IETF Logo Mail Archive

Re: suggested revision for MUST/SHOULD

"D. J. Bernstein" <djb@cr.yp.to> Thu, 27 July 2000 22:53 UTC

Received: from cs.utk.edu (CS.UTK.EDU [128.169.94.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA28274 for <drums-archive@odin.ietf.org>; Thu, 27 Jul 2000 18:53:47 -0400 (EDT)
Received: from localhost (daemon@localhost) by cs.utk.edu with SMTP (cf v2.9s-UTK) id SAA19626; Thu, 27 Jul 2000 18:53:29 -0400 (EDT)
Received: by cs.utk.edu (bulk_mailer v1.13); Thu, 27 Jul 2000 18:53:27 -0400
Received: by cs.utk.edu (cf v2.9s-UTK) id SAA19609; Thu, 27 Jul 2000 18:53:27 -0400 (EDT)
Received: from muncher.math.uic.edu (marvin@localhost) by cs.utk.edu with SMTP (cf v2.9s-UTK) id SAA19596; Thu, 27 Jul 2000 18:53:26 -0400 (EDT)
Received: from muncher.math.uic.edu (131.193.178.181 -> koobera.math.uic.edu) by cs.utk.edu (smtpshim v1.0); Thu, 27 Jul 2000 18:53:26 -0400
Received: (qmail 14524 invoked by uid 1001); 27 Jul 2000 22:53:47 -0000
Date: Thu, 27 Jul 2000 22:53:47 -0000
Message-ID: <20000727225347.4260.qmail@cr.yp.to>
From: "D. J. Bernstein" <djb@cr.yp.to>
To: drums@cs.utk.edu
Subject: Re: suggested revision for MUST/SHOULD
References: <200007271515.LAA17836@astro.cs.utk.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
List-Unsubscribe: <mailto:drums-request@cs.utk.edu?Subject=unsubscribe>

I agree that this needs discussion.

However, I have a different suggestion: the definitions of ``SHOULD'' et
al. should simply be copied from RFC 2119, with no extra text.

(I say ``copied'' rather than ``incorporated by reference'' to avoid the
issue of whether smtpupd is obliged to follow RFC 2119, section 6.)

Most readers will be expecting the RFC 2119 definitions. Many readers
will misinterpret the document if it uses anything else---they'll never
realize that there are new definitions.

I've checked a bunch of ``SHOULD''s in the document, and they all seem
to predate the new definitions, which appeared in smtpupd-09. Presumably
they were written with the RFC 2119 (or almost identical RFC 1123)
definitions in mind. The new definitions seem to have screwed up the
meaning of these ``SHOULD''s.

---Dan

v2.23.0 | Report a Bug | By Email