrote:
> On Sat, 16 Jun 2018, Benjamin Kaduk wrote:
> > There were some remarks in the secdir review that I don't remember seeing a response
> > to yet (though I'm not sure about the "DKIM Hash Algorithms Registry" part) -- it would be
> > good to see a reply to them as well as the updates already made.
> 
> That was about a typo in an earlier version, fixed a while ago.

My ballot remark should be interpreted as "I think there are some
good comments in the secdir review, but I do not understand the part
of the secdir review that talks about the DKIM Hash Algorithms
Registry."  I don't know that an earlier rev of the document is
relevant, since the secdir review
(https://datatracker.ietf.org/doc/review-ietf-dcrup-dkim-crypto-12-secdir-lc-wouters-2018-06-11/)
claims to apply to revision 12, the same revision I initially
reviewed.

Since apparently this is difficult, let me go through the secdir
review item-by-item and comment on it:

% NITS:
% I believe the [FIPS-180-4-2015] reference should be replaced with a
% reference to RFC-6376

Still relevant.  (This is the citation for SHA-256; we generally
prefer IETF references to external references.)

% Remove or indicate the RFC Editor should remove the following text:
% 
%       Discussion Venue:    Discussion about this draft is directed to the
%       [e-mail] [1] mailing list.

Still relevant.

% This sentence doesn't parse easily:
% 
%      This is an additional DKIM signature algorithm added to Section 3.3
%    of [RFC6376] as envisioned in Section 3.3.4 of [RFC6376].
% 
% It should simply say something like "This document adds an
% additional key algorithm type to the DKIM Key Type Registry and a
% new signature type to the DKIM Hash Algorithms Registry"

Still relevant, though AFAICT only the Key Type Registry is
impacted.

% This text reads a little odd:
% 
%    Ed25519 is a widely used cryptographic technique, so the security of
%    DKIM signatures using new signing algorithms should be at least as
%    good as those using old algorithms.
% 
% It seems to suggest that being "widely used" is a guarantee for
% being "at least as good as older stuff". Better would be to just
% point to the Security Considerations of RFC 8032

Addressed in the -13.

% Section 4 and 8 have an introductory lines that says "update as
% follows" followed by a dot instead of a colon. That is a little
% confusing to the reader, as if some text is missing before the dot.

Still relevant (though I guess I would prefer "as described in the
following sections" to using a colon).



More generally, while it's oftentime just as fine to reply to a
directorate review with "thanks for your comments, but this text
will be clear for the document's target audience" as to take the
suggested changes, it's really unclear what conclusion to draw when
the review is met with a curtain of silence.

-Benjamin