It is not uncommon to encounter situations where you need additional data that is not generally available either from your EDI application/interface (outbound) or from your trading partner (inbound).
Consider the situation where your trading partner sends purchase orders but omits the item descriptions…this item description being a necessary piece of information for processing the received orders. One option is to require that your trading partner includes this description for each item on the order – good luck trying to persuade a buyer to accommodate your request. Another option is to modify your application to eliminate this requirement – this may require a considerable amount of redevelopment.
The solution: Create an external table to host descriptions for each partner item that can be accessed during the translation process. Continue reading →
Enterprise Service Busses (ESBs) offer an interesting communications layer that enables an enterprise to expose data to interested parties (i.e. applications, data-feeds, etc.) with a Publish/Subscribe model. The Pub/Sub model originated in the printed media world, utilized as a distribution model for newspapers and magazines. It has evolved with modern times into the electronic age in the form of email-subscribed newsletters, and more recently, RSS feeds such as blogs.
In the enterprise, there is a growing need to share data among systems, both internally (A2A) and externally (B2B). However, as new demands for sharing data surface, we need a way to “bolt in” the new requestors without impacting our current implementations.
ESBs commonly implement a variant of the GoF Observer Pattern. This exposes a Publication/Subscription model allowing information sources (publisher) to expose data (message) on a queue. One or more interested parties (subscriber) consume the data. The key benefit of loose coupling in a Publication/Subscription model is that the Publisher does not need to know, or care, about “who” is subscribing. The data is published and downstream subscribers use the data as they see fit.
Spreadsheets have become a du jour “standard” for some forms of Business-to-Business integration, offering a data representation that is easy to produce and consume, and is widely supported across industries and geographic boundaries. Spreadsheets are also very portable; they can be emailed, viewed across platforms (Microsoft Windows, Apple OS X, Linux) and are accessible by many software packages (Microsoft Excel, OpenOffice, Google Docs).
Using spreadsheets as an integration medium can be challenging, however, because the layout of data within a spreadsheet can vary. Spreadsheets aren’t always simple grids of rows and columns. Data can be represented in a tabular format, e.g., to transmit raw data from a back-end application, or in a forms-based layout, similar to what we might find in business documents like purchase orders and invoices. That flexibility makes the spreadsheet versatile and attractive as a data-publishing tool, but makes integration of some spreadsheet cases less trivial.