Metapattern: Context and Time in Information Models
Angela Poynton
Ranch Hand
Joined: Mar 02, 2000
Posts: 3143
posted
As the author is quick to point out this is an innovative and highly original work in conceptual information modeling. After a formal presentation of metapattern terminology and key concepts Wisse jumps into the middle of several issues in conceptual modeling. For example, the book hopes to explain that an object in context is different from the absolute objects assumed by traditional OO models. The metapattern presumes that an object's behaviors are completely different from one context to another. The metapattern, by emphasizing context over object, enables you to adequately describe the enormous variety found in real life. Traditional OO modelers should be ready to leave behind many dogmas. In the metapattern, time has drastic consequences for objects. In addition, OO designers should be forewarned. The metapattern is not a method for technical design or software engineering. 'Metapattern' is a formalization of context, time and validity in information models. This book is actually several books within one and you ought to pay careful attention to the order you read the chapters. The middle parts compare the metapattern to recent work by authors James J. O'Dell, Haim Kilov, August-Wilhelm Scheer, David C. Hay and Martin Fowler. These chapters supplement the theoretical explanation of the metapattern's most fundamental characteristics. A necessary warm-up before tackling the challenging terminology and concepts in the first four chapters is to read both appendices and the introduction. Wisse's schematic diagrams are simple and elegant but weren't enough. This reviewer wanted additional, concrete explanations of core ideas. Immense value would come from more concrete comparisons and examples. The author's frequent tangential remarks, while doubtless precise and true, were distracting. Frankly, I tried hard to understand the metapattern. I read the book cover to cover, then re-read chapters in a different order, all the while investigating his ideas using concrete examples, as best I could. There were glimmers of excitement as I encountered familiar concepts from metaphysics, epistemology, set theory, information modeling and object-oriented modeling. However, all too often I missed the point of his arguments. Reading does not entail understanding. Perhaps I am too much of a software engineer and not enough of a conceptual modeler. I must apologize to the author, but the fairest thing for me to say about the metapattern is that I still don't understand it.(Scott M. Allman, Feb 2001)
