Re: [secdir] Secdir review of draft-baker-ietf-core-03.txt

Fred Baker <> Sat, 21 November 2009 21:06 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id B04423A6849; Sat, 21 Nov 2009 13:06:06 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -106.435
X-Spam-Status: No, score=-106.435 tagged_above=-999 required=5 tests=[AWL=0.164, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id NO4v5bLcd-k1; Sat, 21 Nov 2009 13:06:05 -0800 (PST)
Received: from ( []) by (Postfix) with ESMTP id 881CC3A6405; Sat, 21 Nov 2009 13:06:05 -0800 (PST)
Authentication-Results:; dkim=neutral (message not signed) header.i=none
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: ApoEAI/kB0tAaHte/2dsb2JhbAC8A5cahDwEgwGKCw
X-IronPort-AV: E=Sophos;i="4.47,264,1257120000"; d="scan'208";a="52372946"
Received: from ([]) by with ESMTP; 21 Nov 2009 21:06:00 +0000
Received: from [] ( []) by (8.13.8/8.14.3) with ESMTP id nALL5xsb012538; Sat, 21 Nov 2009 21:05:59 GMT
Message-Id: <>
From: Fred Baker <>
To: Phillip Hallam-Baker <>, "Steven M. Bellovin" <>, Charlie Kaufman <>
In-Reply-To: <>
Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Apple Message framework v936)
Date: Sun, 22 Nov 2009 06:05:57 +0900
References: <> <> <>
X-Mailer: Apple Mail (2.936)
Cc: " IESG" <>,
Subject: Re: [secdir] Secdir review of draft-baker-ietf-core-03.txt
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Security Area Directorate <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Sat, 21 Nov 2009 21:06:06 -0000

Hmm. I wonder of one of the directorate members would like to suggest  
better text for the section? I don't claim any expertise on that topic.

On Nov 22, 2009, at 5:42 AM, Phillip Hallam-Baker wrote:

> You can however do CMS without S/MIME.
> I agree that we should regard S/MIME as an extension to MIME. But the
> origin was really to take the RSA developed PKCS#7 technology and
> apply it to MIME. There really isn't much of a link there to the PEM
> work. The IETF attempt to resurrect PEM in MIME was MOSS.
> The problem here is trying to cram everything into a layered model in
> the OSI style. These attempts always fail because the IETF protocols
> are not a layered architecture. It is much more like a web of modules
> that expose interfaces to other co-operating modules. While there is a
> rough conceptual organization according to the degree of abstraction,
> this is not the only organization present and it is nowhere near as
> rigid as in the OSI model.
> The point about S/MIME being protection of data at rest is a critical
> one. We do in fact have a security layer for SMTP - TLS. More mail is
> secured using TLS than any other mail encryption protocol.
> The point being that smart grid is not an opportunity for the IETF to
> roll out the architecture that they would have liked to have developed
> rather than the one that did develop. Some of the differences can be
> explained by path dependence, but quite a few can be explained by the
> market having made the better choice.
> On Sat, Nov 21, 2009 at 12:03 PM, Steven Bellovin  
> <> wrote:
>> On Nov 20, 2009, at 11:41 PM, Charlie Kaufman wrote:
>>> Section 3.1.4: I’d be surprised if S/MIME was originally an  
>>> extension to SMTP. Even when S/MIME was PEM, it was largely  
>>> transport independent (and designed to pass over X.400, which was  
>>> a contender in those days). S/MIME – and more generally CMS – is  
>>> not really a networking protocol at all. It is designed to protect  
>>> data at rest. I can take a CMS protected file and send the pile of  
>>> bits to you by floppy disk or paper tape. Years later, if you can  
>>> still read the media, you can still process it. It’s a tough call  
>>> whether it is an Internet Core Protocol. It’s certainly an  
>>> important IETF protocol.
>> That section is incorrect in several respects, in my opinion.   
>> First, S/MIME doesn't protect "SMTP Mail", in my opinion, since to  
>> me "SMTP" is referring to the 821/2821/5321 parts of the protocol.   
>> S/MIME is more for the 822/2822/5322 parts.  More precisely, S/MIME  
>> is a way to put in security at the MIME layer (references omitted),  
>> which in turn extend 822/2822/5322.  The distinction is important  
>> because HTTP uses MIME but not SMTP.  (It could have used S/MIME,  
>> but that's a separate wistful thread.)  Second, it's an interesting  
>> question if CMS should be mentioned separately -- you can't do S/ 
>> MIME without CMS.
>>                --Steve Bellovin,
>> _______________________________________________
>> secdir mailing list
> -- 
> -- 
> New Website:
> View Quantum of Stupid podcasts, Tuesday and Thursday each week,