=======================================================================

The Future of EDI on the Internet


(copyright) Ravi Kalakota & Andrew B. Whinston
=======================================================================

EDI has come a long way from its origins in the 1960s bisynchronous 
communications protocols of the IBM mainframe environments. In the 
70s, EDI evolved into store and forward networking (also called the poor 
man's Internet) that became the predominant model used by Value Added 
Network (VAN) service providers even today.   

But, this form of VAN based  EDI is rapidly being considered by many 
leading experts a dinosaur as we move to the fast paced world of mass 
customization and quick response operations.  Experts see the need for 
customized transaction systems, allowing people who do not know each 
other to transact business without prior bilateral agreement. Forging 
EDI agreements today often requires several "painful" years of 
standardization usually by a committee.  

Several groups are proposing new forms of EDI to deal with the turbulent 
world. The reason for this being that with the business environment 
changing rapidly, the existing EDI standards process is both wasteful and 
useless for business operations, as  competing companies in several 
industries operate in specialized business arrangements which last a couple 
of months or few transactions.  These new forms of EDI are being called 
"open-EDI" or "new-EDI" .  

What is the major difference? Today, in the vast majority of EDI 
transactions, we have negotiated Trading Partner Agreements (TPAs) 
specifying data interchange on a one-to-one basis.  On the Internet and 
electronic commerce arena, there is an emerging trend of what can be 
termed Unilateral TPAs - where one party promulgates the standard to be 
used for a transaction, and invites anyone willing to do business with that 
firm to submit transactions of that type. This is analogous to the paper 
"Order Form" and can be seen in the on-line World Wide Web (WWW) 
"forms interface" expected to be the basis for "electronic commerce" on the Internet. But, note that UTPAs are not fly-by-night forms but do 
require some thinking about syntax and semantics in the context of legal, 
accounting and auditing considerations.

The new "forms" based flexible WWW-EDI is the future. Why? First 
because, these forms use a structured document language (e.g. SGML -- 
Standard Generalized Markup Language) that communicates clearly what 
the intent of the transaction as clearly as the rigid EDI codes. The use of a 
structured document languages provide a great deal of flexibility for post 
processing of forms in various down-stream business processes.  This 
promises to eliminate the problems posed by the competing, vertical 
industry aligned, segregated VAN's that impede effective business process 
creation and management due to their inflexibility.

On the much talked about issue of Internet security. New cryptographic 
methods being integrated into various applications  promise to provide 
transparent security features that operate under-the-hood and let the user 
concentrate on the task rather than the nitty-gritty details of 
encryption/decryption and key management. Examples of these new EDI 
implementations can be seen in  Premenos Inc.  Templar, which enables 
companies to make secure EDI transactions over the Internet rather than 
through their VAN, which level per-character or per-message fees. 
Templar uses RSA Data's encryption method to wrap up EDI transmission 
in a secure "mailer," which then can be sent by Internet E-mail. Templar is 
being tested by Cisco Systems Inc. and three of its trading partners .

In sum, companies cannot afford to adopt an ostrich-like mentality by 
hiding their IS "heads" in the sand and hoping that the Internet revolution 
and it's influence on the seismic upheaval of the underlying bed-rock of 
fundamental EDI concepts is a passing fad.  This criticism is even more 
valid for academic programs at "top" universities where very few have any 
courses that offer an in depth analysis of EDI issues other than superficial 
overviews that teach definitions. This must change and we need to "re-
engineer" our thinking about EDI for effective electronic commerce. 

Interested readers can subscribe to a listserv on the Internet dedicated to 
discussion about these new forms of EDI. To become a member of the list 
just send a message to "majordomo@utu.premenos.com" with the message 
body containing:  "subscribe open-edi".