-----------------------------
In reply to: 61931112091716/0003765414NA1EM
Application message id: 0100044086104841
Importance: Normal
Grade of Delivery: Normal
-----------------------------
I wholeheartily greet ANSI's initiative to enhance X.400 with
another address form: "Internet E-mail Address Format". this could make
it superflouus to encode or otherwise translate addresses between RFC-
822 and X.400 -- X.400 can *incorporate* RFC-822 instead of trying to
replace it.
On the questions raised, --
} (a) A definition of the allowed syntax within an Internet E-mail
address. <
I certainly do not know, but seeing that ANSI proposed to encode it
as an T.61 string, I would not think that there are any restrictions.
} (b) An extension to the X.400 apppendix on how to print OR-
addresses on business cards to cater for the new name format. <
I think this is simple: you just write the "Internet E-Mail Address
Format" as you are used to write an RFC-address, e.g. like this:
jdoe(_at_)host(_dot_)domain(_dot_)top
Does this proposal make gatewaying between X.400 and Internet easier?
Well, the question is: is gatewaying as such then still necessary?
X.400 can assume the role many of its architects hat in mind: to be a
superstructure for all proprietary electronic mail schemes. Internet
addresses can simply be transported in an X.400 message; the
translation of the body part can become a normal routine of the MTS,
just as the translation between T.61 (TTX) body parts and IA5-
Bodyparts. Transition from RFC-822 to X.400 can be done without any
pain, since one could continue to use the same address.
Complementing the RFC-822-Address with a domain id in an Internet E-
Mail Address Form of an X.400 O/R address is the same scheme as with
the X.121 Address. In some cases it might be better to add an X.400
domain Id for more economical routing. Can you conclude from an address
like: meier(_at_)city(_dot_)com or something(_at_)ibmmail(_dot_)com in which
country the host
for this mailbox is located? It can be anywhere in the world, and
actually it is not in the USA, but in Germany resp. Britain. The
unilateral decision of the "mapping authorities" to assign a domain id
of US;;INTERNET to addresses which do not end with an IS3166 country
code or with "UK" does not make sense.
The good idea of the original X.400 architects, to use the normal
personal name as the address, can become a reality only with X.500
based directories. The original way as the mnemonic form has been
perverted by many implementations which put routing information in the
mnemonic address and use organization or org.units to represent host
ids, or put the computer user id into the surname or given name portion
of the personal name. Also, the mailbox id has to be a worldwide unique
key, and a human personal name is not and cannot be uniqe; it is also
not fair to force people to adopt additional name attributes to make
their names unique. It would have been better to allow the numeric user-
id from the numeric address form (in X.400-1988 terms) to be also a
string user-id -- so the existing public e-mail providers like ATTMAIL,
TELEMAIL, MCi, etc would not had to swerve to DDAs to carry their
necessary user-id address information. X.500 directories which allow
duplicate names, free-form names in the address and a simple string-
based user-id are what I am looking for. Maybe one could generalize the
ANSI proposal to do this.
Reading of "mapping authorities" -- this makes me think that this
Internet-X.400 gatewaying generated a more centralised (authoritarian?)
structure than necessary for e-mail. X.400 allows a more decentralised,
cooperative structure, and the ANSI proposal would make these
authorities superflouus -- how nice, one more centralized
administration gone.
I am looking forward to a real convergence of X.400 and RFC-based e-
mail.
Regards,
L.Willms