Traditional EDI

The number of users of Microsoft's Windows technology has been reported as having passed 60,000,000 world wide after a few short years, whereas EDI with decade of history is reported to have less than 100,000 users.

Where have the Traditional EDI (trad-edi) schemes such as ASC X12 and UN/EDIFACT gone wrong and why are so few organisations taking advantage of it?

The ICARIS project set out to find some answers to this question.

After studying the problem, talking to users and sitting in on the various aspects of the standardisation and implementation processes, it was concluded that:

• The "standards" are not standard, but a series of proprietary implementations based on, but not necessarily loyal to, a generic framework.
• Trad-edi is based on a premise that all kinds of businesses can fit into one mould for transaction data and the actual business processes (as implemented by application programs in computerised systems) are quite deliberately ignored. This gives rise to a situation where the EDI implementations do not mesh with the operation of the application programs (ie business processes) and then it is very difficult to integrate them and achieve interoperation.
• Trad-edi translates transaction files into a very different kind of file structure with its own peculiar syntax before transmission of the transaction data to a trading partner. It then sets out to "untranslate" back to a file that can be processed by the application program at the receiving end. Many users receiving trad-edi transactions reported the need to "correct" the "untranslation" process by writing custom-built conversion programs to enable them to be processed. This raises the question of why translation is necessary in the first place.
• The very prolonged processes whereby new transaction "messages" or "transaction sets" are accredited and then implementation conventions drawn up and a similar delay when changes are needed means the result is necessarily static. Business, however, needs to be highly responsive to changes in the market place and business environment; a need trad-edi is unable to meet.
• Most of the problems with the uptake of EDI, particularly by Small and Medium Enteprises (SMEs) appears to stem directly from the translation process. There is good reason to expect the greater proportion of the problems would disappear if the raw transaction files were used with a simple and automatic reformatting operation to compensate for different file layouts at one end. This is the basis of the second generation proposals.

A paper, "The Case for Next Generation EDI", setting out the problems with trad-edi and proposing a move to the next generation technology to overcome them is available for download.

There is also a presentation in powerpoint 4.0 which may be downloaded.

The ICARIS Project has proposed Business System Interoperation (BSI) as a possible solution to the shortcomings of traditional edi.