Updates

Latest Tweet



What's New?

Check out for latest innovation, a computer based training video collection


Like this Page

Software Fortresses: Modeling Enterprise Architectures Review by Michael Tan

Excellent for Enterprise Systems Architecture

I enjoyed the book thoroughly. The concept of using the fortress as a metaphor for system is fluently presented. It starts with simple and well understood characters in a typical fortress and evolves logically into the more difficult concepts, always building on previous discussions and reiterating important concepts if necessary.

I found the book very useful in my role as an enterprise architect (more about that later) for a scientific agency here in Canberra. As an agency, we are concern about data interoperability. Our stakeholders download data from our internet site to use with their researches and other business activities. They need interoperable data. The same can be said about our own researchers who need interoperable data from other sites. So we need a system that facilitates that exchange of data. It will start with manual interface at first but will require machine2machine interfaces in time to come. We know web services technology is the obvious answer. Our challenge is to logically evolve from the big picture (all the requirements) to the technologies we need to build the system. Discussions about heterogenous/homogeneous synchronous/asynchronous drawbridges are excellent way to evolve logical designs to physical designs. Technology selection does not have to base on personal experiences anymore. The fortress software metaphor also gives me a perfect tool in painting the big picture to general users and management. The fortress analogy creates resonance immediately because we, as government agent, are paranoid about security.

From a software development practice viewpoint, the software fortress model carries many of the virtues of object-based programming like reuse and data encapsulation. This is not a coincidence given the fact that author is a guru in this area. What the model gives us is also the opportunity to transition into this important practice from the current amateurish programming habits.

From the point of an enterprise architect, software fortress is an excellent tool to develop the enterprise technology architecture from an enterprise system architecture. An enterprise system architecture satisfies the enterprise information architecture which in turn satisfies the enterprise business architecture. At the moment, that transition from systems to technology is more an art than a science. As the author asserted the software industry has no conceptual model for building enterprise systems.

If I have any objection with the author, it is the usage of the term "enterprise architecture" in the book. Enterprise Architecture actually has a well defined meaning and acceptance, at least in USA. If you point your browser to http://www.whitehouse.gov/omb/egov/a-1-fea.html , you will see "enterprise architecture" actually consist of business, information, system and technology architecture. By "enterprise architectures" in the book, the author actually refers to enterprise system architecture. As an enterprise architect, I hope we all have a consistent use and meaning of the terminology.