Re: [apps-discuss] AppsDir review of draft-dbider-sha2-mac-for-ssh-05

[Jeffrey Hutzelman <[e-mail]>]

On Tue, 2012-04-10 at 17:50 -0700, Mark D. Baushke wrote:
> > In the Abstract Section:
> > 
> >    "It also updates RFC4253 by specifying a new RECOMMENDED data
> >     integrity algorithm."
> > 
> > Should the word "RECOMMENDED" be interpreted as a RFC 2119 key word?
> 
> Yes, the word "RECOMMENDED" given in both the Abstract and in section
> "2. Data Integrity Algorithms" is a RFC 2119 key word as is specified in
> section "1.1. Requirements Terminology" of document
> draft-dbider-sha2-mac-for-ssh-05.

Technically, the abstract itself does not impose a normative requirement
in the language of RFC2119.  However, the quoted sentence describes
section 2, which does use that language; for clarity, the abstract also
uses uppercase "RECOMMENDED".  IMHO, changing this would serve only to
make the abstract less clear and/or less concise.

> > In Section 3:
> > 
> >   "IANA is requested to update the SSH algorithm registry with the
> >    following entries."
> > 
> > Shouldn't that be the Secure Shell MAC Algorithm Names registry?
> 
> Yes. I was uncertain how to properly address the registry.

No.  It should be the "Secure Shell Protocol Parameters" registry, of
which the MAC Algorithm Names table is one part.  So, if you want to
replace "SSH algorithm registry" with "Secure Shell Protocol
Parameters", that would be a reasonable change, but don't spin a new
internet-draft at this time just for that change.

Aside from that, the terminology used in section 3 is consistent with
that used on the registry web page.  Unless IANA indicates that
something is unclear to them, I would not make any other changes.

-- Jeff