Received: from localhost (daemon@localhost) 
        by CS.UTK.EDU with SMTP (cf v2.9s-UTK)
	id WAA21821; Mon, 16 Mar 1998 22:44:47 -0500 (EST)
Received: by cs.cs.utk.edu (bulk_mailer v1.9); Mon, 16 Mar 1998 22:42:36 -0500
Received:  
        by CS.UTK.EDU (cf v2.9s-UTK)
	id WAA21726; Mon, 16 Mar 1998 22:42:35 -0500 (EST)
Received: from snark.thyrsus.com (esr@[192.190.237.102]) 
        by CS.UTK.EDU with ESMTP (cf v2.9s-UTK)
	id WAA21711; Mon, 16 Mar 1998 22:42:28 -0500 (EST)
Received: (from esr@localhost)
	by snark.thyrsus.com (8.8.7/8.8.7) id WAA29476;
	Mon, 16 Mar 1998 22:47:16 -0500
Message-ID: <19980316224715.22711@snark.thyrsus.com>
Date: Mon, 16 Mar 1998 22:47:15 -0500
From: "Eric S. Raymond" <esr@snark.thyrsus.com>
To: Dave Crocker <dcrocker@brandenburg.com>
Cc: DRUMS WG <drums@cs.utk.edu>
Subject: Re: X- fields
References: <199803170013.SAA20527@deptvachi-bh.va.gov> <199803170013.SAA20527@deptvachi-bh.va.gov> <19980316200526.41665@snark.thyrsus.com> <199803170155.RAA24802@baygate.bayarea.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.85e
In-Reply-To: <199803170155.RAA24802@baygate.bayarea.net>; from Dave Crocker on Mon, Mar 16, 1998 at 05:52:36PM -0800
Organization: Eric Conspiracy Secret Labs
X-Eric-Conspiracy: There is no conspiracy

Dave Crocker <dcrocker@brandenburg.com>:
> What we have learned from 20 years of X- use is that as a space for
> experimental header names (that is, a header for which there will later be
> a different, standardized name) it's a massive failure.  
> 
> This suggests making explicit that the X- name space must be used only for
> private header conventions for which there is no intention or expectation
> of later, public standardization.

That (and Paul Hoffman's supporting remarks) is good enough for me.  I withdraw
my previous proposal and suggest the following language:

...............................................................................

3.6.9 Private extension fields

This standard allows for private extension fields.  Such fields are
distinguished by field-names beginning with the strings "X-" or "x-".

X- fields SHOULD NOT be used for experimental fields intended to be
standardized at such future time; practice has shown that such X-
names become fossilized and difficult to change.

Good practice for X- fields is similar to that for trace and
informational fields.  A conforming programs SHOULD accept X- fields
in a message header without raising an error condition, and SHOULD
NOT delete or modify X- fields other than those of which it has
explicit private knowledge.

Interpretation and semantics of X- fields are outside the scope 
of this document.

...............................................................................

Is this acceptable to all parties?
-- 
		<a href="http://sagan.earthspace.net/~esr">Eric S. Raymond</a>

"Government is not reason, it is not eloquence, it is force; like fire, a
troublesome servant and a fearful master. Never for a moment should it be left
to irresponsible action."
	-- George Washington, in a speech of January 7, 1790