Re: [ietf-smtp] Should we update an RFC if people refuse to implement parts of it ?

Viktor Dukhovni <> Mon, 07 June 2021 06:44 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id C80343A394F for <>; Sun, 6 Jun 2021 23:44:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.898
X-Spam-Status: No, score=-1.898 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id ZjgMOXV9ggnl for <>; Sun, 6 Jun 2021 23:44:30 -0700 (PDT)
Received: from ( []) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id CCAE83A394D for <>; Sun, 6 Jun 2021 23:44:30 -0700 (PDT)
Received: by (Postfix, from userid 1001) id 92FA8BF541; Mon, 7 Jun 2021 02:44:29 -0400 (EDT)
Date: Mon, 7 Jun 2021 02:44:29 -0400
From: Viktor Dukhovni <>
Message-ID: <>
References: <E863AF380A69FAD9C7CFBEB2@PSB> <20210606185656.06775CC9D0E@ary.qy>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210606185656.06775CC9D0E@ary.qy>
Archived-At: <>
Subject: Re: [ietf-smtp] Should we update an RFC if people refuse to implement parts of it ?
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: "Discussion of issues related to Simple Mail Transfer Protocol \(SMTP\) \[RFC 821, RFC 2821, RFC 5321\]" <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Mon, 07 Jun 2021 06:44:36 -0000

On Sun, Jun 06, 2021 at 02:56:55PM -0400, John Levine wrote:

> I still don't think it's worth opening up RFC 6531 unless we have a
> lot more to change.

I can continue to ignore the suggested use of U-labels in trace headers
without a document revision, and without any consequente
interoperability downsides.  So there's not much impetus for change

If we're really looking for surgery patient, that'd be RFC6532, in which
message/global is a major design blunder.  It blatantly breaks a key
feature of the design of MIME, in which for good reason composite parts
did not get non-identity transfer encodings.

It is perhaps still not too late (given low adoption) to fix the
blunder, by deprecating message/global.  Bounces of EAI messages can
either optimistically include UTF-8 characters in the headers of nested
messages, or conservatively bounce just a "text/rfc822-headers" portion,
which being non-composite can be transfer-encoded:

bouncing just the headers is a good idea anyway, since it reduces
the odds of the bounces getting lost in transit for being overly
large, or triggering false positives in filters, ...