Re: Recursive look up of base in outer headers

Al Gilman <> Tue, 02 September 1997 21:14 UTC

Received: from cnri by id aa23400; 2 Sep 97 17:14 EDT
Received: from (services.Bunyip.Com []) by (8.8.5/8.7.3) with ESMTPid RAA25456; Tue, 2 Sep 1997 17:17:26 -0400 (EDT)
Received: (from daemon@localhost) by (8.8.5/8.8.5) id RAA29070 for uri-out; Tue, 2 Sep 1997 17:00:30 -0400 (EDT)
Received: from (mocha.Bunyip.Com []) by (8.8.5/8.8.5) with ESMTP id RAA29064 for <>; Tue, 2 Sep 1997 17:00:27 -0400 (EDT)
Received: (from daemon@localhost) by (8.8.5/8.8.5) id QAA19871 for uri@services; Tue, 2 Sep 1997 16:58:07 -0400 (EDT)
Received: from ( []) by (8.8.5/8.8.5) with ESMTP id QAA19868 for <uri@Bunyip.Com>; Tue, 2 Sep 1997 16:58:04 -0400 (EDT)
Received: (from asgilman@localhost) by (8.8.4/8.8.4) id RAA03064; Tue, 2 Sep 1997 17:00:20 -0400 (EDT)
From: Al Gilman <>
Message-Id: <>
Subject: Re: Recursive look up of base in outer headers
To: Jacob Palme <>
Date: Tue, 2 Sep 1997 17:00:20 -0400 (EDT)
In-Reply-To: <v03110700b031e7c5aae8@[]> from Jacob Palme at "Sep 2, 97 05:58:53 pm"
X-Mailer: ELM [version 2.4ME+ PL15 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

to follow up on what Jacob Palme said:

> The MHTML standard says that if there is both a Content-Base
> and a Content-Location header in the same content heading, the
> Content-Base has precedence over the Content-Location. Is there
> any problem with this? Clearly, the standard must specify which
> has precedence, good standard should not leave things like
> this undefined. "undefined" in my opinion is an ugly word
> in standards documents.

Some more loose talk from this end -- I haven't done the homework

I suspect that HTTP is willing to trust that header fields are
received in the order sent, and give precedence to the textually
last header field within a [single message header, or by
extension the headers of a single MIME part].  Since in the email
context you may not wish to trust that header fields are received
in the order that they were sent, you may have a good reason not
use that as the basis for resolving conflicts.

Personally, I see some virtue to a policy which would make the
presence of both a Content-Base and Content-Location header field
within the same header block an error [leaving BASE undefined] if
they do not agree as to the implied BASE.  But I do see the
choice whether to fix this quietly, fix it with a warning, or not
fix it as a judgement call without an iron-clad case for any one

Particularly, if the same Content-Base + Content-Location value
pair are at risk of being interpreted one way when transmitted
in MIME and another way when transmitted via HTTP I see this as
a likely source of trouble and too arcane for prime time.  

Al Gilman