Standards only remain “standard” until the next change. Even in the B2B EDI world where trading partner data exchanges are “supposed to be” standardized, new additions and upgrades to later versions will create an opportunity to re-evaluate how data is being moved and stored for company consumption.
Traditional EDI transactions are built upon the “Header” and “Detail” premise where a single iteration of header information could be followed by many iterations of detail information. This makes it simple to build a database in support of this structure – a single transaction would see one record moved into the header file, and one or more records moved into the detail file. Header and detail would be linked together through common key values and sequential controls.
As requirements mature, changes are made and generally new information is presented. This could come by way of additional (single or multiple pieces of) data that could be stored in either the header or detail file. Trouble begins to brew when this new data is defined [by itself] as repeating information. Continue reading