ietf-822
[Top] [All Lists]

Internet EDI

1992-12-07 16:05:47


We have been following the list for some time now with regards to
the integration of MIME/PEM. As an EDI Software provider we have great
interest in this area for it's potential use in the EDI mission.

As a veteran of numerous religious wars (ANSI X12 - UN/EDIFACT )
I would like to suggest that now that the Internet community has taken
a giant step towards becoming a player in the EDI world, we should
reflect on our history and not repeat the errors of our contentious
past.

To try to put this in perspective I should point out that in the event
that the Internet should become a highly successful transport of EDI
documents, a large number of other networks and transports will
undoubtedly coexist. Among them are ISO X.435, IBM SNA networks,
EDI VANs using 2780/3780 Bisynch Protocol and believe it or not,
tape (I forget who it was that said "never underestimate the bandwidth
of 57 Chevy stationwagon.)

EDI users tend to have a very pragmatic view of transport issues. The
EDI traffic will flow to the environment which can demonstrate the
appropriate balance of functionality, cost, and security.

In all of this we must accept the fact that even today the EDI
environment is a "mixed network" game. Here at Premenos we spend a
good bit of our time trying to figure out how to provide
a reasonable level of interoperability between the 30 different
EDI Comm modules we provide with our EDI products. I should point
out that in a recent survey of our 2400 customers they average just
under 3 different comm modules per customer.

We have put together a team to begin working on "Network 31", the
Internet/MIME module. In reading the MIME/822/PEM lists it is clear
to me that we are at a very crucial moment in the standards process.
The MIME/PEM draft has provided us the foundation for doing EDI on the
Internet, but much work needs to be done to address several key areas
such as authentication, delivery notification, non-repudiation,
directory services, EDI message routing, and gateways.

Many of these issues have already been addressed in (dare I say)
ISO X.435. While this like many other OSI efforts has proceeded
unencumbered by implementations, there has been much work between the
EDI community and ISO to hammer out what details do exist.

I know of at least two sites that are actually trying to do 435
internally while waiting for the future deployment of publicly
available 88 ADMDs.

I believe that it is absolutely imperative that whatever we come up
with to do Internet EDI, it must have a very high level of
interoperability with X.435.

To this end, I recommend that before we begin randomly assigning
MIME content types for the purpose of doing EDI on the Internet,
we spend a brief amount of time to seek some common ground with the
ISO folks to insure that we can achieve a level of interoperability
that we all know would be good for both communities. I have today
contacted Richard Hill in Switzerland. He is the UN/EDIFACT Rapporteur
for Western Europe and was instrumental in the X.435 effort.
Klaus-Dieter Naujok, the UN/EDIFACT-ISO Rapporteur Liaison is
fortunately a member of our staff and I have asked him to join
us in this effort.

We would welcome the help of any other interested parties in trying
to put together a recommendation for doing EDI on the Internet.


Lew Jenkins
Premenos Corp

lew(_at_)premenos(_dot_)sf(_dot_)ca(_dot_)us
+1 510.602.2000
+1 510.602.2133  FAX


<Prev in Thread] Current Thread [Next in Thread>
  • Internet EDI, Lew Jenkins <=