A while back, I did a great deal of research into faceted classification, metadata, ontological schemes, and other informationally challenging topics. The problem? Complex access control systems. As I work more and more towards a practical application of such systems, it’s occuring to me that complex business rule application systems may actually be on the same path.
Both business rule systems and classification systems have been developing in terms of the schema used to describe interrelationships and requirements of various objects and logic in complex systems. When you boil it down, business rule systems collect metadata about the relationships between objects, and enforce well-defined policies on that metadata. Ontological schemes not only capture metadata about objects, but also encode relationship information in ways that can be used similarly.
Big players on the business rule from include IBM, ILOG, Fair Isaac, Corticon, Production Systems Technologies, and even Microsoft.
Whereas, metadata-based systems are used in the open source world aren’t traditionally known as business rule systems. There’s been a lot of discussion about the ways that you can build faceted navigation systems, related links sidebars, and other neato website widgets within the web development world (especially that of information architects). Major players in this sphere include Ontopia, Ontoprise, IBM (again), and more.
Business rule systems are in some ways, the practical application of these metadata-based topics that kind of struck me by surprise. The more I worked on application of metadata towards decision-making process, the more I realized that the work i was doing is remarkably similar to that described by business rule engines.
Although i’m sure i’m simplifying slightly to make a point, it may prove useful for others to be able to see the similarities between business rule engines and metadata-based information architectures, two seemingly different (at least the marketing materials appear different) approaches towards solving two mysteriously similar – perhaps the same – problem spaces.
Uncategorized