Path: utzoo!utgpu!watmath!att!rutgers!ucsd!helios.ee.lbl.gov!ncis.tis.llnl.gov!MIRSA.INRIA.FR!Christian.Huitema
From: Christian.Huitema@MIRSA.INRIA.FR (Christian Huitema)
Newsgroups: comp.protocols.iso.x400
Subject: Re: concise format
Message-ID: <8910230853.AA14326(a)jerry.inria.fr>
Date: 23 Oct 89 09:17:00 GMT
References: <457293(a)QZCOM>
Sender: root@ncis.tis.llnl.gov
Distribution: inet
Organization: The Internet
Lines: 49
Approved: post-x400@tis.llnl.gov

Jacob,

I was proposing table free RFC-987 like addresses, which would yeld something like:
   Internet:	christian.huitema@mirsa.inria.fr
   X.400:	christian.huitema@mirsa.inria.atlas.FR
on a business card. This should have been obvious from examples like:
	grimm@darmstadt..gmd.dbp.DE
	eppenberger@verw.switch.switch.arcom.CH
	steve.kille@cs.ucl.uk\.ac.gold_400.GB
which I gave in the message. It should be fairly clear to anyone that
the addresses should be noted in a way that does not involve tables..

However, this was merely a counter-example to yours, showing ``another
short-hand format'', which would appear much more natural in our
environment than your original proposal. Apart from depending a lot
from different local tradition, all these ``concise formats'' suffers
the same draw backs, i.e.:

1. Requiring a positional syntax implies that one selects a ``standard
address form'' (given.name@ou.o.prmd.admd.c in the example), and that
one resorts to special tricks like double separators or blank fields
for ``deviant addresses'', e.g. when some field is ``missing''. This is
not user friendly.

2. The most used form includes a Personal name and some organisational
designation. As the personnal name can include several tokens, and the
organisational designation a variable number of OUs, one need a special
construct (@ and . in RFC-987 style) for the delimitation. This can be
difficult to explain if the construct is not very clear; the "/" and
"//" of your example is in my opinion difficult to explain to users.

3. The less used forms include rare fields, like X.121 addresses, A ids
or DD attributes. There is no way to incorporate them into a positional
syntax, and one must resort to a key-word syntax.

4. Positional syntaxes, by nature, are less user friendly than keyword
syntaxes. For example, users should be free to enter the addresses in
both little endians and big endians fashion, depending of their mood.

5. The claimed conciseness is not such a big winner. For example, the
difference between:
	<grimm@darmstadt..gmd.dbp.DE> and
	<S=grimm;OU=darmstadt;PRMD=gmd;ADMD=dbp;C=DE>
is only 15 chars. The price is not so large, if it results in less
confusion. Actually, the only relevant criticsm which was expressed in
this discussion is that the keywords used in the RARE syntax are too
terse, and that "Surname" would be preferable to "S", etc.

Chrsitian Huitema