Thought WG-MSG would be interested as well.
I put some additional comments to Jacob's at the end of this message.
Kind regards,
JH
-//-
Forwarded message:
To: IESG <iesg(_at_)cnri(_dot_)reston(_dot_)va(_dot_)us>,
ietf-822(_at_)dimacs(_dot_)rutgers(_dot_)edu
Subject: FYI, RFC-822 addresses as part of X.400
Date: Thu, 11 Nov 93 17:27:25 +0100
From: "Erik Huizer (SURFnet BV)" <Erik(_dot_)Huizer(_at_)surfnet(_dot_)nl>
Although the idea (help save X.400 :-) is good, the proposal below is
frightening, and will complicate both X.400 and RFC-1327 (gatewaying to
RFC-822). I have send out queries on how to intercept this and am going to
try to get an "IETF" point of view into this. Even if you do not think that
worthwile the proposal should at least make you laugh. Jacobs analysis below
seems correct.
Erik
------- Forwarded Message
} Date: 11 Nov 93 16:52:39 +0100
} From: Jacob Palme DSV <jpalme(_at_)dsv(_dot_)su(_dot_)se>
} To: iso-messaging-group(_at_)gec-b(_dot_)rl(_dot_)ac(_dot_)uk, mhsnews
<mhsnews(_at_)ics(_dot_)uci(_dot_)edu>
} Subject: Extending X.400 with Internet e-mail addresses
}
} A few days ago I received a proposal put forward by the U.S. National
} ISO Body (=ANSI) to ISO/IEC JTC 1/SC 18. The proposal wants to extend
} the X.400 naming scheme to add a new kind of permitted name in X.400.
}
} The new name is called "Internet E-mail Address Format". Such an address
} can consist of one internet-email-address, coded as a teletex string
} (T.61 string) plus, conditionally, a country name combined with an ADMD,
} or a country name combined with a PRMD, or domain-defined-attributes.
}
} Domain-defined-attributes are only allowed if an ADMD or a PRMD is given.
}
} The proposal is very interesting, but I lack the following:
}
} (a) A definition of the allowed syntax within an Internet E-mail address.
}
} (b) An extension to the X.400 apppendix on how to print OR-addresses on
} business cards to cater for the new name format.
}
} The advantages with this proposal are:
}
} (1) Easier gatewaying between X.400 and Internet
}
} (2) The possibillity of using Internet names also within X.400,
} to get a neater naming scheme
}
} The disadvantages, as I see it, are
}
} (3) The OR-address format becomes even more complex
}
} (4) Interaction problems with older X.400 systems
}
} Question on the proposal is if it is really necessary to extend
} Internet addresses with country code, ADMD, PRMD and Domain-defined
} attributes. By allowing these, even if they are optional, the whole
} idea of giving X.400 the simple name formats of Internet are
} defeated!
}
JH's comments:
I would predict that using the optional ADMD and PRMD attributes would
become the default if MDs start using it to force routing through their
domain. I thus agree with Jacob that this defeats idea of giving X.400
the simple name formats of Internet.
I am not so sure that gatewaying would become easier. Every gateway
has to know whether it is gatewaying to a new or an old X.400 system.
Since the old systems will be widespread for quite some time, a double
standard will have to be built into every gateway.
I wonder if it is really needed. Within the RARE Task Dorce on Mapping
Authorities, we discussed last week that any domain should be allowed
to overrule any RFC 1327 mapping with a DDA mapping (as long as it
uses its 'own' standard attributes). This means that with a rather
simple enhancement, any X.400 UA can allow the user to type in Internet
addresses directly. There are some issues left to be resolved here, but
it gives me the feeling that ISO may be trying to solve a non-problem.