Monday, July 14, 2008

1. XML for Business Intelligence

Much sooner than later, every company will have to do business the XML way, or it won’t do business at all.
Robert H. Hertz, E. Mary Keegan, and David M. H. Phillips, The Value Reporting Revolution

XML (an acronym for “eXtensible Markup Language”) is not a programming language, and, despite its name, it is not just a language for “marking up” documents. XML is a way of expressing the meaning of information – XML provides us with an unambiguous way of expressing the semantics of data that we can communicate to another person or machine.

It is the ability of XML to communicate meaning to a computer, however, that makes it a revolution whose time is inevitable. Human readers have the ability to derive meaning for data from the position or format of a report. Columns represent categories, rows allow the data to be linked to a person or object, and positions within sentences provide us with textual context for the data. However, machines have a great difficulty in determining visual patterns and require a “markup” language to give data categories, textual context, and the other dimensions of meaning.

XML simply provides us with the ability to attach meaningful tags to pieces of information, allowing a machine that reads that information to detect the critical information. A machine that is programmed to read income summaries from financial statements, get simply scan the document for the “income” tag and read the numbers inside the tag, allowing the machine to find thousands of corporate incomes within minutes.

The intent of this blog is to introduce the reader to the powers of XML in the world of financial information and business intelligence. This is the first post for this blog. In the following blogs we will continue to show how the hierarchical structure of XML allows a document to report orders of magnitude more information than was possible before its inception. See Banking the Past, page 180.