The Enterprise: XML and Enterprise Information Portals

By Clive Finkelstein

Published in DM Review in July 1999

Last month I discussed business reengineering and some white papers that identify opportunities created by the Internet. This month I will address two technologies that will become increasingly important to business and IT over the next five years. The first is the extensible markup language (XML). The second is the evolution of data warehouses to enterprise information portals (EIPs). They promise to move data administration from a backroom function to one that is critical to the future of IT and business.

Like HTML, XML is also based on standard generalized markup language (SGML). XML is more powerful than HTML, yet simpler and not as expensive as SGML. While HTML focuses on the layout and presentation of text, images, audio and video on a Web page, it does not provide any information about the meaning of the text and other data content. XML addresses this deficiency of HTML and more.

XML provides an easy way to define meta data associated with the content of Web resources. It separates meta data content definition using XML from the definition of the style of a Web document. XML styles use the extensible style language (XSL). XML also uses the extensible linking language (XLL) to provide a more powerful multidirectional linking capability than HTML.

However, the main advantage of XML is not as a replacement for HTML. It can also be used to define the structure and content of unstructured data (such as text, graphics, images, audio and video), as well as structured data in legacy files and relational or object databases. Therefore, it provides a powerful capability to integrate dissimilar systems and databases within an enterprise. This integration is the real power of XML. It is what the industry has been looking for now for many years. XML is easy to use, inexpensive and more importantly ? it works!

New development tools will emerge over the next few years that will use meta data, XML, the Internet and intranets for easy integration of unstructured documents, legacy files and relational and object databases. To benefit from these developments, organizations need to have the identification and management of their enterprise meta data under control.

This definition of meta data is what enterprises should be concentrating on today using CASE tools. The meta data captured by these tools is stored in their own or other repositories. And as CASE tools now automatically generate DDL scripts for most DBMS products, it is expected that they will soon be able to also generate the document type definition (DTD) scripts used by XML.

The capabilities offered by XML and Internet technologies will lead to the second IT area that will emerge over the next five years: enterprise information portals (EIPs).

The emphasis over the last few years on the installation and deployment of enterprise resource planning (ERP) packages has implemented operational ERP data. But ERP packages have not provided an effective vehicle to deliver ERP information for management. Enterprise information portals will provide powerful knowledge management capabilities for ERP with easy-to-use business intelligence tools.

Data warehouses and data marts deliver information to management, but today they focus only on structured data. More than 90 percent of the knowledge resource in most enterprises exists not in structured legacy files or databases, but in unstructured data such as in text documents, graphics, images, audio and video.

XML will enable enterprises to integrate this unstructured data with the structured information made available through data warehouses and data marts. Other data sources from the Internet will also be integrated with structured and unstructured enterprise data and information, delivered using intranet technologies. As they provide this capability, data warehouses will evolve into enterprise information portals ?a single entry point for all information and knowledge resources in an enterprise.

To help you understand XML and the development of enterprise information portals, McGraw- Hill will publish a book in September co-authored by Peter Aiken and myself entitled, Building Corporate Portals Using XML. Methods covered in this book include data warehouse planning, implementation and deployment. We hope everyone will find it interesting and useful.

Clive Finkelstein, the "father" of information engineering (IE), is an international consultant and an instructor. He is the managing director of Information Engineering Services Pty Ltd (IES) in Australia and a member of the International Advisory Board of DAMA International. Contact Finkelstein by e-mail at cfink@ies.aust.com.