Re: [apps-discuss] I-D Action: draft-levine-trace-header-registry-01.txt

John Leslie <> Mon, 23 January 2012 13:19 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 37FE821F8745 for <>; Mon, 23 Jan 2012 05:19:54 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -106.298
X-Spam-Status: No, score=-106.298 tagged_above=-999 required=5 tests=[AWL=0.301, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id LgelINNhYR37 for <>; Mon, 23 Jan 2012 05:19:53 -0800 (PST)
Received: from ( []) by (Postfix) with ESMTP id 8F7D121F8742 for <>; Mon, 23 Jan 2012 05:19:53 -0800 (PST)
Received: by (Postfix, from userid 104) id 4DD0333C24; Mon, 23 Jan 2012 08:19:53 -0500 (EST)
Date: Mon, 23 Jan 2012 08:19:53 -0500
From: John Leslie <>
To: John Levine <>
Message-ID: <20120123131953.GA36092@verdi>
References: <20120122220229.87477.qmail@joyce.lan>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20120122220229.87477.qmail@joyce.lan>
User-Agent: Mutt/1.4.1i
Subject: Re: [apps-discuss] I-D Action: draft-levine-trace-header-registry-01.txt
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: General discussion of application-layer protocols <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Mon, 23 Jan 2012 13:19:54 -0000

John Levine <> wrote:
> As promised, I smooshed S.M.'s draft and mine together, so here's one that has
> both his narrative text and my registry additions.

   Nice work!

   But I stumbled over the difference (if any) between RFC2119 SHOULD and
lower-case "should".

   I expect at least one IESG member would also stumble. I advise clarifying
if possible.

   DKIM-Signature's "should" is quoting RFC6376, where it is a "SHOULD",
unless I misunderstand.

   For Authentication-Results, RFC5451 does indeed use lower-case "should"
for "be treated as a Trace field," but it uses upper-case "SHOULD" for
"be added at the top of the message" -- I think folks tend to think of
both of these as RFC2119 "SHOULD".

   For VBR-Info, RFC5518 uses "SHOULD".

   For Auto-Submitted, RFC5436 AFAICT uses "MUST", not "SHOULD" or "should".

   My take on this is that we'd do best to replace those five "should" with

   The remaining "should" in Section 4, I'd simply remove.

John Leslie <>