Subject: Re: Status of 821bis & Munich agenda
From: Matti Aarnio <mea@nic.funet.fi>
To: Chris.Newman@innosoft.com (Chris Newman)
Date: 	Sat, 2 Aug 1997 03:08:15 +0300 (EET DST)
Cc: drums@cs.utk.edu
In-Reply-To: <Pine.SOL.3.95.970801090014.9839A-100000@eleanor.innosoft.com> from "Chris Newman" at Aug 1, 97 09:22:31 am
MIME-Version: 1.0
Content-Type: 	text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Message-Id: <19970802000823Z204-29056+60@nic.funet.fi>

...
> I think you misunderstood.  I'm a firm supporter of IPv6 and expect DRUMS
> to specify a format for IPv6 literals, but RFC 1924 simply isn't a
> candidate.  The reason why RFC 1924 isn't a candidate can be discerned by
> a common feature it shares with RFC 1437.

	Aargh....   Right.  At first view 1924 does look like it is defining
	the form of compressing embedded zero sequence, unless one reads it
	all the way thru.  Had R.Elz decided to write about using base64
	coding, it could be very believable despite of what it is intended
	to be...

	So to reiterate with correct RFC references, and in ABFN;
	I use following form with ZMailer:
		"[ipv6." <rfc-1884-paragraph-2.2-address> "]"

	Reading smtpupd-05 I see a form which I do consider difficult.
	Namely the use of a space as tag/value separator will be most
	cumbersome with the current breed of email clients, which
	often consider space as separator between addresses.

	On the other hand, doing some testing with suggested form(*)
	with proper double quotes to help, does show to me that it
	is encodable, while not easy for average email user.  At the
	same time, I don't think address literals need to be easy for
	non-experts, anyway.  They are not intended for general use,
	after all.

	(*) the form I used was like ABFN:
		"[IPv6" SP <rfc-1884-paragraph-2.2-address> "]"
	and what I actually wrote at the shell command line was:
		elm -V '"mea@[IPv6 ::1]"'
	Without the double-quotes elm did report target address as:
		mea@[IPv6, ::1]
	that is, TWO addresses...

	I am waiting for the smtpupd-06 to emerge from the ID editors
	to be able to comment further.

> 		- Chris

/Matti Aarnio <mea@nic.funet.fi>