EDI (electronic data interchange) is a generic term for
electronically controllable commercial processes which facilitate electronic data
interchange from computer to computer on the basis of various standard formats. XML on the
other hand is primarily a standardized markup language, though one which was developed
with the aim of simplifying data interchange. Like XML, EDI is based on the idea of
linking different applications together. But, will XML be able to gain ground in the
medium term, and possibly even displace EDI?
There is no doubt at all that XML-based data interchange is going to catch up fast. We
don't have to look very closely at traditional EDI to recognize that all its solutions are
proprietary ones. Its implementation is linked to substantial initial investments in
sophisticated technical environments (value added networks with expensive and complicated
data communications systems, for instance). Since not every firm can afford investments on
this scale, the great majority of small and medium-sized business are barred from enjoying
the benefits of efficiency enhancements through EDI-based business processes. Furthermore,
there are a number of different industry-specific EDI standards which are incompatible
with one another (ANSI X12, UN/EDIFACT, ODETTE etc.), with formats that are extremely
difficult - if not impossible - to convert.
This is one reason for the mediocre success of EDI. No more than two or three percent
of all business transactions worldwide involve EDI - most of them by Fortune 500 firms in
the USA. XML takes up the EDI principle, in that it maps the industry-specific document
structures and elements used for standard business processes, such as orders, invoices,
delivery information etc., to DTDs (document type definitions), often making them
available for general referencing in special "schema repositories"
(XML/EDI).
XML is much more dynamic and robust than EDI, and far less time-consuming to harmonize
with business partners. XML documents can be adapted faster and more easily if agreement
terms change, and automatically executing processes are not interrupted if additional
information (such as new elements) is inserted into the incoming data stream. In addition,
the order of incoming, self-explanatory elements is largely irrelevant for XML-capable
applications.
XML is simple to learn and its documents are simple to understand, as well as being
cheaply and universally exchangeable over the Internet. It is this key advantage which
leads us to predict that XML-based EDI will succeed in emerging from the shadow of its
coexistence with EDI and possibly even in ousting it altogether in the not-too-distant
future. This will depend more on the level of a company's investment so far (return on
investment) than on the technically implicit improvements that can be achieved with
XML-based EDI.
EDI and XML are not inherently mutually exclusive.
XML-based EDI standards, all of which share the common goal of making EDI processes
easier and more flexible to handle, can now be developed on the bedrock of this language.
In reality, therefore, classic EDI and XML-based EDI are very similar to one another.
Renowned organizations and firms, such as RosettaNet, CommerceNet, Commerce One, Ariba and
Microsoft (BizTalk), have already proposed XML-based EDI standards, although there are as
yet no signs that any of them will achieve predominance. XML Solutions is moreover in the
process of converting all ANSI X12 and EDIFACT transaction elements to XML.
By far the most likely scenario is that the two forms will continue to complement one
another, at least for a while. Large firms, which may already be practicing EDI with their
business partners, will be unwilling to abandon their established processes. Since each
new EDI project generates huge costs and is extremely time-consuming to implement,
XML-based data exchange will become increasingly attractive to EDI-practicing
corporations. A whole host of new, comparatively small business partners, often
characterized by greater flexibility, will thus acquire low-cost access to electronic
commerce with large-scale enterprises.
Although private VAN networks are still considered to constitute a safer alternative to
the Internet, the present shortcomings in Internet security will soon be a thing of the
past as the volume of electronically conducted business grows, thus leaving the door wide
open for pure XML approaches.