One may ask why with so many standards around do we need to have "Integration/Interface Engines"? Aren't communications "standardized" already? - Not really.  For a useful data exchange to take place a few things have to match between the two systems besides being electronically connected. They should be able to a) recognize each other's syntax (i.e. the order of the data stream, delimiters, and structure) and b) understand each other's vocabularies.

Syntax

Some industries have developed  data interchange standards specific to their business' workflow, as is the case with HL7 for healthcare. Although modeled after X12 in its hierarchical philosophy, the data dictionary and other syntax rules make HL7 very suitable to data interchange only within the healthcare setting. For communicating outside that environment HL7 must be mapped to X12 or XML. 

Likewise the insurance business has its own industry specific standard such as ACORD/XML, and so have other industries. 

In this context, EDI is anything that is in compliance with ANSI X12 and/or EDIFACT standards which are becoming the cross-industry standard of choice.

In 1983, ANSI published the first American National Standards  for EDI, (X12), through its Accredited Standards Committee (ASC). In 1985 the United Nations (UN) established the international standards body, Electronic Data Interchange For Administration, Commerce, and Transport (EDIFACT).  ASC X12 and EDIFACT have issued more than 300 transaction sets intended to satisfy a broad spectrum of data requirements. The EDIFACT standards are primarily used in Europe and Asia. However, in order for everyone to benefit from a single global EDI standard, ANSI X12 agreed to begin  in 1997 a gradual alignment with EDIFACT. The ANSI X12 alignment plan includes administrative alignment (and technical migration) to UN/EDIFACT. 

Top Level Comparison of X12 and UN/EDIFACT EDI Standards

Both standards have the following in common:

1. Using tagged and delimited  ASCII character strings for encoding messages to specify the structure and content of the data.

2. Using a data dictionary of standard (global) business data elements and data segments.

3. Having predefined message types using combinations of data elements to create standard messages, which are composed of a specified sequence of data segments.

Although there are basic similarities between the ANSI X12 transaction set and UN/EDIFACT message, there are some fundamental differences:

1. The application level syntax rules of the ANSI X12 and UN/EDIFACT messages vary greatly. For example, the messages are delimited and terminated by using different character sets and rules.

2. Looping and nesting conventions differ.

3. Number of data element types defined: ANSI X12 has  six data element       types:  Numeric, Decimal, Identifier, String, Date, and Time, whereas UN/EDIFACT, has three data element types: Numerical, Alphabetical, and Alpha-numerical.

4. EDIFACT allows for two levels of syntax and there is no provision for optional fields.

5. EDIFACT uses composite data elements.
   

In summary, at the macro level the interchange structure is very similar, however at the micro level the structure is significantly different. The variance in the way segments and data elements are structured is large enough that these two standards are not easily interchangeable. Often, multiple X12 data elements are needed to define a single EDIFACT data element, which leads to a many-to-one mapping scheme.

Data Dictionaries

In order for information systems to usefully exchange data they need to communicate using a common language. Data dictionaries establish a precise meaning for the shared information, allowing for a meaningful sharing.  The languages of the day are XML-based languages.  Unfortunately there are around 1200 XML vocabularies to deal with.

Several organizations develop and maintain such dictionaries:

• ASC X12 EDI: The ASC X12 Data Element Dictionary represents the collection of basic building blocks on which all Electronic Data Interchange transaction sets are constructed. The dictionary listing (in data element number order) defines each data element and its cross-reference to EDI segments and transaction sets in which it is used including all available codes and attributes. 

• EDIFACT: The UN/EDIFACT Data Dictionary defines each data element and its cross-reference to all UN/EDIFACT messages in which it is used including all available codes and attributes. 

• RosettaNet:  an independent, self-funded, non-profit consortium dedicated to the development and deployment of standard electronic business interfaces. These standards form a common e-Business language, aligning processes between supply chain partners on a global basis.  

• OAG: The Open Applications Group (OAG) has developed the Business Object Document Architecture (BOD). This architecture provides the framework to communicate messages or business documents. BOD consists of 2 major components - control Layer and Business Data Layer. The OAG work group has provided the specification to develop a set of OAG compliant DTD's (Document Type Definition) to support their XML messaging requirement. Both the XML messages and its DTDs make up the BOD.  

• ebXML - Core Components: The ebXML Core Components Project Team is working on the methodology to develop a common business data dictionary. The goal of this effort is to develop a syntactical neutral data dictionary where it can support numerous syntax such as XMl, X12, EDIFACT, etc. This effort is currently a work in progress; the EIDX/CompTIA Technology Subcommittee will publish the final version, as it becomes available.

© 2003 - Transko Corporation